Modeling mobile network performance

BACKGROUND

0001Mobile devices, such as smart phones, tablet computers, laptop computers, and other electronic hand-held devices, are becoming increasingly popular. In order to support the growing number of mobile devices, mobile networks (e.g., third generation (3G) and fourth generation (4G) mobile networks) employ radio network subsystems with macro cells using one or more high-powered base stations. Although advances in technology have made it possible for these base stations to cover relatively large geographical areas to improve mobile communications, this is a one-size-fits-all approach that may not adequately leverage network resources to fully optimize a mobile network for mobile communications.

0002With the advent of fifth generation (5G) systems that further develop the technology of network-function virtualization (NFV) and software-defined networking (SDN), the concept of delivering network infrastructure as a service (NaaS) is being introduced. Such networks may support multi-tenancy and may include an infrastructure that supports multiple operators of different types. Consequently, an individual operator's scope of control may be constrained to one or more portions or “slices” of the network infrastructure subject to an agreement with the infrastructure owner to receive the NaaS. Therefore, different users for a self-optimizing network (SON) may target one or more individual slices of the network, where each network slice may include a different set of network functions.

BRIEF DESCRIPTION OF THE DRAWINGS

0003<figref idref="DRAWINGS">FIGS. 1A-1C</figref> are diagrams of an overview of an example implementation described herein;

0004<figref idref="DRAWINGS">FIG. 2</figref> is a diagram of an example environment in which systems and/or methods, described herein, may be implemented;

0005<figref idref="DRAWINGS">FIG. 3</figref> is a diagram of example components of one or more devices of <figref idref="DRAWINGS">FIG. 2</figref>;

0006<figref idref="DRAWINGS">FIG. 4</figref> is a diagram of example functional components of a self-optimizing network (SON) system depicted in <figref idref="DRAWINGS">FIG. 2</figref>;

0007<figref idref="DRAWINGS">FIG. 5</figref> is a flow chart of an example process for training a predictive model based on call trace information, measurement information, and/or configuration information;

0008<figref idref="DRAWINGS">FIGS. 6A and 6B</figref> are diagrams of an example implementation relating to the example process shown in <figref idref="DRAWINGS">FIG. 5</figref>;

0009<figref idref="DRAWINGS">FIG. 7</figref> is a flow chart of an example process for configuring a mobile network based on a predictive model;

0010<figref idref="DRAWINGS">FIGS. 8A-8C</figref> are diagrams of an example implementation relating to the example process shown in <figref idref="DRAWINGS">FIG. 7</figref>;

0011<figref idref="DRAWINGS">FIG. 9</figref> is a flow chart of an example process for identifying and mitigating interference based on a predictive model; and

0012<figref idref="DRAWINGS">FIGS. 10A-10C</figref> are diagrams of an example implementation relating to the example process shown in <figref idref="DRAWINGS">FIG. 9</figref>.

DETAILED DESCRIPTION

0013The following detailed description of example implementations refers to the accompanying drawings. The same reference numbers in different drawings may identify the same or similar elements.

0014Current mobile networks may fail to fully utilize detailed knowledge of users, the users' mobile devices, and other specific information to better allocate network resources in order to implement a more efficient, focused, and customized network plan. A self-organizing or self-optimizing network (SON) system may use various mechanisms to determine whether a mobile network is performing to a desired level (e.g., optimally, at a desired Quality of Service (QoS), etc.) for a given set of traffic conditions. A base station of a mobile network may contain configuration parameters that control various aspects of a cell site of the mobile network. The SON system may alter these parameters to change network behavior, based on measurements obtained by the base station (e.g., uplink information), call trace information associated with a mobile device (e.g., downlink information), or other acquired data. For example, the SON system may automatically alter various network parameters if such changes would lead to a better user experience for some or all users. The network parameters may include transmit power levels, neighbor cell relation tables, antenna electrical tilts, antenna pointing direction/angles (e.g., elevation, tilt, and/or azimuth), handover thresholds (e.g., a mobile device of a voice user on a heavily-used 4G network may be encouraged to perform a handover to a base station of another network in order to free up 4G resources), or the like.

0015The SON system may make changes to network configuration data in order to improve mobile network performance. For example, the SON system may adjust the network configuration data to alter cell sizes, to balance a load across the mobile network, to improve overall mobile network capacity, to improve overall mobile network coverage, to mitigate interference, or the like. However, some changes may lead to a better outcome than other changes. For example, a first network configuration may cause a greater improvement to mobile network capacity than a second network configuration. Without information describing both uplinks (i.e., data transmitted from mobile devices to the base station) and downlinks (i.e., data transmitted from the base station to mobile devices), the outcomes may be difficult or impossible to predict. Further, without the information describing both the uplinks and the downlinks, sources of interference (e.g., repeaters, harmonic interference, etc.) may be difficult to locate and mitigate.

0016Systems and/or methods, described herein, may provide a SON system that can model outcomes of network reconfiguration of a mobile network, such as a mobile network. The SON system may include an architecture that allows the SON system to receive measurement information, including uplink information, from base stations, and to receive call trace information, including downlink information, from the base stations. Based on the measurement information and the call trace information, the SON system may train predictive models. The predictive models may predict network performance based on observed measurement information and/or call trace information, may estimate missing measurement information/call trace information, and/or may be useful to identify and locate cellular interference (e.g., passive intermodulation products), and determine actions which may mitigate cellular interference. In this way, the SON system can predict outcomes of network reconfiguration, which improves performance of the network reconfiguration, increases mobile network capacity, and reduces mobile network interference.

0017<figref idref="DRAWINGS">FIGS. 1A-1C</figref> are diagrams of an overview of an example implementation <b>100</b> described herein. As shown in <figref idref="DRAWINGS">FIG. 1A</figref>, a base station may communicate with a mobile device via an uplink (e.g., from the mobile device to the base station) and a downlink (e.g., from the base station to the mobile device). As further shown, the base station may provide measurement information, including uplink information relating to the base station, to a SON system. As shown, the base station may provide call trace information, including downlink information relating to the mobile device, to the SON system.

0018As shown, the SON system may train a predictive model based on the uplink information and the downlink information. The predictive model may predict network performance information based on the uplink information and the downlink information. In some implementations, the predictive model may predict other information. For example, the predictive model may output estimated uplink information for base stations that do not provide uplink information to the SON system, or the like.

