System, method, and apparatus for providing dynamic, prioritized spectrum management and utilization.

This application is related to and claims priority to the following US patents and patent applications. This application claims priority to US Patent Application No. 17/085,635, filed October 30, 2020 , which claims the benefit of U.S. Provisional Application No. 63/018,929, filed May 1, 2020. Each of the applications listed above is incorporated herein by reference in its entirety.

BACKGROUND OF THE INVENTION

1. field of invention

The present invention relates to spectrum analysis and management for electromagnetic signals and, more particularly, to providing dynamic and prioritized spectrum management and utilization.

2. Description of the prior art

It is generally known in the prior art to provide wireless communications spectrum management to detect devices and to manage space. Spectrum management includes the process of regulating the use of radio frequencies to promote efficient use and obtain a net social benefit. One problem facing effective spectrum management is the varying numbers of devices emitting wireless signals propagating at different frequencies and through different technological standards. Coupled with different regulations related to spectrum usage around the world, effective spectrum management becomes difficult to obtain and, at best, can only be achieved over a long period of time.

Another problem facing effective spectrum management is the increasing need for spectrum despite the finite amount of spectrum available. Wireless technologies and applications or services that require spectrum have grown exponentially in recent years. Consequently, the available spectrum has become a valuable resource that must be used efficiently. Therefore, systems and methods are needed to effectively manage and optimize the available spectrum being used.

The prior art patent documents include the following:

US Patent Publication No. 2018/0352441 for Devices, Methods and Systems with Dynamic Spectrum Sharing by inventors Zheng, et al., filed on June 4, 2018 and published on December 6, 2018 , is aimed at devices, methods and systems with dynamic spectrum distribution. A wireless communication device includes a software-defined radio, a spectrum sensing subsystem, memory, and an electronic processor. The software defined radio is configured to generate an input signal and communicate wirelessly with one or more radio nodes using a traffic data channel and a transmission control channel. The spectrum detection subsystem is configured to detect local spectrum information of the input signal. The electronic processor is communicatively connected to the memory and the spectrum sensing subsystem and is configured to receive local spectrum information from the spectrum sensing subsystem, receive spectrum information from one or more radio nodes, and allocate resources for the channel. of traffic data based on local spectrum information and spectrum information received from one or more radio nodes.

The US Patent Publication No. 2018/0295607 for Method and Apparatus for Using Adaptive Bandwidth in a Wireless Communication Network by inventors Lindoff, et al, filed on October 10, 2017 and published on October 11, 2018, is addressed to reconfigure the bandwidth of a receiver of the initiating wireless device to match the second scheduling bandwidth, wherein the second scheduling bandwidth is greater than a first scheduling bandwidth currently associated with the wireless device, and wherein the first and second scheduling bandwidth respectively define the bandwidth used to schedule transmissions to the device. wireless.

US Patent No. 9,538,528 for Efficient Coexistence Method for Dynamic Spectrum Sharing by inventors Wagner, et al, filed on October 6, 2011 and issued on January 3, 2017, is directed to a apparatus that defines a set of resources from a first number of orthogonal radio resources and controls a transmission means to simultaneously transmit a respective first radio signal for each resource on all resources in the set. A respective estimated interference is estimated on each of the resources in the set when the first respective radio signals are transmitted simultaneously. A first resource in the set is selected if the estimated interference on the first resource exceeds a first predefined level and, in the set, the first resource is replaced by a second resource from the first number of resources that have not been part of the set. Each of the checks and estimates, selection and substitution is performed in order, respectively, for a predefined time.

US Patent No. 8,972,311 for Intelligent Spectrum Allocation Based on User Behavior Patterns by inventors Srikanteswara, et al., filed on June 26, 2012 and issued on March 3, 2015, is directed to a platform to facilitate the transfer of spectrum rights that includes a database to determine information about the spectrum available for use in wireless communications. A spectrum use request from an entity in need of spectrum may match available spectrum. This iviA/a/zuzz/uid/oo matching involves determining a pattern in user requests over time to optimize spectrum allocation. The Cloud Spectrum Services (CSS) process allows entities to access spectrum they would not otherwise have; allows the end user to complete their download during periods of congestion while maintaining a high quality of service; and allows the spectrum lease holder to receive compensation for an otherwise dormant asset.

U.S. Patent No. 10,536,210 for Interference Suppression Method and Device in a Dynamic Frequency Spectrum Access System by inventors Zhao, et al., filed April 14, 2016 and issued April 14 January 2020, is aimed at an interference suppression method and device in a frequency dynamic spectrum access (DSA) system. The system includes: a frequency spectrum management device, a primary system including a plurality of primary devices, and a secondary system including a plurality of secondary devices. The method includes: transmitting position information of each of the secondary devices to the frequency spectrum management device; determining, by the frequency spectrum management device, a weight factor for a specific secondary device according to the formation of the received position; and performing a second stage precoding, and in the second stage precoding, adjusting, using the weight factor, an estimated power of the specific secondary device leaking to the other secondary device.

The US patent No. 10,582,401 for Large-Scale Radio Frequency Signal Information Processing and Analysis System by inventors Mengwasser, et al., filed April 15, 2019 and issued March 3, 2020, is directed to a large-scale radio frequency signal information analysis and processing that provides advanced signal analysis for applications in telecommunications, including determinations of band capacity and geographic density and detection, classification, identification and geolocation of signals over a wide range of frequencies and across wide geographic areas. The system can use a variety of novel algorithms for bin-wise processing, Rayleigh distribution analysis, telecommunication signal classification, receiver anomaly detection, transmitter density estimation, transmitter detection and location, geolocation analysis , estimate of telecommunications activity, estimate of telecommunications use, estimation of frequency utilization and data interpolation.

U.S. Patent No. 10,070,444 for Coordinated Spectrum Allocation and Deallocation to Minimize Spectrum Fragmentation in a Cognitive Radio Network by inventors Markwart, et al., filed December 2, 2011 and issued December 4 September 2018, is ινΐΛ/a/zuzz/ui or/oo

The US Patent Publication No. 2017/0041802 for Spectrum Resource Management Device and Method by inventors Sun, et al., filed on TJ May 2015 and published on February 9, 2017, is directed to a resource management device spectrum that determines the available spectrum resources of a target communication system, so that the aggregation interference caused by the target communication system and a communication system with a low right versus a communication system with a high right in a management area does not exceed an interference threshold of the communication system with a high right; reduces the available spectrum resources of the low-entitlement communication system, so that the interference caused by the low-entitlement communication system against the target communication system does not exceed an interference threshold of the target communication system; and updates the available spectrum resources of the target communication system according to the reduced available spectrum resources of the communication system with a low entitlement, so that the aggregation interference does not exceed the interference threshold of the communication system with a low entitlement. high.

US Patent No. 9,900,899 for Dynamic Spectrum Allocation Method and Dynamic Spectrum Allocation Device by inventors Jiang, et al., filed on March 26, 2014 and issued on February 20, 2018 , is aimed at a dynamic spectrum allocation method and a dynamic spectrum allocation device. In the method, a centralized node performs spectrum allocation and transmits a spectrum allocation result to each communication node, so that the communication node operates on a corresponding spectrum resource according to the spectrum allocation result and performs communication quality measurement information statistics. The centralized node receives the communication quality measurement information reported by the communication node and determines whether or not to activate spectrum reallocation for the communication node according to the communication quality measurement information. on the communication node. When spectrum reallocation is required to be activated, the centralized node reallocates the spectrum for the communication node.

U.S. Patent No. 9,578,516 for Radio System and Spectrum Resource Reconfiguration Method by inventors Liu, et al., filed on February 7, 2013 and issued on February 21, 2017, is directed to a radio system and a method of reconfiguring spectrum resources thereof. The method comprises: a reconfigurable base station (RBS) dividing the subordinate nodes into groups according to the attributes of the subordinate nodes and sending a reconfiguration command to a subordinate node in a designated group, and the RBS and the subordinate node execute reconfiguring spectrum resources according to the reconfiguration command; or, the RBS executes the reconfiguration of the spectrum resources according to the ινΐΛ/a/zuzz/ui or/oo reconfiguration command; and a subordinate User Equipment (UE) accessing a reconfigured RBS after the outage. Reconfiguration of the spectrum resources of a cognitive radio system can be performed.

U.S. Patent No. 9,408,210 for Method, Device and System for Dynamic Frequency Spectrum Optimization by inventors Pikhletsky, et al., filed on February 25, 2014 and issued on August 2, 2016, It is directed to a method, a device and a system for the dynamic optimization of the frequency spectrum. The method includes: predicting a distribution of terminal(s) traffic in each cell of multiple cells; generating multiple frequency spectrum allocation schemes for multiple cells according to the traffic distribution of the terminal(s) in each cell, wherein each frequency spectrum allocation scheme comprises assigned frequency spectrum(s) for each cell; select a frequency spectrum allocation scheme superior to a current multi-cell frequency spectrum allocation scheme from the multiple frequency spectrum allocation schemes based on at least two network performance indicators of a network in which they are located multiple cells; and allocate frequency spectrum(s) for the multiple cells using the selected frequency spectrum allocation scheme. This improves the utilization rate of the frequency spectrum and optimizes multiple network performance indicators at the same time.

The US patent No. 9,246,576 for Apparatus and Methods for Dynamic Spectrum Allocation in Satellite Communications by Inventors Yanai, et al., filed on March 5, 2012 and issued on January 26, 2016, is directed to a communication system including a satellite communication apparatus that provides communication services to at least a first set of communicators, the first set of communicators including a first plurality of communicators, wherein the communication services are provided to each of the communicators in accordance with a spectrum assignment corresponding thereto, to thus define a first plurality of spectrum assignments that distributes a first predefined portion of spectrum among the first set of communicators; and a dynamic spectrum allocation operating apparatus for dynamically modifying at least one spectrum allocation corresponding to at least one of the first plurality of callers without exceeding the spectrum portion.

U.S. Patent No. 8,254,393 for Leveraging Predictive Models of Channel Availability Durations for Enhanced Opportunistic Allocation of Radio Spectrum by Inventor Horvitz, filed June 29, 2007 and issued August 28, 2012, is aimed at describing an adaptive and proactive radio methodology for opportunistic allocation of radio spectrum. The methods can be used to allocate ινΐΛ/a/zuzz/ui or/oo radio spectrum resources by employing machine learning to learn models, by accumulating data over time, that have the ability to predict the context-sensitive durations of channel availability. Predictive models are combined with cost-benefit analysis and decision theory to minimize service or quality disruptions that may be associated with reactive allocation policies. Instead of reacting to channel losses, proactive policies look for changes before the loss of a channel. Beyond determining the duration of availability for one or more frequency bands, statistical machine learning can also be used to generate price predictions to facilitate the sale or rental of available frequencies, and these predictions can be used in switching analyses. The methods can be employed in distributed non-cooperative allocation models, in centralized allocation approaches and in hybrid spectrum allocation scenarios.

US Patent No. 6,990,087 for Dynamic Wireless Resource Utilization by inventors Rao, et al., filed April 22, 2003 and issued January 24, 2006, is directed to a method for dynamic wireless resource utilization that includes monitoring a wireless communication resource; generate wireless communication resource data; using wireless communication resource data, predicting the appearance of one or more available slots in a future time period; generate prediction data for available spaces; using available slot prediction data, synthesizing one or more wireless communication channels from one or more predicted available slots; generate channel synthesis data; receive data reflecting feedback from a previous wireless communication attempt and data reflecting a network condition; according to the received data and the channel synthesis data, selecting a particular wireless communication channel from one or more synthesized wireless communication channels; generate wireless communication channel selection data; using the wireless communication channel selection data, instructing a radio unit to communicate using the selected wireless communication channel; and instruct the radio unit to discontinue use of the selected wireless communication channel after communication has been completed.

U.S. Patent No. 10,477,342 for Systems and methods of using location, context, and/or one or more wireless communication networks to monitor, prevent, and/or mitigate behavior previously identified by inventor Williams, filed on December 13, 2017 and issued on November 12, 2019, addresses systems and methods of using location, context and/or one or more communication networks to monitor, prevent and/or mitigate previously identified behavior. For example, exemplary modalities described herein may include involuntary, automatic and/or wireless monitoring/mitigation of ινΐΛ/a/zuzz/ui or/or undesirable behavior (e.g., undesirable behavior related to addiction, etc.) of a person (e.g., an addict, a person on probation, a system user, etc.). In an exemplary embodiment, a system generally includes a plurality of devices and/or sensors configured to determine, through one or more communications networks, a location of a person and/or a context of the person at the location; predict and evaluate a risk of a behavior previously identified by the person in relation to the location and/or context; and facilitate one or more actions and/or activities to mitigate the risk of the previously identified behavior, if any, and/or react to the previously identified behavior, if any, by the person.

BRIEF DESCRIPTION OF THE INVENTION

The present invention relates to spectrum analysis and management for electromagnetic signals and, more particularly, to providing prioritized and dynamic spectrum management and utilization. Furthermore, the present invention relates to spectrum analysis and management for electromagnetic signals (e.g. e.g., radio frequency (RF)), and to automatically identify reference data and state changes for signals from a multiplicity of devices in a wireless communications spectrum, and to provide remote access to the measured and analyzed data through a virtualized computer network. In one embodiment, signals and signal parameters are identified and indications of the available frequencies are presented to the user. In another embodiment, the signal protocols are also identified. In a further embodiment, the modulation of the signals, the types of data carried by the signals, and the estimated origins of the signals are identified.

It is an object of this invention to prioritize and manage applications in the wireless communications spectrum, while optimizing application performance.

In one embodiment, the present invention provides a system for the dynamic and prioritized management and utilization of spectrum in an electromagnetic environment that includes at least one monitoring sensor operable to monitor the electromagnetic environment, thereby creating measured data, at least one analysis engine. to analyze the measured data, at least one application, a semantic engine that includes a policy editor and programmable rules, and a suggestion and signal server, wherein the data analysis engine(s) includes a detection engine and a learning engine, wherein the detection engine is operative to automatically detect at least one signal of interest, and wherein the learning engine is operative to learn the electromagnetic environment, wherein the application or applications include an occupancy survey application and a resource brokering application, wherein the occupancy survey application is operative to determine occupancy in frequency bands and schedule occupancy in at least one frequency band, and wherein the d/OO resource brokerage application is operative to optimize resources to improve the performance of at least one client application and/or at least one client device, wherein the programmatic rules and policies editor includes at least one rule and/or at least one policy, and wherein one or more of the rule(s) and/or policy(ies) are defined by at least one client, wherein the suggestion and signal server is operative to use the analyzed data from at least one analysis engine. data to create actionable data, where each of the client application(s) is assigned a priority, and wherein the priority and one or more of the rule(s) and/or policy(ies) are used to dynamically assign the frequency band(s) in the electromagnetic spectrum.

In another embodiment, the present invention provides a system for the dynamic and prioritized management and utilization of spectrum in an electromagnetic environment that includes at least one operational monitoring sensor to monitor the electromagnetic environment, thereby creating measured data, at least one analysis engine. to analyze the measured data, at least one application, a semantic engine that includes a policy editor and programmable rules, and a suggestion and signal server, wherein the data analysis engine(s) include a detection engine, an identification engine, a classification engine, a geolocation engine, and a learning engine, wherein the detection engine may function to automatically detect at least one signal of interest, and wherein the learning engine is operative to learn the electromagnetic environment, wherein the application(s) include an occupancy survey application and a resource brokering application, wherein the occupancy survey application is operative to determine occupancy in frequency bands and schedule occupancy in at least one frequency band. , and wherein the resource brokering application is operative to optimize resources to improve the performance of at least one client application and/or at least one client device, wherein the programmable rules and policies editor includes at least one rule and /or at least one policy, and where one or more of the rules and/or policies are defined by at least one client, wherein the hint and signal server is operative to use the analyzed data from the data analysis engine(s) to create actionable data, wherein each of the client application(s) is assigned a priority, and wherein the priority and one or more of the standard(s) and/or policy(ies) are used to dynamically allocate the frequency band(s) in the electromagnetic spectrum.

In yet another embodiment, the present invention provides a method for the dynamic and prioritized management and utilization of spectrum in an electromagnetic environment that includes providing a semantic engine that includes a programmable policies and rules editor, wherein the programmable policies and rules editor includes at least one rule and/or at least one iviA/a/zuzz/ui or/or policy, and where one or more of the rule(s) and/or policy(ies) are defined by at least one client, monitoring the electromagnetic environment using at least one monitoring sensor, thereby creating measured data, analyzing the measured data using at least one data analysis engine, thereby creating analyzed data, wherein the data analysis engine or engines include a data analysis engine detection and a learning engine, learning the electromagnetic environment using the learning engine, automatically detecting at least one signal of interest using the detection engine, determine occupancy in frequency bands and schedule occupancy in at least one frequency band using an occupancy survey application, optimize resources to improve the performance of at least one customer application and/or at least one customer device using a resource brokering application, assign a priority to each of the client's applications, and dynamically assign at least one frequency band in the electromagnetic spectrum based on priority and one or more of the standard(s) and/or policy(ies).

In yet another embodiment, the present invention provides a method for dynamic and prioritized management and utilization of spectrum in an electromagnetic environment that includes providing a semantic engine that includes a programmable policy and rule editor, wherein the policy and rule editor programmable includes at least one rule and/or at least one policy, and where one or more of the rule(s) and/or the policy(ies) are defined by at least one client, monitoring the electromagnetic environment using at least one monitoring sensor, thereby creating measured data, analyzing the measured data using at least one data analysis engine, thereby creating analyzed data, wherein the data analysis engine or engines includes a data analysis engine detection, a classification engine, an identification engine, a geolocation engine and a learning engine, learning the electromagnetic environment using the learning engine, automatically detect at least one signal of interest using the detection engine, classify the signal or signals of interest using the classification engine, identify the signal or signals of interest using the identification engine, determine the location of the signal or signals of interest using the geolocation engine, determine occupancy in frequency bands and schedule occupancy in at least one frequency band using an occupancy survey application, optimize resources to improve the performance of at least one customer application and/or at least one customer device using a resource brokering application, assign a priority to each of the client application(s), and dynamically assign the frequency band(s) in the electromagnetic spectrum based on priority and one or more of the standard(s) and/or policy(ies).

These and other aspects of the present invention will become apparent to those skilled in the art after reading the following description of the preferred embodiment when considered with the drawings, as they support the claimed invention.

BRIEF DESCRIPTION OF THE DRAWINGS

The patent or application file contains at least one drawing executed in color. Copies of this patent publication or patent application with color drawings will be provided by the Office upon prior request and payment of the necessary fee.

FIG. 1 illustrates one embodiment of an RF recognition and analysis system.

FIG. 2 illustrates another modality of the RF recognition and analysis system.

FIG. 3 is a flow chart of the system according to one embodiment.

FIG. 4 illustrates the acquisition component of the system.

FIG. 5 illustrates one modality of an analog frontend of the system.

FIG. 6 illustrates one embodiment of a radio receiver front-end subsystem.

FIG. 7 continues the modality of the front of the radio receiver shown in FIG. 6.

FIG. 8 is an example of a programmable channelizer in the time domain.

FIG. 9 is an example of a programmable channelizer in the frequency domain.

FIG. 10 is another modality of a programmable channeler.

FIG. 11 illustrates one embodiment of a blind detection engine.

FIG. 12 illustrates an example of an edge detection algorithm.

FIG. 13 illustrates an example of a blind classification engine.

FIG. 14 illustrates details about selection matching with cumulants for modulation selection.

FIG. 15 illustrates a flow chart according to an embodiment of the present invention.

FIG. 16 illustrates the functions of the control panel according to one embodiment.

FIG. 17 illustrates one embodiment of a system RF analysis architecture.

FIG. 18 illustrates one embodiment of a system detection engine.

FIG. 19 illustrates a mask according to an embodiment of the present invention.

FIG. 20 illustrates an automatic signal detection workflow according to an embodiment of the present invention.

FIG. 21 illustrates the components of a Dynamic Spectrum Usage and Sharing model in accordance with an embodiment of the present invention.

FIG. 22 illustrates a results model provided by the system according to an embodiment of the present invention.

FIG. 23 is a table listing problems that can be solved using the present invention.

FIG. 24 illustrates a view of a passive geolocation radio engine system according to an embodiment of the present invention.

ινΐΛ/a/zuzz/ui or/oo

FIG. 25 illustrates one embodiment of an algorithm for selecting a geolocation method.

FIG. 26 is a diagram that describes three pillars of a customer mission solution.

FIG. 27 is a block diagram of an example of a spectrum management tool.

FIG. 28 is a block diagram of one embodiment of a resource brokering application.

FIG. 29 illustrates another example of a system diagram that includes an automated semantic engine and translator.

FIG. 30 illustrates a flowchart of a method for obtaining actionable data based on customer objectives.

FIG. 31 illustrates a flowchart of a method of implementing knowledge decision gates and actionable data from the total signal flow.

FIG. 32 illustrates a flowchart of a method for identifying knowledge decision gates based on operational knowledge.

FIG. 33 illustrates an overview of an example of information used to provide knowledge.

FIG. 34 is a map showing locations of three macro sites, 3 SigBASE units and a plurality of locations evaluated for the deployment of alternative or additional sites for a first example.

FIG. 35 is a user distribution graph by average downlink Physical Resource Block (PRB) allocation for the first example.

FIG. 36 illustrates the rate of overutilization events and the degree of overutilization for the first example.

FIG. 37A is a sector coverage map for three macro sites for the first example.

FIG. 37B illustrates the signal intensity for the sector shown in FIG. 37A for the first example.

FIG. 37C illustrates the subscriber density for the sector shown in FIG. 37A for the first example.

FIG. 37D illustrates the carrier/interference relationship for the sector shown in FIG. 37A for the first example.

FIG. 38A illustrates the reference scenario shown in FIG. 34 for the first example.

FIG. 38B is a map showing the locations of the three original macrosites and two additional macrosites for the first example.

ινΐΛ/a/zuzz/uio/oo

FIG. 39 illustrates the signal intensity of the reference scenario of FIG. 38A on the left and the stage with two additional macrosites of FIG. 38B on the right for the first example.

FIG. 40A illustrates the carrier/interference ratio of the reference scenario of FIG. 38A for the first example.

FIG. 40B illustrates the carrier/interference scenario scenario with two additional macro sites for the first example.

FIG. 41 illustrates a reference scenario for a second example on the left and a map showing the locations of the original macrosites of the reference scenario with three additional macrosites proposed for the second example on the right.

FIG. 42 illustrates the signal intensity of the reference scenario in FIG. 41 on the left and the stage with three additional proposed macrosites from FIG. 41 on the right for the second example.

FIG. 43A illustrates the carrier/interference relationship for the reference scenario of FIG. 41 for the second example.

FIG. 43B illustrates the carrier/interference scenario scenario with three additional proposed macrosites from FIG. 41 on the right for the second example.

FIG. 44 illustrates a signal strength comparison of a first carrier (Carrier 1) with a second carrier (Carrier 2) for 700 MHz for a third example.

FIG. 45 illustrates the carrier/interference relationship for Carrier 1 and Carrier 2 for the third example.

FIG. 46 is a graph of Area vs. RSSI and Traffic vs. RSSI for Carrier 1 and Carrier 2 for the third example.

FIG. 47 is a traffic difference graph for Carrier 1 versus Carrier 2 for the third example.

FIG. 48 is a plot of SNR versus RSRP for each SigBASE for the third example.

FIG. 49 is another graph of SNR vs RSRP for each SigBASE for the third example.

FIG. 50 is a clustering plot of SNR versus RSRP for each SigBASE for the third example.

FIG. 51 is another clustering plot of SNR vs. RSRP for each SigBASE for the third example.

FIG. 52 is a schematic diagram of a system of the present invention.

DETAILED DESCRIPTION

The present invention is generally directed to spectrum analysis and management for electromagnetic signals and, more particularly, to providing prioritized and dynamic spectrum management and utilization.

In one embodiment, the present invention provides a system for the dynamic and prioritized management and utilization of spectrum in an electromagnetic environment that includes at least one monitoring sensor operable to monitor the electromagnetic environment, thereby creating measured data, at least one analysis engine. to analyze the measured data, at least one application, a semantic engine that includes a policy editor and programmable rules, and a suggestion and signal server, wherein the data analysis engine(s) includes a detection engine and a learning engine, wherein the detection engine is operative to automatically detect at least one signal of interest, and wherein the learning engine is operative to learn the electromagnetic environment, wherein the application(s) include an occupancy survey application and a resource brokering application, wherein the occupancy survey application operates to determine occupancy in frequency bands and schedule occupancy in at least one frequency band, and wherein the resource brokering application is operative to optimize resources to improve the performance of at least one frequency band. a client application and/or at least one client device, wherein the programmatic rules and policies editor includes at least one rule and/or at least one policy, and wherein one or more of the rule(s) and/or policy(ies) are defined by at least one client, wherein the suggestion and signal server functions to use the analyzed data from the data analysis engine(s) to create procedural data, where each of the client's application(s) is assigned a priority, and wherein the priority and one or more of the rule(s) and/or policy(ies) are used to dynamically assign the frequency band(s) in the electromagnetic spectrum.

In another embodiment, the present invention provides a system for the dynamic and prioritized management and utilization of spectrum in an electromagnetic environment that includes at least one operational monitoring sensor to monitor the electromagnetic environment, thereby creating measured data, at least one analysis engine. to analyze the measured data, at least one application, a semantic engine that includes a policy editor and programmable rules, and a suggestion and signal server, wherein the data analysis engine(s) include a detection engine, an identification engine, a classification engine, a geolocation engine, and a learning engine, wherein the detection engine may function to automatically detect at least one signal of interest, and wherein the learning engine is operative to learn the electromagnetic environment, wherein the application(s) include an occupancy survey application and a resource brokering application, wherein the occupancy survey application is operative to determine occupancy in frequency bands and schedule occupancy in at least one frequency band. , and wherein the ινΐΛ/a/zuzz/ui or/oo resource brokering application is operative to optimize resources to improve the performance of at least one client application and/or at least one client device, wherein the programmable rules and policies includes at least one rule and/or at least one policy, and where one or more of the rules and/or policies are defined by at least one client, wherein the hint and signal server is operative to use the analyzed data from the data analysis engine(s) to create actionable data, wherein each of the client application(s) is assigned a priority, and wherein the priority and one or more of the standard(s) and/or policy(ies) are used to dynamically allocate the frequency band(s) in the electromagnetic spectrum.

In yet another embodiment, the present invention provides a method for the dynamic and prioritized management and utilization of spectrum in an electromagnetic environment that includes providing a semantic engine that includes a programmable policies and rules editor, wherein the programmable policies and rules editor includes at least one rule and/or at least one policy, and where one or more of the rule(s) and/or the policy(ies) are defined by at least one client, monitoring the electromagnetic environment using at least one monitoring sensor, thereby creating measured data, analyzing the measured data using at least one data analysis engine, thereby creating analyzed data, wherein the data analysis engine(s) includes a detection and a learning engine, learning the electromagnetic environment using the learning engine, automatically detecting at least one signal of interest using the detection engine, determine the location of the signal(s) of interest using the geolocation engine, determine occupancy in the frequency bands and schedule occupancy in at least one frequency band using an occupancy survey application, optimize resources to improve the performance of at least one client application and/or at least one client device using a resource brokering application, assign a priority to each of the customer's application(s), and dynamically assign the frequency band(s) in the electromagnetic spectrum based on the priority and one or more of the standard(s) and/or policy(ies).

In yet another embodiment, the present invention provides a method for the dynamic and prioritized management and utilization of spectrum in an electromagnetic environment that includes providing a semantic engine that includes a programmable policy and rules editor, wherein the policy and rules editor programming includes at least one rule and/or at least one policy, and where one or more of the rule(s) and/or the policy(ies) are defined by at least one client, monitoring the electromagnetic environment using at least one monitoring sensor, thereby creating measured data, analyzing the measured data using at least one data analysis engine, thereby creating analyzed data, wherein the data analysis engine(s) includes a detection, a classification engine, an identification engine, a geolocation engine and a learning engine, learning the electromagnetic environment using the learning engine, automatically detect at least one signal of interest using the detection engine, classify the signal(s) of interest using the classification engine, identify the signal(s) of interest using the identification engine, determine the location of the signal(s) of interest using the geolocation engine, determine the location of the signal(s) of interest using the geolocation engine, determine occupancy in frequency bands and schedule occupancy in at least one frequency band using an occupancy survey application, optimize resources to improve the performance of at least one customer application and/or at least one customer device using a resource brokering application, assign a priority to each of the client application(s), and dynamically assign the frequency band(s) in the electromagnetic spectrum based on priority and one or more of the standard(s) and/or policy(ies).

Traditional spectrum management is static, based on geographic and band-specific licenses. The Federal Communications Commission (FCC) has allocated spectrum in a table. Utilization increases by dividing the spectrum into finer parts. In addition, interference is limited by imposing sanctions through strict geographic band usage rules and licenses. However, these traditional methods of spectrum management do not work with the increase in demand and the emergence of new services. New services would have to be at higher frequencies (e.g. above 10 GHz), which is very expensive and requires an expensive transceiver with a limited distance range.

Spectrum is valuable because it is a finite resource. In addition, the demand for spectrum is increasing. The Shannon-Hartley theorem calculates the maximum rate at which information can be transmitted over a communications channel of a specific bandwidth in the presence of noise as follows:

C = BW log₂(l +SNR) where C is the channel capacity in bits per second, BW is the channel bandwidth in Hz and SNR is the signal-to-noise ratio.

Early attempts to manage spectrum included the development of technology that increases spectrum efficiency (i.e., maximizes SNR). Although this results in more bits per Hz, the logarithmic function limits the gains in channel capacity that result from improved technology. Additional attempts to manage spectrum also include the development of technology to enable the use of alternative spectrum (e.g. (e.g., free space optical (FSO) communication). However, the use of alternative spectrum, such as higher frequencies, leads to smaller ranges, line-of-sight limitations, higher elevation of transmission structures, and/or costly infrastructure.