0019As shown in <figref idref="DRAWINGS">FIG. 1B</figref>, the SON system may determine observed network performance information based on the measurement information and the call trace information. The observed network performance information may include, for example, an uplink/downlink received signal strength indicator, a percentage of dropped calls, a modulation and coding scheme, or the like. As further shown, the SON system may propose a change to the mobile network to cause a network performance improvement. As shown, the SON system may use the predictive model to predict whether the proposed change improves performance with regard to the uplink and the downlink. Here, the proposed change improves performance with regard to the uplink and the downlink, so the SON system causes the proposed change to be implemented. As further shown, the SON system may cause the base station to reconfigure the uplink/downlink based on reconfiguration information to implement the proposed change. In this way, the SON system can change the mobile network configuration to improve network performance, and can predict an effect of the change to determine whether the change is worth implementing.

0020As shown in <figref idref="DRAWINGS">FIG. 1C</figref>, in some cases, interference may cause observed uplink information that is received by the base station to differ from uplink information transmitted by a mobile device. As further shown, to detect the interference, the SON system may determine predicted uplink information based on the predictive model determined in connection with <figref idref="DRAWINGS">FIG. 1A</figref>. As shown, the SON system may compare the observed uplink information to the predicted uplink information, and may identify the interference based on the observed uplink information differing from the predicted uplink information. As further shown, the SON system may characterize the interference (e.g., may determine a frequency of the interference, an amplitude of the interference, a wave pattern of the interference, whether a source of the interference is moving, etc.), may locate a source of the interference (e.g., based on information from the base station, based on information from several base stations, etc.), and may perform actions to mitigate the interference. In this way, the SON system identifies interference based on a predictive model, characterizes and locates the interference, and causes mitigating actions to be performed, which improves uplink performance and reduces interference at the base station.

0021<figref idref="DRAWINGS">FIG. 2</figref> is a diagram of an example environment <b>200</b> in which systems and/or methods, described herein, may be implemented. As shown in <figref idref="DRAWINGS">FIG. 2</figref>, environment <b>200</b> may include a SON system <b>210</b>, a base station <b>220</b>, a mobile device <b>230</b>, a mobile network <b>240</b> with network resources <b>245</b>, and a network <b>250</b>. Devices of environment <b>200</b> may interconnect via wired connections, wireless connections, or a combination of wired and wireless connections.

0022SON system <b>210</b> may include one or more devices capable of receiving, generating, storing, processing, and/or providing information, such as information described herein. For example, SON system <b>210</b> may include one or more computing devices, such as one or more server devices, desktop computers, workstation computers, virtual machines (VMs) provided in a cloud computing environment, or similar devices. In some implementations, SON system <b>210</b> may be utilized by an entity that manages and/or operates one or more portions of environment <b>200</b>, such as, for example, a telecommunication service provider, a television service provider, an Internet service provider, or the like.

0023Base station <b>220</b> may include one or more devices capable of transferring traffic, such as audio, video, text, and/or other traffic, destined for and/or received from mobile device <b>230</b>. In some implementations, base station <b>220</b> may include an eNB associated with an LTE network that receives traffic from and/or sends traffic to network <b>250</b>. Additionally, or alternatively, one or more base stations <b>220</b> may be associated with a RAN that is not associated with the LTE network. Base station <b>220</b> may send traffic to and/or receive traffic from mobile device <b>230</b> via an air interface. In some implementations, base station <b>220</b> may include a small cell base station, such as a base station of a microcell, a picocell, and/or a femtocell. In some implementations, base station <b>220</b> may communicate with another base station <b>220</b> of mobile network <b>240</b> regarding measurement information, network performance information, or the like.

0024Mobile device <b>230</b> may include one or more devices capable of communicating with base station <b>220</b> and/or a network (e.g., mobile network <b>240</b>, network <b>250</b>, etc.). For example, mobile device <b>230</b> may include a wireless communication device, a radiotelephone, a personal communications system (PCS) terminal (e.g., that may combine a cellular radiotelephone with data processing and data communications capabilities), a smart phone, a smart meter, a vehicle, a vending machine, a laptop computer, a tablet computer, a personal gaming system, and/or a similar device. Mobile device <b>230</b> may send traffic to and/or receive traffic from network <b>250</b> (e.g., via base station <b>220</b>).

0025Mobile network <b>240</b> may include a mobile communications network, such as 3G mobile network, a 4G mobile network, a heterogeneous network, and/or a combination of these or other types of networks. In some implementations, mobile network <b>240</b> may correspond to an evolved packet system (EPS) that includes an OSS, a radio access network (e.g., referred to as a long term evolution (LTE) network), a wireless core network (e.g., referred to as an evolved packet core (EPC) network), an Internet protocol (IP) multimedia subsystem (IMS) network, and a packet data network (PDN). The LTE network may include a base station (eNB). The EPC network may include a mobility management entity (MME), a serving gateway (SGW), a policy and charging rules function (PCRF), a PDN gateway (PGW), a base station controller (BSC), a radio network controller (RNC), an operations and maintenance centre (OMC), a network management system (NMS) and/or a network management center (NMC). The IMS network may include a home subscriber server (HSS), a proxy call session control function (P-CSCF), an interrogating call session control function (I-CSCF), and a serving call session control function (S-CSCF).

0026In some implementations, mobile network <b>240</b> may include one or more network resources <b>245</b>, such as, for example, the OSS, the eNB, the MME, the SGW, the PCRF, the PGW, the HSS, the P-CSCF, the I-CSCF, the S-CSCF, or the like. In some implementations, network resources <b>245</b> may exchange information based on an interface (e.g., an X2 interface, a northbound interface (NBI), etc.).

0027Network <b>250</b> may include one or more wired and/or wireless networks. For example, network <b>250</b> may include a mobile network, a public land mobile network (PLMN), a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), a telephone network (e.g., the Public Switched Telephone Network (PSTN)), an ad hoc network, an intranet, the Internet, a fiber optic-based network, a cloud computing network, a private network, and/or a combination of these or other types of networks. In some implementations, network <b>250</b> may include one or more device-to-device wireless networks where communication may occur through direct communication between devices, under the control of network <b>250</b> or independently. In some implementations, direct device-to-device links may comprise one or more hops. Such direct device-to-device links may be used in a cooperative manner together with point-to-point and/or point-to-multi-point links mediated by network <b>250</b>.

0028The number and arrangement of devices and networks shown in <figref idref="DRAWINGS">FIG. 2</figref> are provided as an example. In practice, there may be additional devices and/or networks, fewer devices and/or networks, different devices and/or networks, or differently arranged devices and/or networks than those shown in <figref idref="DRAWINGS">FIG. 2</figref>. Furthermore, two or more devices shown in <figref idref="DRAWINGS">FIG. 2</figref> may be implemented within a single device, or a single device shown in <figref idref="DRAWINGS">FIG. 2</figref> may be implemented as multiple, distributed devices. Additionally, or alternatively, a set of devices (e.g., one or more devices) of environment <b>200</b> may perform one or more functions described as being performed by another set of devices of environment <b>200</b>.

0029<figref idref="DRAWINGS">FIG. 3</figref> is a diagram of example components of a device <b>300</b>. Device <b>300</b> may correspond to SON system <b>210</b>, base station <b>220</b>, mobile device <b>230</b>, and/or network resource <b>245</b>. In some implementations, SON system <b>210</b>, base station <b>220</b>, mobile device <b>230</b>, and/or network resource <b>245</b> may include one or more devices <b>300</b> and/or one or more components of device <b>300</b>. As shown in <figref idref="DRAWINGS">FIG. 3</figref>, device <b>300</b> may include a bus <b>310</b>, a processor <b>320</b>, a memory <b>330</b>, a storage component <b>340</b>, an input component <b>350</b>, an output component <b>360</b>, and a communication interface <b>370</b>.

0030Bus <b>310</b> may include a component that permits communication among the components of device <b>300</b>. Processor <b>320</b> is implemented in hardware, firmware, or a combination of hardware and software. Processor <b>320</b> may include a processor (e.g., a central processing unit (CPU), a graphics processing unit (GPU), an accelerated processing unit (APU), etc.), a microprocessor, and/or any processing component (e.g., a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), etc.) that interprets and/or executes instructions. In some implementations, processor <b>320</b> may include one or more processors capable of being programmed to perform a function. Memory <b>330</b> may include a random access memory (RAM), a read only memory (ROM), and/or another type of dynamic or static storage device (e.g., a flash memory, a magnetic memory, an optical memory, etc.) that stores information and/or instructions for use by processor <b>320</b>.

0031Storage component <b>340</b> may store information and/or software related to the operation and use of device <b>300</b>. For example, storage component <b>340</b> may include a hard disk (e.g., a magnetic disk, an optical disk, a magneto-optic disk, a solid state disk, etc.), a compact disc (CD), a digital versatile disc (DVD), a floppy disk, a cartridge, a magnetic tape, and/or another type of computer-readable medium, along with a corresponding drive.

0032Input component <b>350</b> may include a component that permits device <b>300</b> to receive information, such as via user input (e.g., a touch screen display, a keyboard, a keypad, a mouse, a button, a switch, a microphone, etc.). Additionally, or alternatively, input component <b>350</b> may include a sensor for sensing information (e.g., a global positioning system (GPS) component, an accelerometer, a gyroscope, an actuator, etc.). Output component <b>360</b> may include a component that provides output information from device <b>300</b> (e.g., a display, a speaker, one or more light-emitting diodes (LEDs), etc.).

0033Communication interface <b>370</b> may include a transceiver-like component (e.g., a transceiver, a separate receiver and transmitter, etc.) that enables device <b>300</b> to communicate with other devices, such as via a wired connection, a wireless connection, or a combination of wired and wireless connections. Communication interface <b>370</b> may permit device <b>300</b> to receive information from another device and/or provide information to another device. For example, communication interface <b>370</b> may include an Ethernet interface, an optical interface, a coaxial interface, an infrared interface, a radio frequency (RF) interface, a universal serial bus (USB) interface, a Wi-Fi interface, a mobile network interface, or the like.

0034Device <b>300</b> may perform one or more processes described herein. Device <b>300</b> may perform these processes in response to processor <b>320</b> executing software instructions stored by a computer-readable medium, such as memory <b>330</b> and/or storage component <b>340</b>. A computer-readable medium is defined herein as a non-transitory memory device. A memory device includes memory space within a single physical storage device or memory space spread across multiple physical storage devices.

0035Software instructions may be read into memory <b>330</b> and/or storage component <b>340</b> from another computer-readable medium or from another device via communication interface <b>370</b>. When executed, software instructions stored in memory <b>330</b> and/or storage component <b>340</b> may cause processor <b>320</b> to perform one or more processes described herein. Additionally, or alternatively, hardwired circuitry may be used in place of or in combination with software instructions to perform one or more processes described herein. Thus, implementations described herein are not limited to any specific combination of hardware circuitry and software. For example, implementations described herein may be implemented based on network functions virtualization (NFV), a public (e.g., Internet) cloud computing environment, a private cloud computing environment, or the like.

0036The number and arrangement of components shown in <figref idref="DRAWINGS">FIG. 3</figref> are provided as an example. In practice, device <b>300</b> may include additional components, fewer components, different components, or differently arranged components than those shown in <figref idref="DRAWINGS">FIG. 3</figref>. Additionally, or alternatively, a set of components (e.g., one or more components) of device <b>300</b> may perform one or more functions described as being performed by another set of components of device <b>300</b>.

0037<figref idref="DRAWINGS">FIG. 4</figref> is a diagram of example functional components of SON system <b>210</b>. As shown in <figref idref="DRAWINGS">FIG. 4</figref>, SON system <b>210</b> may include an uplink modeler <b>410</b>, an uplink interference detector <b>420</b>, an uplink calculator <b>430</b>, a network configuration component <b>440</b>, an uplink simulator <b>450</b>, and a derived modeler <b>460</b>.

0038Uplink modeler <b>410</b> may generate and/or train predictive models to predict network performance based on call trace information and/or measurement information. In some implementations, uplink modeler <b>410</b> may receive call trace information and/or measurement information, and may generate/train predictive models accordingly. Uplink modeler <b>410</b> may provide predictive models, call trace information, and/or measurement information to other components of SON system <b>210</b> (e.g., uplink interference detector <b>420</b>, uplink calculator <b>430</b>, network configuration component <b>440</b>, uplink simulator <b>450</b>, derived modeler <b>460</b>, etc.). In some implementations, one or more functions of SON system <b>210</b> may be distributed and operate on network elements under a distributed control system. In some implementations, the one or more functions of SON system <b>210</b> may operate in a centralized manner. In such implementations, the one or more functions of SON system <b>210</b> may be determined by a centralized controlling node, may operate in a hybrid manner, or the like. In some implementations, uplink modeler <b>410</b> may perform the operations described above with regard to a downlink between base station <b>220</b> and mobile device <b>230</b>.