The missing component in spectrum management is bandwidth management. Bandwidth management provides flexible spectrum utilization, allows management of spectrum resources and users, while allowing spectrum usage to be quantified. Most applications that use spectrum can coexist if each application knows the spectrum needs of other applications and how they plan to use the spectrum. However, since the needs of each application are dynamic, a dynamic spectrum management system is needed. The present invention enables autonomous and dynamic sharing of electromagnetic spectrum to enable maximum utilization by various applications in accordance with specific utilization standards (dynamic and/or static) while maintaining minimal interference between applications. This requires new tools that provide dynamic spectrum environment awareness of all signals present in the electromagnetic environment (e.g., RF) to correctly execute utilization rules, which are operational to describe or facilitate the sharing of data resources. spectrum between several competing users or protecting one user from the service of others, among others.

5G requires spectrum awareness. Larger blocks of spectrum are required to support higher speeds. Dynamic spectrum sharing is necessary to make spectrum assets available. Additionally, visibility into spectrum activity is required to support reliability objectives. Interference prevention and resolution must be integrated. The wireless reliance on machine/IoT communication raises the need for real-time RF visibility to avoid outages and security issues.

The system of the present invention provides scalable processing capabilities at the edge. Edge processing is fast and reliable with low latency. Environmental sensing processes streamline collection and analysis, making data sets manageable. Advantageously, the system minimizes backhaul requirements, allowing procedural data to be delivered more quickly and efficiently.

Deep learning techniques extract and deliver knowledge from large data sets in near real time. These deep learning techniques are essential for identifying and classifying signals. Dordes analysis allows third-party data (e.g., social media system, podation information, real estate information, traffic information, geographic information) to further enrich the captured data sets. A semantic engine and inference reasoner leverage insights generated by machine learning and edge analysis. Ontologies are established that allow the creation of knowledge that works to inform and guide actions and/or decisions.

Referring now to the drawings in general, the illustrations are intended to describe iviA/a/zuzz/ui or/oo one or more preferred embodiments of the invention and are not intended to limit the invention thereto.

The present invention provides systems, methods and apparatus for spectrum analysis and management by identifying, classifying and cataloging at least one or a multiplicity of signals of interest based on measurements of the electromagnetic spectrum (e.g., measurements of the radio frequency spectrum), location and other measurements. The present invention uses real-time and/or near real-time signal processing (e.g. e.g., parallel processing) and corresponding parameters and/or signal characteristics in the context of historical, static and/or statistical data for a given spectrum, and more particularly, all use reference data and changes in the state of the data compressed to enable near real-time analysis and results for individual monitoring sensors and for aggregated monitoring sensors to make unique comparisons of data.

The systems, methods and apparatus according to the present invention are preferably operable to detect in near real time, and more preferably to detect, measure and/or analyze in near real time, and more preferably to perform any operation in near real time in about 1 second or less. In one embodiment, near real-time is defined as calculations completed before the data marks an event change. For example, if an event occurs every second, near real time means that calculations are completed in less than a second. Advantageously, the present invention and its real-time functionality described herein provides and allows the system to compare acquired spectrum data with historical data, to update the data and/or information and/or to provide more data and/or information in open spaces. . In one embodiment, the information (p. e.g., open space) is provided in an apparatus unit or a device that occupies the open space. In another embodiment, the system compares acquired data with historically scanned data (e.g., 15 min to 30 days) and/or historical database information in near real time. Additionally, data from each monitoring sensor, appliance unit or device and/or aggregate data from more than one monitoring sensor, appliance unit and/or device is communicated over a network to at least one server and stored. in a database on a virtualized or cloud-based computer system, and the data is available for secure remote access over the network from distributed remote devices that have software applications (applications) operable thereon, for example, through web access (mobile application) or computer access (desktop application ). The server or servers are operational to analyze the data and/or the aggregated data.

The system is operational to monitor the electromagnetic environment (e.g., RF) through at least one monitoring sensor. The system is then operational to analyze the data acquired from the control sensor(s) to detect, classify and/or identify at least one signal in the electromagnetic environment. The system is operational to learn the electromagnetic environment, which ινΐΛ/a/zuzz/ui ó/oo allows the system to extract knowledge from the environment. In a preferred embodiment, the system extracts knowledge from the environment by including the client's objectives. Environmental awareness is combined with client objectives, client-defined policies, and/or standards (e.g., client-defined standards, government-defined standards) to extract useful information to help the client optimize performance according to customer objectives. Processed information is combined and correlated with additional information sources to improve customer insight and user experience through dynamic spectrum utilization and prediction models.

The systems, methods and apparatus of the various modalities allow the management of spectrum utilization by identifying, classifying and cataloging signals of interest based on electromagnetic (e.g., radio frequency) measurements. In one embodiment, signals and signal parameters are identified. In another embodiment, indications of available frequencies are presented to a user and/or user equipment. In yet another embodiment, signal protocols are also identified. In a further embodiment, the modulation of the signals, the types of data carried by the signals, and the estimated origins of the signals are identified. Identification, classification and cataloging signals of interest are preferably produced in real time or near real time.

The embodiments are directed to a spectrum monitoring unit that is configurable to obtain spectrum data over a wide range of wireless communication protocols. The embodiments also provide the ability to acquire data from and send data to database repositories that are used by a plurality of spectrum management clients and/or applications or services that require spectrum resources.

In one embodiment, the system includes at least one spectrum monitoring unit. Each of the spectrum monitoring unit or units includes at least one monitoring sensor that is preferably in network communication with a database system and a spectrum management interface. In one embodiment, the spectrum monitoring unit(s) and/or the monitoring sensor(s) are portable. In a preferred embodiment, one or more of the spectrum monitoring unit(s) and/or the monitoring sensor(s) is a stationary installation. The spectrum monitoring unit(s) and/or the monitoring sensor(s) are operative to acquire different spectrum information including, but not limited to, frequency, bandwidth, signal power, time and location of the spectrum. signal propagation, as well as the type and format of modulation. The spectrum monitoring unit or units are preferably operational to provide identification, classification and/or geolocation of signals. Furthermore, the spectrum monitoring unit(s) preferably includes a processor to allow the spectrum monitoring unit(s) to process iviA/a/zuzz/uid/oo spectrum power density data as received and/or to process complex raw Phase and Quadrature (I/Q) data. Alternatively, the spectrum monitoring unit(s) and/or the monitoring sensor(s) transmit the data to at least one data analysis engine for storage and/or processing. In a preferred embodiment, the transmission of the data is performed through a backhaul operation. Spectrum power density data and/or raw complex I/Q data can be used for signal processing, signal identification, and additional data extraction.

Preferably, the system functions to manage and prioritize spectrum utilization based on five factors: frequency, time, space, signal space, and application objectives.

The frequency range is preferably as large as possible. In one embodiment, the system supports a frequency range between 1 MHz and 6 GHz. In another embodiment, the system supports a frequency range with a lower limit of 9 kHz. In yet another embodiment, the system supports a frequency range with an upper limit of 12.4 GHz. In another embodiment, the system supports a frequency range with an upper limit of 28 GHz or 36 GHz. Alternatively, the system supports a frequency range with an upper limit of 60 GHz. In yet another embodiment, the system supports a frequency range with an upper limit of 100 GHz. The system preferably has an instantaneous processing bandwidth (IPBW). ) of 40 MHz, 80 MHz, 100 MHz or 250 MHz per channel.

The time range is preferably as wide as possible. In one embodiment, the number of samples per dwell time in a frequency band is calculated. In one example, the system provides a minimum coverage of 2 seconds. The number of samples per time spent in the frequency band is calculated as follows:

N_yes > (IPBW)(2)/channel

The required buffering is a minimum of 2 seconds per channel per dwell time, which is calculated as follows:

Storage = (/PBW)(2)(2 Bytes)(channels)/(dwell time)

Spatial processing is used to divide a coverage area by a range of elevation and azimuth angles. The coverage area is defined as an area under a certain azimuth and range. This is implemented through antenna array processing, steerable beamforming, array processing, and/or directional antennas. In one embodiment, the directional antennas include at least one steerable electrical or mechanical antenna. Alternatively, directional antennas include a steerable antenna array. More antennas require more signal processing. Advantageously, spatial processing allows for better signal separation, reduction of noise and interfering signals, geospatial separation, increased signal processing gains, and provides a spatial component for signal identification. Additionally, this allows for simple integration of geolocation, iviA/a/zuzz/ui or/oo techniques such as TDOA, AOA and/or FDOA. This also allows for the implementation of a geolocation engine, which will be discussed in detail later.

Each signal has inherent signal characteristics including, but not limited to, a type of modulation (e.g., FM, AM, QPSK, QAM, BPSK, etc.), a protocol used (e.g., without protocol for analog signals, DMR, LRM, P25, NXDN, cellular, LTE, UMTS, 5G), an enveloping behavior (e.g. e.g., bandwidth (BW), center frequency (FC), symbol rate, data rate, constant envelope, peak power to average power ratio (PAR), cyclostationary properties), an interference index, and statistical properties ( e.g., stationary decomposition, cyclostationary decomposition, higher momentum decomposition, nonlinear decomposition (e.g., Volterra series to cover nonlinearities, basic learning model).

The application objectives depend on the particular application used within the system. Examples of applications used in the system include, but are not limited to, traffic management, telemedicine, virtual reality, video streaming for entertainment, social media, autonomous and/or unmanned transportation, etc. Each application works to be prioritized within the system based on the client's objectives. For example, traffic management is a higher priority application than streaming video for entertainment.

As described above, the system functions to monitor the electromagnetic environment (e.g., RF), analyze the electromagnetic environment, and extract knowledge from the electromagnetic environment. In a preferred embodiment, the system extracts knowledge of the electromagnetic environment by including the client's objectives. In another embodiment, the system uses recognition of the environment with the client's objectives and/or user-defined policies and standards to extract processable information to help the client optimize the client's objectives. The system combines and correlates other information sources with extracted actionable information to improve customer insight through dynamic spectrum utilization and prediction models.

FIG. 1 illustrates one embodiment of an RF recognition and analysis system. The system includes an RF recognition subsystem. The RF recognition subsystem includes, but is not limited to, an antenna subsystem, an RF conditioning subsystem, at least one front receiver, a programmable channeler, a blind detection engine, a blind classification engine, a module envelope feature extraction, a demodulation bank, an automatic gain control (AGC) dual-loop subsystem, a signal identification engine, a feature extraction engine, a learning engine, a geolocation engine, a data analysis engine and/or a database that stores information related to at least one signal (e.g., metadata, timestamps, power measurements, frequencies, etc. ). The system further includes an alarm system, a visualization subsystem, a knowledge engine, an operational semantic engine, a customer optimization module, a database of customer objectives and operational knowledge and/or a database of decisions and actionable data.

The antenna subsystem monitors the electromagnetic environment (e.g., RF) to produce monitoring data. Monitoring data is then processed through the RF conditioning subsystem before being processed through the front-end receivers. The AGC double loop subsystem functions to perform AGC adjustment. Front receivers convert data from analog to digital.

The digital data is then sent through the programmable pipeliner and undergoes I,Q buffering and masking. A fast Fourier transform (FFT) is performed and the blind detection engine performs blind detection. Additionally, the blind sorting engine performs blind sorting. Information (e.g., observed channels) is shared from the blind detection engine to the blind classification and/or programmable channeler (e.g. (e.g., to inform selection logic and processes). The information from the blind detection engine is also sent to the surround feature extraction module. The information from the blind sorting engine is sent to the demodulation bank.

The information from the envelope feature extraction module, the demodulation bank and/or the blind classification engine functions for use by the signal identification engine, the feature extraction engine, the learning engine and/or the geolocation. The information of AGC double loop subsystem, I,Q buffer, masking, programmable channeler, signal identification engine, feature extraction engine, learning engine and geolocation engine, extraction module of envelope characteristics, the demodulation bank and/or the blind classification engine are operational to be stored in the database that stores the information related to the signal(s) (e.g. (e.g. signal data, metadata, timestamps).

The database information (i.e., the database that stores information related to the signal or signals), the signal identification engine, the feature extraction engine, the learning engine and/or the Geolocation can be sent to the data analysis engine for further processing.

The alarm system includes database information that stores information related to the signal(s) and/or the customer's target database and operating knowledge. Alarms are sent from the alarm system to the display subsystem. In a preferred embodiment, the display subsystem customizes a graphical user interface (GUI) for each client. The visualization system functions to present information from the base ινΐΛ/a/zuzz/ui ó/oo of decision data and actionable data. In one embodiment, alarms are sent via text message and/or email. In one embodiment, alarms are sent to at least one IP address.

The database of client objectives and operational knowledge also functions to send information to a semantic engine (e.g., client objectives and alarm conditions) and/or an operational semantic engine (e.g., operational knowledge the client's). The semantic engine translates the information into constraints and sends the constraints to the client optimization module, which also receives information (e.g., signal metadata) from the data analysis engine. The client optimization module may function to send actionable data related to the electromagnetic environment to the operational semantic engine. The client optimization module works to discern which information (e.g., environmental information) has the greatest statistically sufficient impact relative to the client's objectives and operation.

In one embodiment, the system includes at least one monitoring sensor, at least one data analysis engine, at least one application, a semantic engine, a policy and programmable rules editor, a suggestion and signal server, and/or a control panel as shown in FIG. 2.

The monitoring sensor or sensors include at least one radio server and/or at least one antenna. The antenna(s) is a single antenna (e.g., unidirectional or directional) or an antenna array consisting of multiple antennas resonating in different frequency bands and antenna configuration configured in ID (linear), 2D (planar), or 3D (area). The monitoring sensor or sensors are operational to explore the electromagnetic spectrum (e.g. e.g., RF) and measure properties of the electromagnetic spectrum, including, but not limited to, receiver I/Q data. Preferably, the monitoring unit(s) are operative to autonomously capture the electromagnetic spectrum with respect to frequency, time and/or space. In one embodiment, the monitoring sensor(s) are operative to perform array processing.

In another embodiment, the monitoring sensor or sensors are mobile. In one embodiment, the monitoring sensor or sensors are mounted on a vehicle or a drone. Alternatively, the monitoring sensor(s) are fixed. In one embodiment, the monitoring sensor or sensors are fixed in or on a road luminaire and/or traffic light. In yet another embodiment, the monitoring sensor or sensors are fixed to the top of a building.

In one embodiment, the monitoring sensor or sensors are integrated with at least one camera. In one embodiment, the camera or cameras capture video and/or still images.

In another embodiment, the monitoring sensor or sensors include at least one monitoring unit. Examples of monitoring units include those described in US Pat.

ινΐΛ/a/zuzz/ui or/oo

η.°10,122,479, 10,219,163, 10,231,206, 10,237,770, 10,244,504, 10,257,727, 10,257,728, 10,257,729, 10,271,233, 10,299,149, 10,498,951, and 10,529,241, and U.S. Publication Nos. 20190215201, 20190364533, and 20200066132, each of which It is incorporated herein by reference in its entirety.

In a preferred embodiment, the system includes at least one data analysis engine for processing data captured by the monitoring sensor(s). An engine is a collection of functions and algorithms used to solve a class of problems. The system preferably includes a detection engine, a classification engine, an identification engine, a geolocation engine, a learning engine and/or a statistical inference and machine learning engine. For example, the geolocation engine is a group of geolocation functions and algorithms that are used together to solve multiple geolocation problems.

The detection engine preferably operates to detect at least one signal of interest in the electromagnetic (e.g., RF) environment. In a preferred embodiment, the detection engine may operate to automatically detect the signal or signals of interest. In one embodiment, the automatic signal detection process includes creating masks and analyzing the environment using masks. Masking is a process of creating a representation of the electromagnetic environment by analyzing a spectrum of signals over a certain period of time. A desired frequency range is used to create a mask, and FFT streaming data is also used in the mask creation process. A first derivative is calculated and used to identify possible maximum power values. A second derivative is calculated and used to confirm the maximum power values. A moving average value is created as FFT data is received over a user-selected time period for mask creation. For example, the time period is 10 seconds. The result is an FFT array with an average of maximum power values, which is called a mask.

The classification engine preferably functions to classify the signal or signals of interest. In one embodiment, the classification engine generates a query to a static database to classify the signal or signals of interest based on its components. For example, the information stored in the static database is preferably used to determine the spectral density, center frequency, bandwidth, baud rate, modulation type, protocol (e.g. GSM, CDMA, OFDM, LTE, etc.), the system or carrier using licensed spectrum, the location of the signal source, and/or a timestamp of the signal(s) of interest. In one embodiment, the static database includes frequency information collected from various sources, including, but not limited to, the Federal Communications Commission, the International Telecommunications Union, and user data. As an example, the static database iviA/a/zuzz/uid/oo may be an SQL database. The data warehouse can be updated, downloaded, or merged with other devices or with your main relational database. In one embodiment, software API applications are included to enable merging of databases with third-party spectrum databases that can only be operated for secure access. In a preferred embodiment, the classification engine functions to calculate second, third and fourth order cumulants to classify the modulation schemes along with other parameters, including center frequency, bandwidth, baud rate, etc.

The identification engine preferably functions to identify a device or transmitter that transmits the signal or signals of interest. In one embodiment, the identification engine uses signal profiling and/or comparison with known database(s) and previously recorded profile(s) to identify the device or emitter. In another embodiment, the identification engine establishes a level of trust related to the identification of the device or the issuer.

The geolocation engine preferably functions to identify a location from which the signal or signals of interest are broadcast. In one embodiment, the geolocation engine uses statistical approaches to eliminate causes of error from noise, timing and power measurements, multipath and non-line-of-sight (NLOS) measurements. As an example, the following methods are used for geolocation statistical approximations and variances: maximum likelihood (nearest neighbor or Kalman filter); least squares approximation; Bayesian filter if prior knowledge data is included; and the like. In another embodiment, time difference of arrival (TDOA) and frequency difference of arrival (FDOA) equations are derived to help resolve inconsistencies in distance calculations. In yet another embodiment, the angle of arrival (AOA) is used to determine geolocation. In yet another embodiment, measurements of power distribution ratio versus azimuth are used to determine geolocation. In a preferred embodiment, geolocation is performed using measurements of angle of arrival (AOA), time difference of arrival (TDOA), frequency difference of arrival (FDOA) and power distribution ratio. Various methods or combinations of these methods can be used with the present invention because geolocation is performed in different environments, including, but not limited to, indoor environments, outdoor environments, hybrid environments (stadium), city center environments , etc.

The learning engine preferably works to learn the electromagnetic environment. In one embodiment, the learning engine uses statistical learning techniques to observe and learn an electromagnetic environment over time and identify temporal characteristics of the electromagnetic environment (e.g., signals) during a learning period. In a preferred embodiment, the learning engine is operative to learn information from the detection engine, the classification engine, the identification engine and/or the geolocation engine. In a ινΐΛ/a/zuzz/ui or/oo mode, the system learning function is operative to enable and disable. When the learning motor is exposed to a stable electromagnetic environment and has learned what is normal in the electromagnetic environment, it will stop its learning process. In a preferred embodiment, the electromagnetic environment is reassessed periodically. In one embodiment, the learning engine reassesses and/or updates the electromagnetic environment in a predetermined time frame. In another embodiment, the learning engine re-evaluates and/or updates the electromagnetic environment after detecting a problem.

The statistical inference and machine learning (ML) engine uses statistical learning and/or control theory techniques to learn the electromagnetic environment and make predictions about the electromagnetic environment.

The occupancy survey application is operational to determine occupancy in frequency bands. In another embodiment, the occupancy survey application is operative to schedule occupancy in a frequency band. The occupancy survey application is also used to preprocess at least two signals that exist on the same band based on interference between the two signals or more.

The resource brokering application is operational to optimize resources and improve application performance. In a preferred embodiment, the resource brokering application is operative to use processed data from the monitoring sensor(s) and/or additional information to determine environmental awareness (e.g., environmental situation awareness). Awareness of the environment and/or capabilities of a device and/or resource are used to determine policies and/or reasoning to optimize the device and/or resource. The resource brokering application is operative to control the device and/or the resource. Additionally, the resource brokering application is operative to control the monitoring sensor(s).

The certification and compliance application is operational to determine whether applications and/or devices behave in accordance with standards and/or policies (e.g., customer policies and/or standards, government regulations). In another embodiment, the certification and compliance application is operational to determine whether applications and/or devices are sharing frequency bands in accordance with standards and/or policies. In yet another embodiment, the certification and compliance application is operative to determine whether applications and/or devices are behaving in accordance with non-interference standards and/or policies.

Application sharing is operative to determine the optimization of how applications and/or devices share frequency bands. In a preferred embodiment, the sharing application uses a plurality of rules and/or policies (e.g., a plurality of customer rules and/or policies, government rules) to determine the optimization of how ινΐΛ/a/zuzz/ ui or/or applications and/or devices share the frequency bands. Therefore, the sharing application satisfies the plurality of rules and/or policies defined by at least one client and/or the government.

The application of statistical utilization, prediction and inference works to utilize predictive analytics techniques including, but not limited to, machine learning (ML), artificial intelligence (AI), neural networks (NN), historical data and/or mining of data to make future predictions and/or models. The system is preferably operational to recommend and/or perform actions based on historical data, external data sources, ML, AI, NN and/or other learning techniques.

The semantic engine functions to receive data in forms including, but not limited to, audio data, text data, video data, and/or image data. In one embodiment, the semantic engine uses a set of system rules and/or a set of system policies. In another embodiment, the system rule set and/or system policy set is created using a prior knowledge database. The semantic engine preferably includes an editor and a language dictionary.

The semantic engine preferably also includes a programmable rules and policy editor. The programmable policies and rules editor is operative to include at least one rule and/or at least one policy. In one embodiment, the rule(s) and/or policy(ies) are defined by at least one client. Advantageously, this allows the client or clients to dictate rules and policies related to the client's objectives.

The system also includes a suggestion and signal server. The hint and signal server is operational and uses environmental awareness of the data processed by the data analysis engine(s) in combination with additional information to create actionable data. In a preferred embodiment, the hint and signal server uses information from a set of specific rules (e.g. e.g., a customer-defined set of rules), further enhancing the system's optimization capabilities. The set of specific rules translates into optimization objectives, including constraints associated with signal characteristics. In a preferred embodiment, the hint and signal server is operative to activate at least one alarm and/or provide at least one report. In another embodiment, the hint and signal server is operative to activate the alarm(s) and/or provide the report(s) in accordance with the specific set of standards.

Advantageously, the system is operative to function autonomously and continuously. The system learns from the environment and, without operator intervention, can operate to detect anomalous signals that did not exist before or that have changed in power or bandwidth. Once detected, the system works to send alerts (e.g. e.g. via text message or email) iviA/a/zuzz/ui or/oo bandwidth, gain, etc.) and then stored in the I,Q buffer before the analysis is completed.

Advantageously, the system is hardware independent. The system is operational to provide a hardware suggestion for a particular set of frequencies. Additionally, the hardware-independent nature of the system allows the established architecture to persist. The system is cost-effective because it also allows the use of cheaper antennas, as well as less expensive filters, because calibration can be performed using the system instead of the antennas and/or filters, as well as post-ADC processing to rectify any loss of performance. Because the system processes all the signals present in the spectrum and their interrelationships to extract knowledge of the environment, the analog front-end does not require filtering to avoid interference and provide optimal dynamic range. Furthermore, the analog front-end does not require optimal antennas for all bands and frequency ranges for environmental awareness.

For a programmable time domain channelizer, all the impulse responses of the filters must be programmable and the number of filters must be programmable. Additionally, the channel bandwidth resolution must be programmable from a minimum bandwidth. The center frequency of each channel must also be programmable. Decimation is based on channel bandwidth and desired resolution. However, these requirements are difficult to implement for channels with variable bandwidth and center frequency. Wavelet filters can be used effectively if the center frequency and channel bandwidth follow a tree structure (e.g., Harr and Deubauchi wavelets). FIG. 8 is an example of a time domain programmable channelizer.

In a preferred embodiment, the system includes a frequency domain programmable channelizer as shown in FIG. 9. The programmable pipeliner includes buffering services, preprocessing, bin selection, at least one bandpass filter (BPF), an inverse fast Fourier transform (IFFT) function, frequency decomposition and/or downconversion and correction to produce the baseband I, Q for channels 1 to R. The IFFT function is performed to obtain each I,Q channel decomposed to the appropriate sampling rate. Advantageously, the programmable channelizer in the frequency domain is more computationally efficient than a programmable channelizer in the time domain because each filter is just a vector in the frequency domain and the filtering operation is just a multiplication of vectors. , Decomposing the input signal into multiple channels of different bandwidths is parsing the vector that represents the frequency domain content of the input signal into a subvector of different length.

FIG. 10 is another embodiment of a programmable channeler. Data enters filter and channel generators with pipeline selector logic for a table lookup of filter coefficients and pipeline vectors. The programmable channelizer includes a comparison on each channel, providing anomalous detection using a mask with frequency and power, which is then sent to the learning engine and/or alarm system (A). The data processed with the FFT is sent to the blind detection engine and/or for average processing (B). The filter coefficient lookup table data and pipeline vectors are preprocessed with a mixed circular rotator to produce Di blocks of Ri points. A sum is taken from the Di block, and an IFFT is taken from the Ri point. to produce OLi discord overlay samples. This process occurs (p. e.g., in parallel) for Di blocks of Ri points through Dr blocks of Rr points to produce OLi to OLr, which are then sent to the classification engine (C). All data in the I,Q buffer is preferably stored in a buffered database (D). In one embodiment, the buffer I,Q is divided into N blocks with L oversamples. In one embodiment, the original sampling rate is decimated by D¡.

FIG. 11 illustrates one embodiment of a blind detection engine. The programmable channelizer data is subjected to an N-point FFT. A power spectral density (PSD) is calculated for each N-point FFT and then a complex average FFT is obtained for the P blocks of the N-point FFT. PSD is sent to a noise floor estimator, edge detection algorithm, and/or isolator. The noise floor estimates from the noise floor estimator are sent to the signal database. The edge detection algorithm passes information to a signal separator (e.g., bandwidth, center frequency). The isolator obtains information including, but not limited to, the per-channel PSD, bandwidth and center frequency, complex average FFT, and/or N-point FFT. The isolator information is sent to the programmable pipeliner, the envelope feature extraction module, and/or the classification engine.

FIG. 12 illustrates one embodiment of an edge detection algorithm. Peaks are detected for all power values ​​above the noise floor. The spikes are recorded on a power array and/or an index array. Consecutive power values ​​are found by cycling through the arrays. For each group of consecutive power values, a subpower array and/or a subindex array is created. The blind detection engine cycles through each power value starting with a predetermined ascending threshold. If consecutive N values ​​increase above the ascending threshold, a first value of the N values ​​is set as a rising edge and the index of the first value of the N values ​​is recorded. The N value is recorded as an ascending reference point. The rising threshold is updated based on the rising setpoint and the blind detection engine continues searching for rising values. If the blind detection engine does not detect rising values ​​and detects consecutive M values ​​decreasing below a falling threshold, a first value of the M values ​​is set as the falling edge ινΐΛ/a/zuzz/ui or/oo and the index of the first value of M values. The M value is recorded as a descending reference point. The falling threshold is updated based on the falling reference point. In one embodiment, x is a value between 1 dB and 2.5 dB. In one embodiment, y is a value between 1 dB and 2.5 dB.

The blind classification engine receives information from the blind detection engine as shown in FIG. 13. Signals are separated based on bandwidth and/or other envelope properties (e.g., duty cycle). An IFFT is performed on R signals for narrowband and/or wideband signals. Next, decimation is performed based on bandwidth. Momentum calculations are performed for each I,Q signal using the decimated values ​​and/or channelizer information. In a preferred embodiment, the moment calculations include a second moment and/or a fourth moment for each signal. A cumulant-based match is selected for each I,Q stream, which is sent to the demodulation band and/or the geolocation engine.

From the definitions of the second and fourth moments, the following equations are used to calculate the cumulants:

n=N c₂₀47 iw iN n=ln=N

71=1 n=N

71=1 n=N

c₄₁ =-^Y³(n)Y\n-)-3C₂₀c₂₁ n=ln=N

c₄₂=^|T(n)|⁴ - |C20|² - 2C21²n=l

Assuming that the transmitted constellations are normalized to the unit average power, which is easily completed with a power factor equal to 0 dB, the result is C₂₁ ~1. To calculate a normalized fourth moment, it is calculated using the following equation:

c₄, 4 ^c< for J= 0, 1, 2 ' U21

Advantageously, the normalization of the fourth moment cumulants eliminates any scaling power problem.

FIG. 14 illustrates details on cumulant-based selection matching for modulation selection. As described above, the cumulants preferably include a second moment and/or a fourth moment for each signal. For example, a

MA/a/ZUZZ/Ul d/OO fourth moment between -0.9 and 0.62 is a QAM signal, a fourth moment greater than or equal to 1 is an AM signal, a fourth moment equal to -1 is an envelope signal constant (e.g., GMSK, FSK, or PSK), a fourth moment between -1.36 and 1.209 is a PAM signal, and a fourth moment equal to -2 is a BPSK signal. A type is selected using a lookup table, the signal I,Q is labeled with the type, and the information is sent to the demodulation bank.

Additional information on cumulant-based selection matching for modulation selection is available in Table 1 below.