0039Uplink interference detector <b>420</b> may detect interference in mobile network <b>240</b> based on a predictive model and based on measurement information. Uplink interference detector <b>420</b> may receive measurement information from base station <b>220</b> and/or uplink modeler <b>410</b>, and may receive predictive models from uplink modeler <b>410</b>. Uplink interference detector <b>420</b> may characterize and/or locate sources of interference, and may provide information and/or cause an action to be performed to notify a network administrator of the sources of interference, to mitigate the sources of interference, or the like.

0040Uplink calculator <b>430</b> may determine estimated measurement information for base stations <b>220</b> and/or areas (e.g., mobile network cells, geographic areas, etc.) for which actual measurement information is not satisfactory (e.g., not received, incomplete, inadequate, does not satisfy an accuracy threshold, etc.). Uplink calculator <b>430</b> may determine the estimated measurement information for a particular base station <b>220</b>/area based on predictive models received from uplink modeler <b>410</b> and based on measurement information and/or call trace information (e.g., for base stations <b>220</b>/areas near the particular base station <b>220</b>/area, for base stations <b>220</b>/areas with historically similar measurement information and/or call trace information to the particular base station <b>220</b>/area, etc.). In some implementations, uplink calculator <b>430</b> may receive information identifying which measurement information to estimate from uplink modeler <b>410</b>, network configuration component <b>440</b>, or another component/device.

0041In some implementations, uplink calculator <b>430</b> may determine network performance information based on measurement information, call trace information, and/or predictive models received from uplink modeler <b>410</b>. Uplink calculator <b>430</b> may provide the network performance information to network configuration component <b>440</b> or another component or device.

0042Network configuration component <b>440</b> may generate network configurations, and may predict network performance effects associated with the generated network configurations, based on measurement information, call trace information, predictive models, and/or uplink metrics. In some implementations, network configuration component <b>440</b> may receive measurement information, call trace information, and predictive models from uplink modeler <b>410</b>, and may receive uplink metrics and/or estimated measurement information from uplink calculator <b>430</b>. Network configuration component <b>440</b> may provide information and/or cause an action to be performed based on the predicted network performance effects and/or the generated network configurations. For example, network configuration component <b>440</b> may cause base station <b>220</b> and/or mobile device <b>230</b> to be reconfigured, may provide information to a network administrator, or the like. In some implementations, network configuration component <b>440</b> may generate and/or train predictive models relating to a downlink network configuration of base station <b>220</b>.

0043Uplink simulator <b>450</b> may adjust uplink characteristics based on network configuration data generated by network configuration component <b>440</b>. For example, uplink simulator <b>450</b> may receive information identifying updated network configuration data and call trace information/measurement information, and may adjust network configuration information to implement the updated network configuration data with regard to an uplink between base station <b>220</b> and mobile devices <b>230</b>.

0044Derived modeler <b>460</b> may receive predictive models and call trace information from uplink modeler <b>410</b> and/or base station <b>220</b>, and may derive one or more secondary predictive models based on the predicted models and the call trace information. In some implementations, derived modeler <b>460</b> may provide the secondary predictive models and/or information determined based on the one or more secondary predictive models to network configuration component <b>440</b>.

0045The number and arrangement of functional components shown in <figref idref="DRAWINGS">FIG. 4</figref> are provided as an example. In practice, SON system <b>210</b> may include additional functional components, fewer functional components, different functional components, or differently arranged functional components than those shown in <figref idref="DRAWINGS">FIG. 4</figref>. Additionally, or alternatively, a set of functional components (e.g., one or more functional components) of SON system <b>210</b> may perform one or more functions described as being performed by another set of functional components of SON system <b>210</b>. For example, SON system <b>210</b> may interact with another SON system <b>210</b> associated with one or more other entities and/or mobile networks <b>240</b>.

0046<figref idref="DRAWINGS">FIG. 5</figref> is a flow chart of an example process <b>500</b> for training a predictive model based on call trace information, measurement information, and/or configuration information. In some implementations, one or more process blocks of <figref idref="DRAWINGS">FIG. 5</figref> may be performed by SON system <b>210</b>. In some implementations, one or more process blocks of <figref idref="DRAWINGS">FIG. 5</figref> may be performed by another device or a group of devices separate from or including SON system <b>210</b>, such as base station <b>220</b>, mobile device <b>230</b>, and network resources <b>245</b>.

0047As shown in <figref idref="DRAWINGS">FIG. 5</figref>, process <b>500</b> may include receiving call trace information identifying mobile devices, one or more cells associated with the mobile devices, and downlink information relating to the mobile devices (block <b>510</b>). For example, SON system <b>210</b> may receive call trace information. The call trace information may identify mobile devices <b>230</b> (e.g., based on a device identifier, such as an international mobile subscriber identity (IMSI), an international mobile station equipment identity (IMEI), a mobile device number (MDN), a subscriber identifier that identifies a subscriber associated with mobile device <b>230</b>, or the like). The call trace information may identify the one or more cells based on a cell identifier, such as a cell global identity (CGI), an enhanced cell global identity (E-CGI), a location area code (LAC), or the like.

0048SON system <b>210</b> may receive the call trace information from base station <b>220</b>, which may serve the one or more cells, and base station <b>220</b> may gather the call trace information for the mobile devices <b>230</b> associated with the one or more cells. For example, mobile devices <b>230</b> in the one or more cells may access mobile network <b>240</b> via base station <b>220</b>, and base station <b>220</b> may determine the call trace information based on the mobile devices <b>230</b> accessing mobile network <b>240</b> via base station <b>220</b>. Additionally, or alternatively, base station <b>220</b> may determine and provide the call trace information continuously.

0049Base station <b>220</b> may determine call trace information based on information that is collected by mobile devices <b>230</b>. Base station <b>220</b> may determine call trace information at a particular interval (e.g., every thirty seconds, every one minute, every five minutes, every fifteen minutes, every thirty minutes, once per hour, daily, etc.), and may provide the call trace information to SON system <b>210</b>. Additionally, or alternatively, SON system <b>210</b> may request the call trace information, and may receive the call trace information based on requesting the call trace information. In some implementations, SON system <b>210</b> may obtain call trace information from base station <b>220</b> and/or network resource <b>245</b> based on an application programming interface (API). For example, distributed nodes of SON system <b>210</b> may collect information based on the API, may determine call trace information based on the collected information, and may provide the call trace information to a central node of SON system <b>210</b>.