TABLE 1

<td>Guy</td><td>and₄₀</td><td>and₄₂</td><td>σ(0₄₀)</td><td> &lt;¿₄₂)</td>

<td>A.M</td><td></td><td> &gt; 1,0</td><td></td><td></td>

<td>FM</td><td></td><td> -1</td><td></td><td></td>

<td>GMSK</td><td></td><td> -1</td><td></td><td></td>

<td>FSK</td><td></td><td> -1</td><td></td><td></td>

<td>BPSK</td><td> -2,00</td><td> -2,00</td><td> 0</td><td> 0</td>

<td>PAM (4)</td><td> -1,36</td><td> -1,36</td><td> 2,56</td><td> 2,56</td>

<td>PAM (8)</td><td> -1,238</td><td> -1,238</td><td> 4,82</td><td> 4,82</td>

<td>PAM (16)</td><td> -1,2094</td><td> -1,2094</td><td> 5,52</td><td> 5,52</td>

<td>PSK (4)</td><td> -1,00</td><td> -1,00</td><td></td><td></td>

<td>QAM (4)</td><td> -0,68</td><td> -0,68</td><td></td><td></td>

<td>QAM (16)</td><td> -0,64</td><td> -0,64</td><td> 3,83</td><td> 2,24</td>

<td>QAM (32)</td><td> -0,61</td><td> -0,61</td><td> 3,89</td><td> 2,31</td>

FIG. 15 illustrates a flow chart according to an embodiment of the present invention. The data in the I/Q buffer is processed using a library of functions. The library of functions includes, but is not limited to, FFT, peak detection, characterization and/or rate tuning. As described above, the system preferably includes at least one data analysis engine. In one embodiment, the data analysis engine(s) include a plurality of engines. In one embodiment, the plurality of engines includes, but is not limited to, a detection engine, a classification engine, an identification engine, a geolocation engine, and/or a learning engine. Each of the plurality of motors is operative to interact with the other motors in the plurality of motors. The system is operational to scan spectrum occupancy, create a mask, detect drones and/or analyze data.

The control panel manages all data flow between the I/Q buffer, library functions, multiple engines, applications and user interface. A collection of basic functions and a particular sequence of operations are called from each of the plurality of engines. Each of the plurality of engines is operative to pass partially processed data and/or analyze to other engines to improve the functionality of other engines and/or applications. The data from the engines is then combined and processed to create applications and/or features that are customer or market specific.

In one embodiment, a plurality of state machines perform a particular analysis for a client application. In one embodiment, the plurality of state machines is a plurality of nested state machines. In another embodiment, a state machine is used for each engine application. The plurality of state machines are used to control the flow of functions and/or the input/output utilization of an engine to perform the required analyses.

FIG. 16 illustrates the functions of the control panel according to one embodiment. The control panel is operational to detect spectrum occupancy, activate an alarm, perform drone detection and direction finding, geolocation, artificial spectrum verification and provide at least one user interface. The user interface(s) is preferably a graphical user interface (GUI). The user interface(s) (UI) are operative to present output data from the plurality of engines and/or applications. In one embodiment, the UI interface(s) incorporates third-party GIS for coordinate display information. The UI(s) are also operative to display alarms, reports, utilization statistics and/or client application statistics. In one embodiment, the UI(s) include an administrator UI and at least one client UI. The client UI(s) are specific to each client.

In one embodiment, the systems and methods of the present invention provide detection of unmanned vehicles (e.g., drones). The overall system is capable of surveying the spectrum from 20 MHz to at least 6 GHz, not just the common 2.4 GHz and 5.8 GHz bands as in the prior art. The systems and methods of the present invention are operative to detect UV and its protocol drivers.

In one embodiment, the systems and methods of the present invention maintain a state-of-the-art learning system and protocol library to classify detected signals by manufacturer and controller type. The state-of-the-art learning system and protocol library are updated as new protocols emerge.

In one embodiment, classification by protocol chipset is used to provide valuable intelligence and insights for risk mitigation and threat defense. Valuable intelligence and knowledge include effective operating range, compatible peripherals (e.g., external or internal camera, barometers, GPS, and dead reckoning capabilities), integrated obstacle avoidance systems, and interference mitigation techniques.

ινΐΛ/a/zuzz/ui or/oo

Advantageously, the system is operative to detect drones that are not in the protocol library. In addition, the system is operational to detect drones without demodulating the command and control protocols. In one embodiment, the system does not include a protocol library. New protocols and new drones are constantly being launched. Additionally, a nefarious operator can change a drone's chipset, which would leave a vulnerable area on the modified drone because a system would not be able to identify the signal as a drone if the protocol is not in the protocol library. In one embodiment, the system generates procedural data indicating that at least one signal behaves like a drone. The system performs blind detection, allowing the system to detect the drone signal without the protocol library. In one embodiment, the system is operative to detect drones by evaluating an envelope of the command and control signal. In one embodiment, the system detects the drone signal based on a duty cycle and/or changes in signal envelope power levels. In one example, the system classifies an LTE signal as a drone when moving at high speed.

FIG. 17 illustrates one embodiment of a system RF analysis subarchitecture. The control panel interacts with the I/Q buffer, library functions, engines, applications, and/or the user interface. The engines include a data analysis engine. The analyzed data from the data analysis engine generates an alarm when an alarm condition is met. The alarm is transmitted by text message and/or email, or displayed on a graphical user interface (GUI) of at least one remote device (e.g., smartphone, tablet, laptop, desktop computer) .

FIG. 18 illustrates one embodiment of a system detection engine. The detection engine receives data from the monitoring unit or units. The detection engine includes blind feature extraction algorithms. A mask is created. The detection engine then performs a mask utilization assessment and the mask is compared to previous masks. Then, anomalies are detected.

As described above, in one embodiment, the data analysis engine is operative to perform masking and analyze an electromagnetic environment (e.g., RF) using masks. Masking is a process of creating a representation of an electromagnetic environment by analyzing a spectrum of signals over a certain period of time. A mask is created with a desired frequency range (e.g. e.g., as entered into the system via user input), and the FFT streaming data is also used in the mask creation process. A first derivative is calculated and used to identify the maximum power values. A moving average value is created as FFT data is received over a selected time period for mask creation (e.g., through user input). For example, the time period is 10 seconds. The result is an arrangement ινΐΛ/a/zuzz/ui or/oo

FFT with an average of maximum power values, which is called a mask. FIG. 19 illustrates a mask according to one embodiment of the present invention.

In one embodiment, the mask is used for analysis of the electromagnetic environment. In one embodiment, the mask is used to identify potential unwanted signals in an electromagnetic (e.g., RF) environment. The system is operational to use masks based on a priori knowledge and/or masks based on the expected behavior of the electromagnetic environment.

Each mask has an analysis time. During its analysis time, a mask is scanned and live FFT streaming data is compared to the mask before the next mask arrives. If a value above the mask range is detected, a trigger analysis is performed. Each mask has a set of trigger conditions and an alarm is triggered in the system if the trigger conditions are met. In one embodiment, there are three main trigger conditions including an alarm duration, a decibel (dB) offset, and a count. The alarm duration is a window of time in which an alarm must occur to be considered a trigger condition. For example, the time window is 2 seconds. If a signal is seen for 2 seconds, it goes to the next condition. dB offset is a threshold value (i.e. dB value) that a signal needs to be above the mask to be considered a potential alarm. The count is the number of times the first two conditions must occur before an alarm is triggered in the system.

FIG. 20 illustrates an automatic signal detection workflow according to an embodiment of the present invention. A mask definition is specified by the user for an automatic signal detection process that includes creating masks, storing masks, and performing electromagnetic analysis of the environment (e.g., RF) based on the created masks and the FFT data stream from a radio server. In one embodiment, if the trigger conditions are met, the alarms are triggered and stored in a local database for display.

FIG. 21 illustrates the components of a dynamic spectrum utilization and sharing model according to one embodiment of the present invention. By employing the dynamic spectrum utilization and sharing model, the present invention is operative to realize a plurality of radio frequency (RF) environment recognition functionalities including, but not limited to, monitoring and/or detection, identification and/or or classification. Monitoring and/or sensing functionalities include, but are not limited to, broadband frequency range sensing, real-time or near-real-time broadband capture, initial processing and/or post-event processing, autonomous monitoring 24 hours and/or reconfiguration options related to time, frequency and spatial configuration. Identification functionalities include, but are not limited to, detecting anomalous signals, signaling anomalous signals, recording timestamps of anomalous signals, providing a database of anomalous signals, and/or utilizing a spectrum mask. In one embodiment, the spectrum mask is a dynamic spectrum mask. Classification functionalities include, but are not limited to, correlation of signal events with known signaling protocols, correlation of signal events with known variables, correlation of signal events with known databases, correlation of events of signals with existing wireless signal formats and/or the correlation of signal events with existing cellular protocol formats. Each of the mentioned functionalities incorporates learning processes and/or procedures. These include, but are not limited to, historical data analysis, data curation tools, and/or learning analytics. Incorporating machine learning (ML), artificial intelligence (AI) and/or neural networks (NN) ensures that all aspects of detection, monitoring, identification and/or classification are performed autonomously. This is complemented by the use of learning analytics, enabling the use of utilization masks for continuous ML, predictive modeling, location analysis, intermodulation analysis and/or the integration of third-party data sets to increase capabilities. and/or general learning capabilities of the platform. Additionally, these capabilities and/or functionalities are supported through secure data retention services, which provides a secure platform environment and/or data implementation documentation (i.e. legal documents). Additionally, the platform is operational to provide push notifications, programmable event triggers, customizable rules and/or policies, and tips and signals practices. Push notifications include, but are not limited to, alerts, alarms and/or reports. Advantageously, this functionality allows the platform to react to specific regulations and/or policies, in addition to incorporating the platform's own recognition and knowledge, creating a platform optimized for any RF environment and/or mission.

The prediction models used by the platform provide an accurate view of the dynamic spectrum allocation and utilization functionalities. These prediction models allow the platform to autonomously create forecasts for future spectrum usage. Additionally, the prediction models used by the platform incorporate descriptive analytics, diagnostic analytics, predictive analytics and/or prescriptive analytics. Descriptive analysis refers specifically to the data stored, analyzed and/or used by the platform. Descriptive analysis provides data that allows the platform to act and/or provide a suggested action. Diagnostic analysis refers to how and/or why descriptive analysis acted and/or suggested an action. Predictive analytics specifically refers to the use of techniques including, but not limited to, ML, AI, NN, historical data and/or data mining to make predictions and/or future models. Prescriptive analysis refers to the act and/or suggested act generated by the descriptive analysis. Once this predictive model is implemented, the platform is operational to recommend and/or perform actions based on historical data, external data sources, ML, ΙΑ, NN and/or other learning techniques.

FIG. 22 illustrates a results model according to one embodiment of the present invention. The results model provided by the present invention focuses on four main practices: proactive, predictive, preventive and conservation. Predictive practice refers to the use of the functionalities and learning capabilities mentioned above to evolve the platform, allowing the characterization of events that led to an interference scenario and/or modeling of interference sources to forecast future probabilities and/or or conflictive events. Predictive practice is intertwined with the platform remaining proactive, identifying potential signals of interference. While the identification of potential interference signals is a combination of the platform's predictive and proactive capabilities, the platform also remains proactive in the form of wireless location characterization for pre- and post-event scenarios. Additionally, the platform's proactive capabilities include, but are not limited to, identifying all potential sources of conflict based on previous events. Furthermore, the platform also focuses on preventive practices. These include, but are not limited to, maintaining a set of conflict resolution rules, providing trigger warning notifications and/or early warning notifications, and/or maintaining compatibility with multiple government agencies, including government management offices. (PMO) and any sources of interference. In one embodiment, the platform automatically establishes the deconfliction rule set, where the deconfliction rule set is operational for editing. In one embodiment, the platform is operative to autonomously edit the set of deconfliction rules. In another modality, the platform allows the editing of the set of conflict resolution rules through user input. Finally, the platform includes conservation components and/or functionalities. These include, but are not limited to, test storage, learning capabilities, and modeling functionality. Each of these four main practices is interconnected within the platform, allowing spectrum to be used and shared dynamically.

Geolocation

Geolocation is an additional aspect related to the electromagnetic (e.g., RF) analysis of an environment. The main functions of electromagnetic environment analysis include, but are not limited to, detection, classification, identification, learning and/or geolocation.

Additionally, electromagnetic analysis is operational to generate environment recognition data.

The system includes a geolocation engine, operational to use passive and/or active radio geolocation methods. In general, radio geolocation refers to the geographical location of artificial emission sources that are propagated by radio (electromagnetic) waves when they hit an artificial geolocator or receiver. Passive radio geolocation does not require the transmission of signals by a geolocator, while active radio geolocation involves a geolocator transmitting signals that interact with a broadcast source. Passive geolocation methods include, but are not limited to, unidirectional beam antenna response, multidirectional beam antenna response (amplitude ratio), multi-antenna element response (array processing), LOB-to-position solutions and/or or general optimization. Multi-antenna element response methods include, but are not limited to, phased interferometry, beamforming, conventional multi-array processing approaches, and/or high-resolution multi-array processing approaches using signal subspace. While these passive methods are primarily applied to approaches for direction search (DF) such as spatial filters, passive methods that are applied to approaches other than DF such as spatial filters are operational for use by the system. DF refers to the process of estimating the direction of arrival of signals from the sender that propagate as they impinge on a receiver. Passive methods include other DF approaches based on general optimization including, but not limited to, digital predistortion (DPD), convex programming and/or distributed swarm approaches.

In addition to the passive approaches mentioned above, the system is operational to apply approaches based on range observations including, but not limited to, receiver signal strength indicators (RSSI), time of arrival (TOA) and/or or time difference of arrival (TDOA) methods. RSSI approaches relate to the generation of observable data and/or location estimation. TOA and/or TDOA approaches relate to the generation of observable data from distributed multi-antenna systems and/or single-antenna systems, and/or location estimation using non-linear optimization and/or constrained linear optimization.

In a preferred embodiment, geolocation is performed using measurements of angle of arrival (AOA), time difference of arrival (TDOA), frequency difference of arrival (FDOA) and power distribution ratio.

FIG. 23 is a table listing the problems that are operational to be solved using the present invention, including serviceability, interference, monitoring and prediction, anomalous detection, planning, compliance and/or spectrum sharing or leasing.

ινΐΛ/a/zuzz/uio/oo

FIG. 24 illustrates a view of a passive geolocation radio engine system according to an embodiment of the present invention. First, a radio frequency (RF) front end receives at least one RF signal. The RF front-end includes, but is not limited to, a sensor array, a sensor subsystem, at least one analog-to-digital converter (ADC), and/or an ADC sensor processing subsystem. Once the RF signal(s) have been analyzed by the RF front end and/or the sensor subsystem, the RF signal(s) are converted into at least one analyzed RF signal. The analyzed RF signal or signals are sent to a measurement subsystem. The measurement subsystem is operational to generate radio location measurements. Radio location measurements are based on envelope signals and/or feature-based signals. The measurement subsystem is further operative to generate contextual measurements and/or conventional measurements related to TOA, AOA, TDOA, receiver signal strength (RSS), RSSI and/or FDOA. The generated conventional measurements are then analyzed using position algorithms, further improving measurement accuracy. Once contextual measurements are generated and/or conventional measurements are analyzed using position algorithms, the analyzed RF signal(s) are sent to a position engine subsystem. The position engine subsystem includes a position display. Each of the components, systems and/or subsystems mentioned above are operational for network communication.

The geolocation engine is operative to use a plurality of algorithms to determine a location of the signal(s). The plurality of algorithms include, but are not limited to, TDOA, FDOA, AOA, power level measurements and/or graphical geolocation, which is described below. Geolocation is operational to autonomously decide which algorithm(s) to use to determine location.