0050In some implementations, the call trace information may include downlink information. The downlink information may relate to information received by and/or provided by mobile device <b>230</b>. For example, the downlink information may include information relating to circuit-switched calls placed by mobile device <b>230</b> (e.g., a quantity of calls, a duration of calls, etc.), packet-switched calls received and/or provided by mobile device <b>230</b> (e.g., a quantity of calls, a duration of calls, etc.), Voice-over-LTE calls received and/or provided by mobile device <b>230</b> (e.g., a quantity of calls, a duration of calls, etc.), a transmit power level associated with mobile device <b>230</b>, a downlink path loss between base station <b>220</b> and mobile device <b>230</b>, a frequency associated with a downlink channel between base station <b>220</b> and mobile device <b>230</b>, a downlink received signal code power (RSCP) associated with mobile device <b>230</b>, a downlink received energy per chip (Ec) for mobile device <b>230</b>, a downlink noise power density (NO) for mobile device <b>230</b>, a received signal reference quality (RSRQ) for mobile device <b>230</b>, a power headroom identifier for mobile device <b>230</b>, a channel quality indication (CQI), a sub-band CQI, an estimate of a channel rank, information according to an E-UTRA standard (e.g., LTE 36.331, LTE 36.423 or the like.

0051As further shown in <figref idref="DRAWINGS">FIG. 5</figref>, process <b>500</b> may include receiving measurement information including uplink information related to base stations of the one or more cells (block <b>520</b>). For example, SON system <b>210</b> may receive measurement information from base station <b>220</b>. Base station <b>220</b> may be associated with the one or more cells identified by the call trace information. For example, base station <b>220</b> may serve mobile devices <b>230</b> associated with the one or more cells, and may provide measurement information relating to the one or more cells. In some implementations, base station <b>220</b> may determine measurement information at a particular interval (e.g., every five minutes, every fifteen minutes, once per hour, daily, etc.), and may provide the measurement information to SON system <b>210</b>. Additionally, or alternatively, SON system <b>210</b> may request the measurement information, and may receive the measurement information based on requesting the measurement information. Additionally, or alternatively, base station <b>220</b> may provide the measurement information continuously. In some implementations, SON system <b>210</b> may obtain measurement information from base station <b>220</b> and/or network resource <b>245</b> based on an application programming interface (API). For example, distributed nodes of SON system <b>210</b> may collect information based on the API, may determine measurement information based on the collected information, and may provide the measurement information to a central SON system <b>210</b>.

0052In some implementations, the measurement information may include uplink information. The uplink information may relate to information received by and/or provided by base station <b>220</b>. For example, the uplink information may identify a frequency of a downlink channel provided by base station <b>220</b>, an uplink RSSI associated with base station <b>220</b>, an uplink SNR for information received by base station <b>220</b>, an uplink modulation and coding scheme (MCS) associated with base station <b>220</b>, a noise floor, a throughput associated with base station <b>220</b> (e.g., in bits per second, kilobits per second, megabits per second, a quantity of calls routable, a quantity of sessions that base station <b>220</b> can maintain, etc.), a downlink propagation loss for base station <b>220</b>, a difference between an uplink channel frequency and a downlink channel frequency, a received total wideband power (RWTP), a training sequence code (TSC), a preamble, a cycle prefix, or the like.

0053In some implementations, the measurement information may relate to multiple base stations <b>220</b>. For example, the multiple base stations <b>220</b> may intercommunicate (e.g., via an X2 interface, etc.) to determine measurement information for one or more cells. The measurement information relating to multiple base stations <b>220</b> may include, for example, one or more high uplink interference indicators, one or more uplink interference overload indicators, a relative narrowband transmit power of two or more base stations <b>220</b>, an almost-blank-subframe (ABS) message transmitted between two or more base stations <b>220</b>, intended uplink/downlink configuration information, cooperative multipoint information (CoMP information), a CoMP hypothesis, network assisted interference cancellation information, or the like. The measurement information may relate to a carrier, a sub-carrier, a sub-band, a resource block, and/or a cell.

0054In some implementations, SON system <b>210</b> may receive configuration information relating to the base stations <b>220</b> and/or the mobile devices <b>230</b>. The configuration information may relate to a configuration of base station <b>220</b> and/or other network devices, based on which to provide network services. For example, the configuration information may include neighbor cell relation tables, antenna electrical tilts, antenna pointing direction/angles (e.g., elevation, tilt, and/or azimuth), handover thresholds, or the like. In some implementations, SON system <b>210</b> may receive the configuration information from base station <b>220</b>. Additionally, or alternatively, SON system <b>210</b> may receive the configuration information from another source (e.g., network resource <b>245</b>, a user input, a planning tool, etc.).

0055As further shown in <figref idref="DRAWINGS">FIG. 5</figref>, process <b>500</b> may include processing the call trace information based on the one or more cells (block <b>530</b>). For example, SON system <b>210</b> may process the call trace information based on the one or more cells identified by the call trace information. In some implementations, SON system <b>210</b> may associate mobile devices <b>230</b>, in a particular cell, with the particular cell and/or with a base station <b>220</b> that serves the particular cell. Additionally, or alternatively, SON system <b>210</b> may determine aggregate call trace information for a particular cell. For example, SON system <b>210</b> may combine call trace information from mobile devices <b>230</b> in a particular cell (e.g., by averaging the call trace information, etc.).

0056As further shown in <figref idref="DRAWINGS">FIG. 5</figref>, process <b>500</b> may include matching processed call trace information with the measurement information (block <b>540</b>). For example, SON system <b>210</b> may match processed call trace information with measurement information. In some implementations, SON system <b>210</b> may match the processed call trace information with the measurement information based on the cell with which the call trace information and/or the measurement information are associated. For example, SON system <b>210</b> may match a particular base station <b>220</b>, corresponding to a particular cell, with each mobile device <b>230</b> that the particular base station <b>220</b> serves (e.g., each mobile device <b>230</b> in the particular cell, each mobile device <b>230</b> that establishes a session with the particular base station <b>220</b>, etc.). In this way, SON system <b>210</b> may associate uplink information for a particular cell and/or base station <b>220</b>, with downlink information for the particular cell and/or base station <b>220</b>, which improves accuracy of predicted network performance effects relating to the particular cell and/or base station <b>220</b>.