FIG. 25 illustrates one embodiment of a method for autonomously selecting one or more of the plurality of algorithms. Carrier frequency offset and timing corrections are made to the I,Q data and sent to the signal detection engine. The I,Q data (e.g., I,Qo, I,Qi, I,Qz, I,Qa) is sent to the signal detection engine. The information from the signal detection engine is sent to the blind classification engine. The information from the blind sorting engine is sent to the demodulation bank. Error estimates are made on envelope (Doppler) measurements from the signal detection engine, signal (time) domain measurements from the blind sorting engine, and timing, protocol, and Doppler measurements from the demodulation bench. A fidelity evaluation is approximately equal to an SNR of the envelope measurements (Ai), signal measurements (Ας), and protocol measurements (Aa). Error analysis for AOA, TDOA, Correlation Ambiguity Function (CAF) for graphical geolocation, FDOA, and power ratio are used in fidelity evaluation. Ct is calculated and minimized over all methods to select the geolocation method(s), where Ct is the cost function to be minimized and t denotes a block of time used to calculate the geolocation solution.

In one embodiment, the geolocation engine uses graphical geolocation techniques. An area is represented pictorially on a grid. The resolution of the grid determines a position in space. The system is operational to detect the signal or signals in space and determine a location of the signal or signals using graphical geolocation techniques. In one embodiment, the outputs (e.g. e.g., location) of a nonlinear equation are used to determine the possible inputs (e.g., power measurements). The possible exits are placed on a two-dimensional map. The inputs are then mapped to form a hypothesis of possible outputs. In one embodiment, graphical geolocation techniques include an image comparison between the two-dimensional map of possible outputs and the signal data. In another embodiment, graphical geolocation techniques further include topology (e.g., mountains, valleys, buildings, etc.) to create a three-dimensional map of possible exits. Graphical geolocation techniques in this modality include an image comparison between the three-dimensional map of possible outputs and the signal data.

The geolocation engine is operational to make use of DF rotation, by using rotating directional antennas and estimating the direction of arrival of an emitter. Rotating directional antennas measure received power as a function of direction, calculating an assumed local maximum direction of the emitter. The geolocation engine is also operational to account for any transient signals that escape detection based on rotation speed. This is achieved by using at least one wide antenna, which reduces the chance of the system losing a signal, as well as reducing angular resolution. Practical considerations for these calculations include, but are not limited to, the antenna rotation speed (co), the signal arrival rate (y), and/or the spatial sampling rate (F_P.S.).

The system is furthermore operational to use amplitude ratio methods for geolocation. These methods involve a multi-lobe amplitude comparison. This is done using a set of fixed directional antennas pointing in different directions. A ratio corresponding to two responses is calculated, taking into account the antenna patterns. This relationship is used to obtain a direction estimate. By not using moving parts and/or antennas, the system responds better to transient signals. However, this requires exact antenna patterns, as these patterns also control the resolution of the system.

General antenna array processing assumes that a signal, s(t), remains coherent when incident on each antenna in the array. This allows the delay (r_m ) of the signal at an mth sensor relative to the signal at the origin of the coordinate system can be expressed as:

T_m = - (q_msin(0) + r_mcos(0))/c

Where c is the light propagation and Θ is the angle of the signal incident on the sensor in relation to the r axis. Since the signal is assumed to have a Taylor series decomposition, the propagation delay, r_m , is equivalent to the phase change of:

Vm= -WT_m = >

Therefore, the vector x(t) of the antenna responses can be written as:

•^1(0 *«(0_andj(wt+ φ;

Where &lt;Pm(w,e) = [q_msin(0) + r_mcos(e)]w/c

More generally, the sensor has different directionality and frequency characteristics which are modeled by applying different gains and phases to the above model, where the gain and phase of the mth sensor are denoted as: g_m(w, 0) and (p_m(w, Θ)

Then, the above equation for x(t) can be expressed as:

^(0' g_M(w,6)e^J.'^^w^ _and¡(wt+ φ) = a(w, Θ) e^Áwt+®

Where a{w, Θ) is known as the response vector of the array.

The collection of all array response vectors for all angles Θ and all frequencies, w, is known as an array manifold (i.e., a vector space). In general, if the manifold arrangement is known and unambiguous, then obtaining the angles k-1 (0_x ...0/ci) of the k-1 signals if their corresponding array response vector is linearly independent is done by correlating x(t) with the array response vector of the appropriate angle. In one embodiment, ambiguities refer to the multiple arrangement that lacks range deficiencies for k if the system is trying to resolve addresses k-1 at the same frequency. The array manifold does not usually have a simple analytical form and therefore the array manifold is approximated using discrete angles for each frequency of interest.

In more general cases, where multiple sinusoidal signals arrive at the array with additive noise, then x(t) can be expressed as:

x(O = a(w, ejSijt) + n(t) i = ls^t) = = [a(w,0i) ···a(w,0_z)][s^t) ··· $₇(ί)]^Γ +n(t) = A(w, 9)s(t) + n(t)

In one embodiment, additive noise refers to thermal noise from sensors and associated electronics, background noise from the environment and/or other man-made interference sources including, but not limited to, diffuse signals.

When one or more signals are non-sinusoidal (i.e. broadband), the equivalent can be expressed by their Taylor series over the relevant frequencies. However, when searching for a narrow frequency band of interest, it is operative to assume that an array response vector, a(w,e), is approximately constant with respect to w at all angles, Θ. This implies that the reciprocal of the time required for the signal to propagate through the array is much less than the bandwidth of the signal. If the sensor characteristics do not vary significantly across the bandwidth, then the dependence on w can be removed from the array and/or array response vector, resulting in:

x(t) = A(0)s(t) + n(t) ινΐΛ/a/zuzz/ui or/oo

For example, in an antenna array using a uniform linear array (ULA), a signal source, s(t) = incident on the ULA at an angle Θ. Therefore, if the signal received at a first sensor is XjCt) = s(t), then it is delayed at sensor m by:

. /(ml)dsin (0)\

- ^(t)

In vector form, this is represented as:

s(t) = a(w,0)s(t) (Μ —1) d sin(0)^land

If there are source signals received by the ULA, then:

x(t) = A (Q)s(t) + n(t)

Where

Α(Θ) =

<img file="MX2022013786A_D0001.tif" />

<img file="MX2022013786A_D0002.tif" />

'(Μ —1) d sin (θχ)' _j_w((Ml)dsin (8;) x(t) is the received signal vector (M times l),$(t) = [s^t) ··· s_;(t)]^r is the source signal vector (I times l), n(t) is the noise signal vector (M times 1), and Λ(Θ) = [αΟ,βυ, ···, a(w ,0J] an array (M by I) -> multiple array. In this example, typical assumptions include, but are not limited to, the signal source(s) are independent and narrow band relative to the dimensions of the ULA (d, Md) and are located around the same maximum frequency , all antenna elements are equal, d < to avoid range ambiguities, the system can resolve direction angles Ml without range and/or noises are uncorrelated.

In another example, array processing is performed for DF using beamforming. Given the knowledge of the multiple array, the array can be maneuvered by taking linear combinations of the response of each element. This is similar to how a single fixed antenna can be mechanically maneuvered. Therefore, y(t) = w^hx(t), where w is interpreted as a finite impulse response (FIR) of a filter in the spatial domain. To calculate the power of y(t), assuming a discretization to N samples, the system uses the following:

Py = <lyfa)|²>w = w^h{x(n) x(n)^h}Nw = w^hR_xxw

Where (. )_N denotes the time average over sample times N and R_xx is the array spatial autocorrelation measure of the received array output data.

In another example, array processing is performed for DF using beamforming, where R_xx — {x(n)x^h(n))ny Rxx — ((Α(Θ)χ(τι) + n(n))(4(0)s(n) + n(n')')^h)N

In one embodiment, the system assumes that a source signal is uncorrelated with a noise source, resulting in:

R_xx = A(&amp;)R_H.HTO^h (Θ) + R_nn

Therefore, the power of the linear combination and/or spatial filtering of the response elements of the array vector is expressed as:

Py = w^h(A(Q)RssA^h(G) + Rnn)w

In examples where array processing for DF is performed by beamforming, for a sinusoidal magnitude of a single unit incident on the array at angle θ₀ without noise it becomes:

PyW = w^ha(0o)a^h (eo)w = |w^ha(fl„)|²

Taking into account the Cauchy-Schwarz inequality |w^ha(e0)|² &lt; ||w||² ||α(Θ0)||², for all vectors n/with equality if, and only if, ives proportional to α(θ₀), the spatial filter that matches the array response in the direction of arrival, θ_ο, produces a maximum value for Ρ_γ(θ).

Furthermore, a DF can be achieved by searching all possible angles to maximize Pyíd), and/° by searching all filters ivthat are proportional to some array vectors that respond to an angle of incidence θ, αίθ), where Max (71 (0)] ,,,. When^J. over alL angles =>filters w=a(0) If this method is used, the system behaves like a rotating DF system where the resulting beam changes for each search angle. Advantageously, this method does not find blind spots due to the rotation and/or arrival speed of the source signal.

Additionally, when the system uses beamforming techniques and/or processes, the system is operational to search for multiple directions of arrival from different sources with resolutions depending on the width of the formed beam and the height of the sidelobes. For example, a local maximum of the average filter output power is operative to be diverted from the true direction of arrival (DOA) of a weak signal by a strong interference source in the vicinity of one of the sidelobes. Alternatively, two closely spaced signals result in only one spike or two spikes in the wrong location.

In yet another example, array processing for DF is performed using a Capon Minimum Variance Undistorted Response (MVDR) approach. This is necessary in cases where signals from multiple sources are present. The system obtains more accurate estimates of DOA by formatting the array beam using degrees of freedom to form a beam in the looking direction and any remaining degrees of freedom to make it null in the remaining directions. The result is a simultaneous beam and zero shaping filter. Null formation in other directions is achieved by minimizing Ρ_γ(θ) while constraining a beam in the looking direction. This avoids the trivial solution of w- 0. Thus:

^m!ⁿover allw SUbjeCt tO W^h α(θ) = 1

The resulting filter, w_c(f)), is shown as:

w_c(0) =

Using this filter, the filter output power is expressed as:

PycW = w^0)R_xxw_cW =

Therefore, Capon's approach seeks at all DOA angles for the above power to be maximized, using max_{over alllosángu}Yo_you A Capon approach is capable of discerning multiple signal sources because while observing signals incident at 0, the system attenuates a signal reaching fifteen degrees by a formed beam.

A Capon approach is a method for estimating an angular decomposition of the average power received by the array, sometimes called the spatial spectrum of the array. The Capon approach is a similar approach to estimating and/or modeling the spectrum of a linear system.

The system is also operable to employ additional resolution techniques including, but not limited to, multiple signal classifier (MUSIC), signal parameter estimation via rotational invariance techniques (ESPRITE) and/or any other DOA algorithm. high resolution. These resolution techniques allow the system to find DOA for multiple sources simultaneously. Additionally, these resolution techniques generate high spatial resolution compared to more traditional methods. In one embodiment, these techniques are applied only when determining DOAs for narrowband signal sources.

For example, when using MUSIC-based methods, the system computes an N x N correlation array using R_x = E{x(t)x^h(t)) = AR_YesTO^h + σ$1, where

R_yes = E{s(t)s^h (ty} = diag.{a?,...,a?}. If the signal sources are correlated so that Rs is not diagonal, geolocation will continue to work as long as Rs is in full range. However, if the signal sources are correlated such that Rs is range deficient, the system will deploy spatial smoothing. This is important, since Rs defines the subspace dimension of the signal. However, for N > I, the ARSA array^h is singular, where detE/l/?^] = det[Rx - σ₀²/] = 0. But this implies that σ0² is an eigenvalue of Rx. Since the dimension of the null space AR_YesTO^h is N - I, there are N - /eigenvalues ​​σ0² of Rx. Furthermore, since both R_x like AR_YesTO^h are non-negative, are there /eigenvalues ​​a? such that σ/ > σ₀² > 0.

In a preferred embodiment, geolocation is performed using measurements of angle of arrival (AOA), time difference of arrival (TDOA), frequency difference of arrival (FDOA) and power distribution ratio. Advantageously, using all four measurements to determine geolocation results in a more accurate location determination. In many cases, only one type of geolocation measurement is available which forces the use of a particular approach (e.g. e.g., AOA, TDOA, FDOA), but in many cases geolocation measurements are operational to be derived from signal behavior, allowing the use of multiple measurements (e.g., all four measurements) that are combined to obtain a more robust geolocation solution. This is especially important when most of the measurements associated with each approach are extremely noisy.

Learning engine

Additionally, the system includes a learning engine, operational to incorporate a plurality of learning techniques including, but not limited to, machine learning (ML), artificial intelligence (AI), deep learning (DL), neural networks (NN). ), artificial neural networks (ANN), support vector machines (SVM), Markov decision process (MDP) and/or natural language processing (NLP). The system is operational to use any of the aforementioned learning techniques alone or in combination.

Advantageously, the system is operative for autonomous operation using the learning engine. Additionally, the system is operational to be continually refined, resulting in greater accuracy with respect to data collection, analysis, modeling, prediction, measurements and/or output.

The learning engine is further operative to analyze and/or calculate a conditional probability set. The conditional probability set reflects the optimal outcome for a specific scenario, and the specific scenario is represented by a data model used by the learning engine. This allows the system, when given a set of ινΐΛ/a/zuzz/ui or/oo data inputs, to predict an outcome using a data model, where the predicted outcome represents the outcome with the lowest probability of error and /or a false alarm.

Without a learning engine, prior art systems are still operational to create parametric models to predict various outcomes. However, these prior art systems cannot capture all inputs and/or outputs, thereby creating inaccurate data models related to a specific set of input data. This results in a system that continually produces the same results when given completely different sets of data. In contrast, the present invention uses a learning engine with a variety of fast and/or efficient computational methods that simultaneously calculate the conditional probabilities that are most directly related to the outcomes predicted by the system. These computational methods are performed in real time or near real time.

Additionally, the system employs control theory concepts and methods within the learning engine. This allows the system to determine whether each data set processed and/or analyzed by the system represents a sufficient statistical data set.

Additionally, the Learning Engine includes a Learning Engine Software Development Kit (SDK), which allows the system to prepare and/or manage the lifecycle of data sets used in any learning application on the system. Advantageously, the learning engine SDK operates to manage system resources related to monitoring, recording and/or organizing any learning aspect of the system. This allows the system to train and/or run models locally and/or remotely using automated ML, AI, DL and/or NN. Models are operational for configuration, where the system is operational for modifying model configuration parameters and/or training data sets. By operating autonomously, the system is operationalized through algorithms and/or hyperparameter settings, creating the most accurate and/or efficient model to run predictive system applications. Additionally, the Learning Engine SDK can be operated to implement web services to convert any training model into services that can run in any application and/or environment.

Therefore, the system is operational to function autonomously and/or continuously, refining each predictive aspect of the system as the system acquires more data. Although this functionality is controlled by the learning engine, the system is not limited to employing these learning techniques and/or methods only in the learning engine component, but throughout the entire system. This includes RF fingerprinting, RF spectrum reconnaissance, modification of the autonomous RF system configuration, and/or autonomous system operations and maintenance.

The learning engine uses a combination of physical models and convolutional neural network algorithms to calculate a set of possible conditional probabilities that represent the set of all possible outcomes based on the input measurements that provide the most accurate prediction of the solution, where exact means minimizing the false probability of the solution and also the probability of error for the prediction of the solution.

FIG. 26 is a diagram that describes three pillars of a customer mission solution. The three pillars include environmental awareness, policy management and spectrum management. The system obtains recognition of the environment through a plurality of sensors. The plurality of sensors preferably captures real-time information about the electromagnetic environment. Additionally, the system includes machine learning and/or predictive algorithms to improve environmental awareness and support resource scheduling. Policy management is flexible, adaptive and dynamic, preferably taking into account real-time information about device configurations and the electromagnetic environment. The system is preferably operational to manage heterogeneous networks of devices and applications. Spectrum management preferably makes use of advanced device capabilities including, but not limited to, directionality, waveforms, hopping and/or aggregation.

FIG. TI is a block diagram of an example of a spectrum management tool. The spectrum management tool includes environmental information obtained from at least one monitoring sensor and at least one sensor processor. The spectrum management tool further includes a policy manager, reasoner, optimizer, targets, device information, and/or device manager. Objectives include information from a mission information database. The policy manager obtains information from a policy information database. In another embodiment, the policy manager uses information (e.g., from the policy information database, measurements of the electromagnetic environment) to create policies and/or rules for the conditional allocation of resources per signal using the spectrum. These policies and/or rules are then passed to the reasoner to determine the conditional optimization constraints to be used by the optimizer with the goal of optimizing spectrum utilization (e.g., based on mission information and objectives). for all signs present in accordance with policies and/or standards. At the output of the optimizer, resources (bandwidth, power, frequency, modulation, spatial azimuth, and elevation focus for TX/RX sources) as well as interference levels per application are recommended for each signal source. After that, the environment collection and awareness loop is passed to the policy manager and reasoner.

FIG. 28 is a block diagram of one embodiment of a resource brokering application. As described above, the resource brokering application is ινΐΛ/a/zuzz/ui or/oo preferably operational to use processed data from the monitoring sensor(s) and/or additional information to determine environment awareness (e.g. , recognition of the surrounding situation). Awareness of the environment and/or capabilities of a device and/or resource are used to determine policies and/or reasoning to optimize the device and/or resource. The resource brokering application is operative to control the device and/or the resource. Additionally, the resource brokering application is operative to control the monitoring sensor(s).

Semantic engine

The system also includes an automated semantic engine and/or translator, as shown in FIG. 29. The translator is operative to receive data inputs that include, but are not limited to, at least one use case, at least one objective, and/or at least one signal. In one embodiment, the use case(s) is a single signal use case. In another embodiment, the use case or cases is a multiple signal use case. Once the translator receives the data input, the translator uses natural language processing (NLP), and/or similar data translation processes and techniques, to convert the data input into actionable data for the automated semantic engine.

By separating the data translation process from the automated semantic engine, the system is operational to provide more processing power once the data input is sent to the automated semantic engine, reducing the overall processing load on the system.

The automated semantic engine includes a rule component, a syntax component, a logic component, a quadrature (Q) component, and/or a conditional set component. Furthermore, the semantic engine is operational for network communication with a prior knowledge database, an analysis engine, and/or a tracking and capture engine. The data is initially sent to the automated semantic engine through the translator. The automated semantic engine is operative to receive data from the translator in forms including, but not limited to, audio data, text data, video data, and/or image data. In one embodiment, the automated semantic engine is operative to receive a query from the translator. The logical component and/or the rules component are operative to establish a set of system rules and/or a set of system policies, where the set of system rules and/or the set of system policies is created using the prior knowledge database.

Advantageously, the automated semantic engine is operative to function autonomously using any of the learning and/or automation techniques mentioned above. This allows the system to run continuously, without the need for user interaction or input, resulting in a system that is constantly learning and/or refining data inputs, creating predictions, models and/or suggested actions further. exact.

Additionally, the automated semantic engine allows the system to receive queries, searches, and/or any other type of search-related function using natural language, rather than requiring a user and/or client to adapt to a particular computer language. This functionality is performed using semantic search using natural language processing (NLP). Semantic search combines traditional word searches with logical relationships and concepts.

In one embodiment , the automated semantic engine uses Latent Semantic Indexing (LSI) within the automated semantic engine. The LSI organizes existing information within the system into structures that support high-order associations of words with text objects. These structures reflect the associative patterns found within the data, enabling data retrieval based on the semantic context latent in the existing system data. Additionally, LSI is operational to account for noise associated with any input data set. This is done through LIS's ability to increase retrieval functionality, a restriction of traditional Boolean queries and vector space models. LSI uses automated categorization, mapping a set of input data to one or more predefined data categories contained within the prior knowledge database, where the categories are based on a conceptual similarity between the input data set and the content of the prior knowledge database. Furthermore, LSI makes use of dynamic clustering, grouping the input data set into data within the prior knowledge database using conceptual similarity without using example data to establish a conceptual basis for each group.

In another embodiment, the automated semantic engine uses Latent Semantic Analysis (LSA) within the automated semantic engine. LSA functionalities include, but are not limited to, occurrence array creation, classification, and/or derivation. Creating the occurrence array involves using a term-document array that describes the occurrences of terms in a data set. Once the occurrence array is created, the LSA uses classification to determine the most accurate solution given the data set. In one embodiment, a low-rank approximation is used to classify the data within the occurrence array.

In another embodiment, the automated semantic engine uses semantic fingerprints. Semantic fingerprinting converts a set of input data into a Boolean vector and creates a semantic map using the Boolean vector. The semantic map can be used in any context and provides an indication of each data match for the input data set. This allows the automated semantic engine to convert any set of input data ινΐΛ/a/zuzz/ui or/oo into a semantic fingerprint, where the semantic fingerprints are operational to be combined with additional semantic fingerprints, providing an exact solution given the set of input data. The functionality of semantic fingerprints also includes, but is not limited to, risk analysis, document search, classification indication and/or classification.

In yet another embodiment, the automated semantic engine uses semantic hashing. Using semantic hashing, the automated semantic engine maps a set of input data to memory addresses using a neural network, where sets of semantically similar data inputs are located at nearby addresses. The automated semantic engine is operational to create a graphical representation of the semantic hashing process using counting vectors from each set of data inputs. Therefore, sets of data entries similar to a target query can be found by accessing all memory addresses that differ by only a few bits from the address of the target query. This method extends the efficiency of hashing to approximate matching much faster than locality-sensitive hashing.

In one embodiment, the automated semantic engine is operative to create a semantic map. The semantic map is used to create target data in the center of the semantic map, while analyzing related data and/or data with similar characteristics to the target data. This adds a secondary layer of analysis to the automated semantic engine, providing secondary context for the target data using similar and/or alternative solutions based on the target data. The system is operational to create a semantic map visualization.

Search systems based on traditional semantic networks suffer from numerous performance problems due to the scale of an expansive semantic network. For semantic functionality to be useful in locating exact results, a system that stores a large volume of data is required. Furthermore, such a wide network creates difficulties in processing many possible solutions to a given problem. The system of the present invention resolves these limitations through the various learning techniques and/or processes incorporated within the system. When combined with the ability to operate autonomously, the system is operational to process a greater amount of data than systems that use only traditional semantic approaches.

By incorporating the automated semantic engine within the system, the system has a greater understanding of possible solutions, given a given set of data. Semantic engines are regularly associated with semantic searches or searches with meaning or searches with an understanding of the general meaning of the query, therefore, understanding the searcher's intent and the contextual meaning of the search generates more relevant results. The semantic engines of the present invention, together with a specific iviA/a/zuzz/ui or/oo spectrum ontology (vocabulary and operational domain knowledge), help automate spectrum utilization decisions based on dynamic observations and knowledge. extracted from the environment, and create and extend spectrum management awareness for multiple applications.

Suggestion and signal processes

The system uses a set of hinting and signaling processes, which generally relate to detecting, processing and/or providing alerts by creating actionable data from RF environment awareness information acquired along with a specific rule set, which further enhances the optimization capabilities of the system. The specific set of rules translates into optimization objectives, including constraints associated with signal characteristics. The suggestion and signaling processes of the present invention produce actionable data to solve a plurality of user problems and/or objectives.

The processes of suggestions and signals are carried out by a recognition system. The recognition system is operative to receive input data that includes, but is not limited to, a set of use cases, at least one objective and/or a set of rules. The input data is then parsed by a translator component, where the translator component normalizes the input data. Once normalized, the input data is sent to a semantic engine. The semantic engine is necessary to analyze unstructured data inputs. Therefore, a semantic engine is needed to understand the data inputs and also apply contextual analysis, leading to a more accurate output result. This accuracy is primarily achieved using the learning techniques and/or technologies mentioned above.

The semantic engine uses the input data to create a set of updated rules, a syntax, a logic component, a set of conditional data, and/or quadrature (Q) data. The semantic engine functions for network communication with components including, but not limited to, a prior knowledge database, an analysis engine, and/or a monitoring and capture engine. The monitoring and capture engine operates with an RF environment and includes a client application programming interface (API), a radio server, and/or a coverage management component. The radio client and server API can function to generate a set of I-phase and Q-phase (I/Q) data using a fast Fourier transform (FFT). The I/Q data set demonstrates the amplitude and phase changes in a sine wave. The monitor and capture engine also serve as an optimization point for the system.

The recognition engine functions as a platform optimization unit and as a client optimization unit. The recognition engine may function to perform functions including, but not limited to, detection, classification, demodulation, ινΐΛ/a/zuzz/ui or/or decoding, localization and/or alarm signaling. The detection and/or classification functions assist with the acclimation of the incoming RF data and further include a supervised learning component, where the supervised learning component is operational to make use of any of the learning techniques and/or technologies before mentioned. The demodulation and/or decoding functionalities are operational to access RF data from WIFI networks, land mobile radio (LMR) networks, long term evolution (LTE) networks and/or unmanned aircraft systems (UAS). The location component of the recognition engine can be operated to apply location techniques including, but not limited to, DF, geolocation, and/or Internet Protocol (IP) based location. The recognition engine is operational to indicate alarms using FASD and/or masks. In one embodiment, the masks are dynamic masks.

The analytics engine is operational to perform functions including, but not limited to, data qualification, data transformation, and/or data computation.

The recognition engine, analysis engine and semantic engine are operational for network communication with the prior knowledge database. This allows each of the engines mentioned above to compare the input and/or output with the data already processed and analyzed by the system.

The various engines present in the suggestion and signaling process further optimize the client's output in the form of dynamic spectrum utilization and/or allocation. The system uses the suggestion and signal process to provide actionable information and/or actionable knowledge to be used by at least one application to mitigate problems of the application or applications and/or to optimize services or objectives of the application or applications.

In a preferred embodiment, each customer has a service level agreement (SLA) with the system administrator that specifies spectrum usage. The system administrator can act as an intermediary between a first client and a second client in spectrum-related conflicts. If the first client's signals interfere with the second client's signals in violation of one or more of the SLAs, the system is operational to provide an alert of the violation. Data related to the violation is stored in at least one database within the system, making resolution of the violation easier. The control plane is operational to communicate directly with the first client (i.e., the client violating the SLA) and/or at least one base station to modify parameters to resolve the violation.

In one embodiment, the system is used to protect at least one critical asset. Each of the critical asset or assets is located within a protection area. For example, a first critical asset is within a first protection area, a second critical asset is within a second protection area, etc. In one embodiment, the protection area is defined by the sensor coverage of the at least one monitoring sensor. In other embodiments, the protection area is defined by sensor coverage from the monitoring sensor(s), a geofence, and/or GPS coordinates. The system is operational to detect at least one signal within the protection area and send an alarm for the signal or signals when it is outside the permitted use of the spectrum within the protection area.

The system is further operational to determine what information is necessary to provide actionable information. For example, sensor processing requires a large amount of power. Incorporating only the sensors necessary to provide sufficient variables for the client's objectives reduces computing and/or power requirements.

FIGS. 30-32 are flowcharts that illustrate the process of obtaining actionable data and using knowledge decision gates. FIG. 30 illustrates a flowchart of a method for obtaining actionable data based on customer objectives 3000. An objective is rephrased as a question in Step 3002. The information required to answer the question is identified in Step 3004. Next, quality, quantity, temporal and/or spatial attributes are identified for each piece of information in Step 3006. In a preferred embodiment, the four attributes (i.e., quality, quantity, temporal and spatial) are identified in step 3006. Quality, quantity, temporal and/or spatial attributes are classified by importance in Step 3008. For each information and attribute pair, information from the corresponding physical layer of the wireless environment is associated in Step 3010. All information obtained in steps 3004-3010 is operational to be transmitted to the semantic engine.

Additionally, the wireless information is associated with a more statistically relevant combination of measurements extracted in at least one dimension in Step 3012. The dimension(s) include, but are not limited to, time, frequency, signal space and/or characteristics. signal, spatial and/or application objectives and/or impact on the client. In a preferred embodiment, the dimension(s) include time, frequency, signal space and/or signal characteristics, spatial and application objectives and/or customer impact. RF recognition measurements are then qualified in Step 3014 and actionable data is provided in Step 3016 based on the relationship established in Steps 3002-3012. The efficiency of actionable data is qualified in Step 3018 based on Step 3014. All actionable data and its statistical significance are provided in Step 3020.

FIG. 31 illustrates a flowchart of a method of implementing knowledge decision gates and actionable data from the total signal flow 3100. The client objective is rephrased as a question in Step 3102. The client objective is provided to the semantic engine having a suitable dictionary in Step 3104 (as shown in Steps 3002-3012 of FIG. 30). Statistically relevant constraints in Step 3104 and extracted electromagnetic reconnaissance information (e.g., RF) from sensors in Step ινΐΛ/a/zuzz/ui or/oo

3106 are used in an optimization cost function in Step 3108 (as shown in Step 3014 of FIG. 30). The results of the optimization cost function in Step 3108 are provided to an optimization engine in Step 3110 (as shown in Steps 30163020 of FIG. 30) to provide processable data and its statistical relevance in Step 3112. .

FIG. 32 illustrates a flowchart of a method for identifying knowledge decision gates based on operational knowledge 3200. The client's operational description of the utilization of actionable data is provided in Step 3202. The customer's operational description of the use of actionable data from Step 3202 is used to identify a common state of other information used to express the customer's operational description and/or required to make decisions in Step 3204. In addition, the operational description of the customer client operational utilization of the actionable data in Step 3202 is used to provide the parameterization of the client's operational utilization of the actionable data in Step 3206. The parameterization of the client's operational use of actionable data from Step 3206 is used to identify conditions and create a conditional tree in Step 3208. In one embodiment, the information from Step 3204 is used to identify the conditions and create the conditional tree in Step 3208. The information in steps 32063208 is operational to be transmitted to the semantic engine. The actionable data is provided in Step 3210 and is used to calculate the statistical properties of the actionable data as it changes over time in Step 3212. The information from Steps 3208 and 3212 is used by a decision engine to traverse a decision tree to identify the decision gates in Step 3214. The identified decision gates from Step 3214 are provided along with the information from Step 3204 to allow the client to make decisions in Step 3216.