0057In some implementations, SON system <b>210</b> may determine that certain information is missing, unusable, or the like. For example, SON system <b>210</b> may determine that a particular base station <b>220</b> has not provided measurement information, has provided distorted measurement information, or the like. In such a case, SON system <b>210</b> may estimate measurement information for the particular base station <b>220</b>, as described in more detail in connection with <figref idref="DRAWINGS">FIG. 7</figref>, below.

0058As further shown in <figref idref="DRAWINGS">FIG. 5</figref>, process <b>500</b> may include training a predictive model, to predict network performance information based on an input, based on the processed call trace information and the measurement information (block <b>550</b>). For example, SON system <b>210</b> may train one or more predictive models based on the processed call trace information and the measurement information. The one or more predictive models may predict network performance based on an input. For example, SON system <b>210</b> may input uplink information, downlink information, and/or configuration information to a predictive model, and the predictive model may output predicted network performance information. SON system <b>210</b> may generate and/or train the predictive model based on received measurement information, call trace information, and/or configuration information, as described in more detail above.

0059The predictive model may output predicted network performance information. The predicted network performance information may predict an uplink RSSI associated with base station <b>220</b>, an uplink SNR for information received by base station <b>220</b>, a throughput associated with base station <b>220</b> (e.g., in bits per second, kilobits per second, megabits per second, a quantity of calls routable, a quantity of sessions that base station <b>220</b> can maintain, etc.), or the like. In some implementations, the predictive model may output predicted network performance information for a set of mobile devices <b>230</b>. For example, the predictive model may output an average predicted value of the above information, a fraction of mobile devices <b>230</b> that satisfy a threshold relating to the above information, or the like. In some implementations, SON system <b>210</b> may output statistical information relating to network performance information. For example, SON system <b>210</b> may output an average of a set of network performance information, a variance of a set of network performance information, an interquartile range of a set of network performance information, a standard deviation of a set of network performance information, a ratio of a set of network performance information that satisfies a particular threshold or a target value, or the like.

0060In some implementations, SON system <b>210</b> may generate a predictive model. For example, SON system <b>210</b> may receive call trace information and measurement information, and may generate a predictive model relating the call trace information and the measurement information. In some implementations, to generate a predictive model, SON system <b>210</b> may determine a correlation between an input variable (e.g., in call trace information and/or measurement information) and an output variable (e.g., in observed network performance information). SON system <b>210</b> may determine one or more operations to perform on the input variable to predict a value of the output variable. In this way, SON system <b>210</b> may generate a predictive model to predict network performance, based on call trace information, measurement information, and observed network performance information.

0061Additionally, or alternatively, SON system <b>210</b> may train an existing predictive model. For example, SON system <b>210</b> may predict a value of network performance information based on particular network configuration data, and may implement the particular network configuration data. SON system <b>210</b> may receive call trace information and/or measurement information after implementing the particular network configuration data. SON system <b>210</b> may determine observed network performance information based on the call trace information and/or measurement information, and may compare the observed network performance information to the predicted value of the network performance information. If the predicted value of the network performance information is inaccurate, SON system <b>210</b> may adjust the predictive model to improve accuracy of predicted network performance information. In this way, SON system <b>210</b> trains a predictive model, which improves accuracy of the predicted network performance information and thus improves network performance.

0062In some implementations, SON system <b>210</b> may generate/train a secondary predictive model based on an existing predictive model. For example, SON system <b>210</b> (e.g., derived modeler <b>460</b> of SON system <b>210</b>) may receive a predictive model, and call trace information and/or measurement information for use as inputs to the predictive model. Assume that the predictive model outputs a predicted uplink RSSI based on call trace information and/or measurement information. In such a case, derived modeler <b>460</b> may determine a relationship between the predicted uplink RSSI and a secondary predicted value (e.g., a quantity of dropped calls observed at base station <b>220</b>, a quantity of blocked calls observed at base station <b>220</b>, etc.). Derived modeler <b>460</b> may, in some implementations, perform an analysis (e.g., a linear regression analysis, a multiple regression analysis, etc.) on the predicted uplink RSSI, the secondary predicted value, and one or more variables in the call trace information and/or measurement information, to generate a secondary predictive model. In this way, SON system <b>210</b> may generate a secondary predictive model to predict network performance information based on a predicted uplink RSSI, which increases a breadth of information that SON system <b>210</b> may predict, and improves network resiliency.

0063Although <figref idref="DRAWINGS">FIG. 5</figref> shows example blocks of process <b>500</b>, in some implementations, process <b>500</b> may include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in <figref idref="DRAWINGS">FIG. 5</figref>. Additionally, or alternatively, two or more of the blocks of process <b>500</b> may be performed in parallel.

0064<figref idref="DRAWINGS">FIGS. 6A and 6B</figref> are diagrams of an example implementation <b>600</b> relating to example process <b>500</b> shown in <figref idref="DRAWINGS">FIG. 5</figref>. <figref idref="DRAWINGS">FIGS. 6A and 6B</figref> show an example of training a predictive model based on call trace information, measurement information, and/or configuration information.

0065As shown in <figref idref="DRAWINGS">FIG. 6A</figref>, mobile device <b>230</b> may periodically determine information <b>605</b> (e.g., an RXLEV value, an RXQUAL value, an RSCP value, an EcN0 value, an RSRP value, and an RSRQ value), and may provide the determined information <b>605</b> to base station <b>220</b>. Base station <b>220</b> may provide measurement information <b>610</b> to SON system <b>210</b>. The measurement information <b>610</b> may include uplink information relating to base station <b>220</b>. As further shown, base station <b>220</b> may determine call trace information <b>615</b> based on the determined information <b>605</b>, and may provide the call trace information <b>615</b> to SON system <b>210</b>. In some implementations, base station <b>220</b> may provide call trace information <b>615</b> to SON system <b>210</b> via one or more network resources <b>245</b> (e.g., an intermediate controller node, management node, one or more mobile edge computing (MEC) resources, etc.). In situations where call trace information <b>615</b> is collected by one or more MEC resources, the one or more MEC resources may collect information from mobile devices <b>230</b>, may process the information to determine call trace information <b>615</b>, and may provide call trace information <b>615</b> to SON system <b>210</b>.

0066As shown in <figref idref="DRAWINGS">FIG. 6B</figref>, uplink modeler <b>410</b> of SON system <b>210</b> receives measurement information <b>610</b> and call trace information <b>615</b>. As shown by reference number <b>620</b>, uplink modeler <b>410</b> groups call trace information <b>615</b> based on cell identifiers included in call trace information <b>615</b>. For example, uplink modeler <b>410</b> may receive call trace information <b>615</b> for multiple mobile devices <b>230</b> and may group mobile devices <b>230</b> that are associated with the same cell identifiers. As shown by reference number <b>625</b>, uplink modeler <b>410</b> may determine processed call trace information based on grouping the call trace information.

0067As shown by reference number <b>630</b>, uplink modeler <b>410</b> may match processed call trace information <b>625</b> with measurement information <b>610</b> based on cell identifiers included in processed call trace information <b>625</b> and measurement information <b>610</b>. For example, uplink modeler <b>410</b> may match measurement information <b>610</b>, for a particular base station <b>220</b> and/or cell, with processed call trace information <b>625</b> for mobile devices <b>230</b> associated with the particular base station <b>220</b> and/or cell. As shown by reference number <b>635</b>, uplink modeler <b>410</b> may generate/train a predictive model to output predicted network performance information based on measurement information <b>610</b> and call trace information <b>615</b>.

0068As indicated above, <figref idref="DRAWINGS">FIGS. 6A and 6B</figref> are provided merely as an example. Other examples are possible and may differ from what was described with regard to <figref idref="DRAWINGS">FIGS. 6A and 6B</figref>.

0069<figref idref="DRAWINGS">FIG. 7</figref> is a flow chart of an example process <b>700</b> for configuring a mobile network based on a predictive model. In some implementations, one or more process blocks of <figref idref="DRAWINGS">FIG. 7</figref> may be performed by SON system <b>210</b>. In some implementations, one or more process blocks of <figref idref="DRAWINGS">FIG. 7</figref> may be performed by another device or a group of devices separate from or including SON system <b>210</b>, such as base station <b>220</b>, mobile device <b>230</b>, and/or network resources <b>245</b>.

0070As shown in <figref idref="DRAWINGS">FIG. 7</figref>, process <b>700</b> may include obtaining information identifying a predictive model, a network configuration, call trace information, and/or measurement information (block <b>710</b>). For example, SON system <b>210</b> may obtain information identifying a predictive model, a network configuration, call trace information, and/or measurement information. In some implementations, SON system <b>210</b> may obtain the information identifying the network configuration, the call trace information, and/or the measurement information from base station <b>220</b>, as described in more detail in connection with <figref idref="DRAWINGS">FIG. 5</figref>, above. In some implementations, SON system <b>210</b> may generate and/or train the predictive model, as described in more detail in connection with <figref idref="DRAWINGS">FIG. 5</figref>, above. In some implementations, SON system <b>210</b> may obtain the information identifying the predictive model, the network configuration, the call trace information, and/or the measurement information from another device. For example, another SON system <b>210</b>, network resource <b>245</b>, or another device may provide the information to SON system <b>210</b>.

0071As further shown in <figref idref="DRAWINGS">FIG. 7</figref>, process <b>700</b> may include determining estimated uplink information for cells without associated measurement information (block <b>720</b>). For example, SON system <b>210</b> may determine estimated uplink information for a cell that is associated with unsatisfactory (e.g., missing, inaccurate, incomplete, etc.) measurement information. The estimated uplink information may include projected values of uplink information (e.g., uplink information, as described in more detail in connection with <figref idref="DRAWINGS">FIG. 5</figref>, above). In some implementations, SON system <b>210</b> may determine the estimated uplink information based on the predictive model obtained in connection with block <b>710</b>. For example, SON system <b>210</b> may input the call trace information for the cell to the predictive model, and may determine an output of the predictive model, including the estimated uplink information.

0072In some implementations, SON system <b>210</b> may determine the estimated uplink information for a particular cell based on uplink information for another cell. For example, SON system <b>210</b> may determine the estimated uplink based on uplink information for neighboring cells, based on uplink information for cells associated with similar mobile devices <b>230</b> and/or historically similar call trace information, based on another cell specified by a network administrator, based on an average of all cells monitored by SON system <b>210</b>, or the like.

0073As further shown in <figref idref="DRAWINGS">FIG. 7</figref>, process <b>700</b> may include determining observed network performance information based on the obtained information (block <b>730</b>). For example, SON system <b>210</b> may determine observed network performance information based on the information identifying the predictive model, the network configuration, the call trace information, the obtained measurement information, and/or the estimated measurement information.

0074The observed network performance information may include an uplink RSSI associated with base station <b>220</b>, an uplink SNR for information received by base station <b>220</b>, a throughput associated with base station <b>220</b> (e.g., in bits per second, kilobits per second, megabits per second, a quantity of calls that were successfully routed, a quantity of sessions that base station <b>220</b> can maintain, etc.), or the like. In some implementations, the observed network performance information may include a combination of the above information (e.g., an average of network performance information values across a set of mobile devices <b>230</b>, a combined throughput of multiple base stations <b>220</b>, etc.).

0075As further shown in <figref idref="DRAWINGS">FIG. 7</figref>, process <b>700</b> may include changing network configuration data, which may be based on the information identifying the network configuration, to generate changed network configuration data (block <b>740</b>). For example, SON system <b>210</b> may change the network configuration data of mobile network <b>240</b> to determine a possible changed network configuration data. SON system <b>210</b> may change the network configuration data to attempt to improve network performance, as identified by the network performance information. For example, SON system <b>210</b> may change the network configuration data to attempt to improve an RSCP, an RSSI, an Ec, an NO, an SNR, a throughput, or the like, of base station <b>220</b>.

0076In some implementations, SON system <b>210</b> may change the network configuration data by rerouting calls from a first cell/base station <b>220</b> to a second cell/base station <b>220</b>, by changing a power level of a signal transmitted by base station <b>220</b> and/or mobile device <b>230</b>, by changing a cell geometry, by changing a signal modulation scheme, by changing a signal coding scheme, or by performing a similar action.