FIG. 33 illustrates an overview of an example of information used to provide knowledge. Information including, but not limited to, network information (e.g., existing site locations, existing site configurations), real estate information (e.g., candidate site locations), signal data (e.g. (e.g. LTE demodulation), signal sites, site issues, crowdsourced information (e.g. e.g., geographic traffic distribution) and/or geographic information services (GIS) is used to perform propagation modeling. Propagation models are used to evaluate the possible outcomes and expected impact of any changes (e.g., addition of macro sites, tower). In one modality, an additional analysis is carried out on the possible results and/or the expected impact.

EXAMPLE ONE

In a first example, the system is used by a tower company to evaluate whether the performance of a carrier can be improved by placing at least one additional macro site on at least one additional tower. If the evaluation shows that the performance of the carrier ινΐΛ/a/zuzz/ui or/oo can be improved, it supports an argument by the tower company to place the macro site(s) on the additional tower(s), which would generate revenue for the tower company.

FIG. 34 is a map showing the locations of three macro sites (1 (green), 2 (orange), and 3 (purple)), 3 SigBASE units (orange diamond), and a plurality of locations evaluated for additional or alternative site deployment ( green circles).

FIG. 35 is a user distribution graph by average downlink Physical Resource Block (PRB) allocation. Real-time monitoring shows the downlink resources allocated to each user. Assignments occur many times per second. A significant concentration of users on 739 MHz have resources assigned to voice service. Most users on 2165 MHz have common resources allocated for high-speed data.

FIG. 36 illustrates the rate of overutilization events and the degree of overutilization. Real-time monitoring shows the percentage of downlink resources used when utilization exceeds 50%. Usage statistics are generated per second as configured. The rate at which a sector's utilization exceeds 50% (overutilized) is presented per hour. Average utilization levels when overutilization occurs describe the severity.

FIG. 37A is a sector coverage map for the three macrosites (1 (green), 2 (orange) and 3 (purple)).

FIG. 37B illustrates the signal intensity for the sector shown in FIG. 37A. This figure shows areas of poor coverage.

FIG. 37C illustrates the subscriber density for the sector shown in FIG. 37A. In one embodiment, data from external sources is used to determine the distribution and density of subscribers. This figure shows areas of high subscriber demand.

FIG. 37D illustrates the carrier/interference relationship for the sector shown in FIG. 37A. This figure shows areas of poor quality.

FIG. 38A illustrates the reference scenario shown in FIG. 34. FIG. 38B is a map showing the locations of the three original macrosites (1 (green), 2 (orange), and 3 (purple)) and two additional macrosites (4 (dark blue) and 5 (light blue)).

FIG. 39 illustrates the signal intensity of the reference scenario of FIG. 38A on the left and the stage with two additional macrosites of FIG. 38B on the right. Adding a 2-sector eNodeB to a tower increases expected coverage by 3 km², as shown in Table 2 below. A total service area for the baseline is 9.89 km² and the total service area increases to 13.15 km² with the two additional macro sites. A total area with a carrier-to-interference ratio less than 5 dB decreases by 1.10 km² for the baseline at 0.38 km² with the two additional macro sites. A total area with a carrier-to-interference ratio greater than 5 iviA/a/zuzz/ui or/oo dB increases by 8.79 km² for the baseline at 12.77 km² with the two additional macro sites. The traffic served without harmful interference increases from 16.73 Erlands for the baseline to 25.23 Erlands with the two additional macro sites. Additionally, an increase in traffic served of 40% is expected. A further 30% utilization reduction is expected for pre-existing sectors. Areas of poor coverage are also reduced.

TABLE 2

<td>Metrics</td><td>Baseline</td><td>2 sector site added</td>

<td>Total service area, km²</td><td> 9,89</td><td> 13,15</td>

<td>Total area with C/I &lt;5 dB, km²</td><td> 1,10</td><td> 0,38</td>

<td>Total area with C/I&gt;5 dB, km²</td><td> 8,79</td><td> 12,77</td>

<td>Traffic served without harmful interference, Erlands</td><td> 16,73</td><td> 25,23</td>

FIG. 40A illustrates the carrier/interference relationship for the reference scenario of FIG. 38A. FIG. 40B illustrates the carrier/interference scenario scenario with two additional macro sites. The two additional macro sites reduce areas with poor carrier/interference ratios.

EXAMPLE TWO

In a second example, a tower company also uses the system to evaluate whether the performance of a carrier can be improved by placing at least one additional macro site on at least one additional tower. If the evaluation shows that carrier performance can be improved, it supports an argument by the tower company to place the macro site(s) on the additional tower(s), which would generate revenue for the tower company.

FIG. 41 illustrates a baseline scenario for the second example on the left and a map showing the locations of the original macrosites from the reference scenario with three additional proposed macrosites on the right.

FIG. 42 illustrates the signal intensity of the reference scenario in FIG. 41 on the left and the stage with three additional proposed macrosites from FIG. 41 on the right. Adding a 3-sector eNodeB to a tower increases expected coverage by 0.5 km², as shown in Table 3 below. A total service area for the baseline is 21.3 km²and the total service area increases to 21.8 km² with the three additional macro sites. A total area with a carrier-to-interference ratio less than 5 dB decreases by 3.0 km² for the baseline at 3.1 km² with the three additional macro sites. A total area with a carrier-to-interference ratio greater than 5 dB increases from 18.3 km² for the baseline at 18.7 km² with the three additional macro sites. The traffic served without negative interference increases from 79.7 Erlands for the baseline to 80.9 Erlands with the three additional macro sites. Additionally, an increase in traffic served of 2% is expected. A further 2% utilization reduction is expected for pre-existing sectors.

iviA/a/zuzz/uij/oo

TABLE 3

<td>Metrics</td><td>Baseline</td><td>487044 added</td>

<td>Total service area, km²</td><td> 21,3</td><td> 21,8</td>

<td>Total area with C/I &lt;5 dB, km²</td><td> 3,0</td><td> 3,1</td>

<td>Total area with C/I&gt;5 dB, km²</td><td> 18,3</td><td> 18,7</td>

<td>Traffic served without harmful interference, Erlands</td><td> 79,7</td><td> 80,9</td>

FIG. 43A illustrates the carrier/interference relationship of the reference scenario of FIG. 41. FIG. 43B illustrates the carrier/interference scenario scenario with three additional proposed macrosites from FIG. 41 on the right. The three additional proposed macrosites slightly reduce the areas with poor carrier/interference ratio.

Although adding the 3-sector eNodeB slightly improves performance, this performance improvement is not significant enough to support adding the three proposed macro sites to the tower.

EXAMPLE THREE

In a third example, the system is used to evaluate which carrier provides better service.

FIG. 44 illustrates a signal strength comparison of a first carrier (Carrier 1) with a second carrier (Carrier 2) for 700MHz.

FIG. 45 illustrates the carrier/interference relationship for Carrier 1 and Carrier 2.

FIG. 46 is a graph of Area vs. RSSI and Traffic vs. RSSi for Carrier 1 and Carrier 2. Carrier 1 and Carrier 2 serve approximately the same amount of area in the sector.

FIG. 47 is a graph of the traffic difference for Carrier 1 versus Carrier 2. Carrier 2 serves more traffic than Carrier 1 at the extremes of coverage, while Carrier 1 serves more traffic in the middle range of coverage.

FIGS. 44-47 illustrate the traffic composition for each SigBASE. Different types of traffic require different signal-to-noise ratios (SNR) versus reference received signal power (RSRP). For voice traffic, the SNR is -6 dB to 0 dB, while the SNR exceeds 20 dB for video transmission.

FIG. 48 is a plot of SNR versus RSRP for each SigBASE for the third example.

FIG. 49 is another graph of SNR vs RSRP for each SigBASE for the third example.

FIG. 50 is a clustering plot of SNR versus RSRP for each SigBASE for the third example.

FIG. 51 is another clustering plot of SNR vs. RSRP for each SigBASE for the third example.

FIG. 52 is a schematic diagram of an embodiment of the invention illustrating a computing system, generally described as 800, having a network 810, a plurality of computing devices 820, 830, 840, a server 850, and a database 870 .

The server 850 is constructed, configured and coupled to enable communication over a network 810 with a plurality of computing devices 820, 830, 840. The server 850 includes a processing unit 851 with an operating system 852. The operating system 852 allows the server 850 to communicate over the network 810 with remote and distributed user devices. The database 870 is operative to host an operating system 872, memory 874, and programs 876.

In one embodiment of the invention, system 800 includes a network 810 for distributed communication via a wireless communication antenna 812 and processing via at least one mobile communication computing device 830. Alternatively, wireless and wired communication and connectivity between the devices and components described herein include wireless network communication such as WI-FI, Worldwide Interoperability for Microwave Access (WIMAX), radio frequency (RF) communication, including RF identification (RFID), NEAR FIELD COMMUNICATION (NFC), BLUETOOTH, including BLUETOOTH LOW ENERGY (BLE), ZIGBEE, infrared (IR) communication, cellular communication, satellite communication, universal serial bus (USB), Ethernet communications, communication via fiber optic cables, coaxial cables, twisted pair cables, and/or any other type of wireless or wired communication. In another embodiment of the invention, system 800 is a virtualized computing system capable of executing any or all aspects of the software and/or application components presented herein on computing devices 820, 830, 840. In certain aspects, the computing system 800 is operable for implementation using hardware or a combination of software and hardware, whether on a dedicated computing device, or integrated into another entity, or distributed among multiple computing entities or devices.

By way of example, and not as a limitation, the computing devices 820, 830, 840 are intended to represent various forms of electronic devices that include at least one processor and memory, such as a server, blade server, mainframe, mobile phone, assistant personal digital (PDA), smartphone, desktop computer, netbook computer, tablet, workstation, laptop and other similar computing devices. The components shown here, their connections and relationships, and their functions, are intended to be an example only and are not intended to limit the implementations of the invention described and/or claimed in the present application.

In one embodiment, the computing device 820 includes components such as a processor 860, a system memory 862 having a random access memory (RAM) 864 and a read-only memory (ROM) 866, and a system bus 868 that couples memory 862 to processor 860. In another embodiment, the computing device 830 is operative to additionally include components such as a storage device 890 for storing the operating system 892 and one or more application programs 894, a network interface unit 896 and/or a controller. input/output 898. Each of the components is operative to couple together through at least one bus 868. The input/output controller 898 is operative to receive and process inputs from, or provide outputs to, a variety of other devices 899, including, but not limited to, alphanumeric input devices, mice, electronic styluses, drives. displays, touch screens, signal generating devices (e.g. loudspeakers) or printers.

By way of example, and without limitation, processor 860 is operative to be a general purpose microprocessor (e.g. e.g., a central processing unit (CPU), a graphics processing unit (GPU), a microcontroller, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a programmable gate array (FPGA), a programmable logic device (PLD), a controller, a state machine, gate or transistor logic, discrete hardware components or any other entity or combinations thereof that can perform calculations, process instructions for execution and/or other manipulations of information.

In another implementation, shown as 840 in FIG. 52, multiple processors 860 and/or multiple buses 868 are operative to be used, as appropriate, together with multiple memories 862 of multiple types (e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in along with a DSP core).

Additionally, multiple computing devices can be connected, with each device providing parts of the necessary operations (e.g., a bank of servers, a cluster of blade servers, or a multiprocessor system). Alternatively, some steps or methods are operative to be performed by a circuit that is specific for a given function.

According to various embodiments, the computing system 800 may operate in a networked environment using logical connections to local and/or remote computing devices 820, 830,840 over a network 810. A computing device 830 may operate to connect to a network 810 through a network interface unit 896 connected to a bus 868. The computing devices are operative to communicate media over wired networks, direct wired connections or wirelessly, such as acoustic, RF or infrared, through an antenna 897 in communication with the network antenna 812 and the unit of network interface 896, which can function to include digital signal processing circuitry when necessary. The network interface unit 896 is operative to provide communications under various modes or protocols.

In one or more example aspects, the instructions may operate to be implemented in hardware, software, firmware, or any combination thereof. A computer-readable medium functions to provide volatile or non-volatile storage for one or more sets of instructions, such as operating systems, data structures, program modules, applications, or other data that incorporate one or more of the methodologies or functions described in this document. The computer-readable medium is operative to include the memory 862, the processor 860 and/or the storage media 890 and is operative on a single medium or on multiple media (e.g., a centralized or distributed computing system) that stores one or more instruction sets 900. Non-transitory computer-readable media includes all computer-readable media, with the sole exception of a transient propagation signal itself. The instructions 900 are also operative to be transmitted or received over the network 810 through the network interface unit 896 as a communication medium, which is operative to include a modulated data signal such as a carrier wave or other transmission mechanism. transportation and includes any means of delivery. The term modulated data signal means a signal that has one or more of its characteristics changed or configured so as to encode information in the signal.

Storage devices 890 and memory 862 include, but are not limited to, volatile and non-volatile media such as cache, RAM, ROM, EPROM, EEPROM, FLASH memory or other solid state memory technology; discs (p. e.g., digital versatile discs (DVD), HDDVD, BLU-RAY, compact discs (CD) or CD-ROM) or other optical storage; magnetic cassettes, magnetic tape, magnetic disk storage, floppy disks or other magnetic storage devices; or any other medium that can be used to store the computer-readable instructions and that can be accessed by the computer system 800.

In one embodiment, the computing system 800 is within a cloud-based network. In one embodiment, server 850 is a physical server designated for distributed computing devices 820, 830, and 840. In one embodiment, server 850 is a cloud-based server platform. In one embodiment, the cloud-based server platform hosts serverless functions for distributed computing devices 820, 830, and 840.

In another embodiment, the computing system 800 is within an edge computing network. Server 850 is an edge server and database 870 is an edge database. The edge server 850 and edge database 870 are part of an edge computing platform. In one embodiment, the edge server 850 and edge database 870 are designated for distributed computing devices 820, 830, and 840. In one embodiment, the edge server 850 and the edge database 870 are not designated for distributed computing devices 820, 830, and 840. The distributed computing devices 820, 830, and 840 connect to an edge server on the network. edge computing based on proximity, availability, latency, bandwidth and/or other factors.

It is also contemplated that the computing system 800 is operative to not include all of the components shown in FIG. 52, is operative to include other components not explicitly shown in FIG. 52, or is operative to use a completely different architecture than that shown in FIG. 52. The various illustrative logical blocks, modules, elements, circuits, and algorithms described in connection with the embodiments described herein may function to be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various components, blocks, modules, circuits and illustrative steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the particular application and the design constraints imposed on the overall system. Those skilled in the art may implement the described functionality in various ways for each particular application (e.g., arranged in a different order or divided in a different manner), but such implementation decisions should not be construed as deviating from the scope of the present invention.

The examples mentioned above are provided to serve the purpose of clarifying aspects of the invention, and it will be apparent to one skilled in the art that they do not serve to limit the scope of the invention. By nature, this invention is highly adjustable, customizable and adaptable. The examples mentioned above are just a few of the many configurations that the mentioned components can take. All modifications and improvements have been omitted from this document for reasons of conciseness and readability, but are within the scope of the present invention.