0077As further shown in <figref idref="DRAWINGS">FIG. 7</figref>, process <b>700</b> may include determining predicted network performance information based on the changed network configuration data and based on the predictive model (block <b>750</b>). For example, SON system <b>210</b> may determine predicted network performance information based on the changed network configuration data and based on the predictive model. In some implementations, SON system <b>210</b> may input parameters of the changed network configuration data to the predictive model, and the predictive model may output the predicted network performance information. In some implementations, SON system <b>210</b> may determine the predicted network performance information based on an output of the predictive model. For example, the predictive model may output predicted call trace information and/or predicted measurement information based on the changed network configuration data information, and SON system <b>210</b> may determine the predicted network performance information based on the predicted call trace information and/or predicted measurement information.

0078As further shown in <figref idref="DRAWINGS">FIG. 7</figref>, process <b>700</b> may include selectively transmitting the changed network configuration data based on comparing the predicted network performance information to the observed network performance information (block <b>760</b>). For example, SON system <b>210</b> may compare the predicted network performance information to the observed network performance information. If the predicted network performance information indicates an improvement of network performance relative to the observed network performance information, SON system <b>210</b> may transmit the changed network configuration data to cause the changed network configuration data to be implemented.

0079If the predicted network performance information does not indicate an improvement of network performance based on the changed network configuration data, SON system <b>210</b> may not cause the network configuration data to be adjusted. In such a case, SON system <b>210</b> may generate another changed network configuration data, may determine predicted network performance information for the other changed network configuration data, and may accordingly cause the network configuration data to be adjusted or generate yet another changed network configuration data. In this way, SON system <b>210</b> may iteratively adjust the network configuration data, which iteratively improves network performance over time.

0080In some implementations, SON system <b>210</b> may determine configuration actions to perform to cause the network configuration data to be adjusted. For example, SON system <b>210</b> may determine one or more configuration actions to be performed (e.g., by base station <b>220</b>), with regard to an uplink, based on an updated power level, an updated cell geometry, an updated signal modulation/coding scheme, or the like. The configuration action may include, for example, modifying transmitted power levels, neighbor cell relation tables, antenna electrical tilts, antenna mechanical tilts, antenna pointing direction/angles (e.g., elevation, tilt, and/or azimuth), handover thresholds, or the like. In some implementations, SON system <b>210</b> may recommend deactivating one or more base stations <b>220</b> and/or activating one or more base stations <b>220</b>. In this way, SON system <b>210</b> may determine configuration actions to perform with regard to an uplink of base station <b>220</b> to adjust network configuration data, which improves performance of the uplink and thus improves mobile network performance.

0081Although <figref idref="DRAWINGS">FIG. 7</figref> shows example blocks of process <b>700</b>, in some implementations, process <b>700</b> may include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in <figref idref="DRAWINGS">FIG. 7</figref>. Additionally, or alternatively, two or more of the blocks of process <b>700</b> may be performed in parallel.

0082<figref idref="DRAWINGS">FIGS. 8A-8C</figref> are diagrams of an example implementation <b>800</b> relating to example process <b>700</b> shown in <figref idref="DRAWINGS">FIG. 7</figref>. <figref idref="DRAWINGS">FIGS. 8A-8C</figref> show an example of configuring a mobile network based on a predictive model.

0083As shown in <figref idref="DRAWINGS">FIG. 8A</figref>, and by reference number <b>805</b>, uplink modeler <b>410</b> may provide information identifying a predictive model to uplink calculator <b>430</b>. As shown by reference number <b>810</b>, uplink calculator <b>430</b> may estimate uplink information for cells that are missing uplink information based on the predictive model. For example, uplink calculator <b>430</b> may input call trace information relating to the cells to the predictive model, or the like. As shown by reference number <b>815</b>, uplink calculator <b>430</b> may provide estimated uplink information to network configuration component <b>440</b>. As shown by reference number <b>820</b>, uplink modeler <b>410</b> may provide the predictive model, processed call trace information, and measurement information to network configuration component <b>440</b>.

0084As shown by reference number <b>825</b>, network configuration component <b>440</b> may determine observed network performance information based on the processed call trace information and the measurement information received from uplink modeler <b>410</b>. As shown by reference number <b>830</b>, network configuration component <b>440</b> may generate possible changes to network configuration data, such as rerouting calls, changing power levels of base station <b>220</b> and/or mobile device <b>230</b>, changing cell geometry, changing signal modulation, changing signal coding, or the like.

0085As shown in <figref idref="DRAWINGS">FIG. 8B</figref>, and by reference number <b>835</b>, network configuration component <b>440</b> may input, to the predictive model, the possible changes to the network configuration data, the processed call trace information, and the measurement information. As shown by reference number <b>840</b>, the predictive model may output predicted network performance information. Assume that network configuration component <b>440</b> compares the predicted network performance information to the observed network performance information to determine whether to implement the changes to the network configuration data. Assume further that network configuration component <b>440</b> determines to implement the changes to the network configuration data. As shown by reference number <b>845</b>, network configuration component <b>440</b> may provide information identifying the possible changes to the network configuration data to uplink simulator <b>450</b>. As shown by reference number <b>850</b>, based on the possible changes to the network configuration, uplink simulator <b>450</b> may provide uplink reconfiguration information to network configuration component <b>440</b>. The uplink reconfiguration information may identify operations to perform to reconfigure base station <b>220</b> based on the changes to the network configuration data.

0086As shown in <figref idref="DRAWINGS">FIG. 8C</figref>, and by reference number <b>855</b>, SON system <b>210</b> may provide reconfiguration information, including the uplink reconfiguration information, to base station <b>220</b>. As shown by reference number <b>860</b>, base station <b>220</b> may reconfigure uplinks and/or downlinks based on the reconfiguration information. In this way, SON system <b>210</b> reconfigures mobile network <b>240</b> based on a predictive model, which improves efficiency of mobile network <b>240</b> and reduces uncertainty in implementing network changes.

0087As indicated above, <figref idref="DRAWINGS">FIGS. 8A and 8B</figref> are provided merely as an example. Other examples are possible and may differ from what was described with regard to <figref idref="DRAWINGS">FIGS. 8A and 8B</figref>.

0088<figref idref="DRAWINGS">FIG. 9</figref> is a flow chart of an example process <b>900</b> for identifying and mitigating interference based on a predictive model. In some implementations, one or more process blocks of <figref idref="DRAWINGS">FIG. 9</figref> may be performed by SON system <b>210</b>. In some implementations, one or more process blocks of <figref idref="DRAWINGS">FIG. 9</figref> may be performed by another device or a group of devices separate from or including SON system <b>210</b>, such as base station <b>220</b>, mobile device <b>230</b>, and network resources <b>245</b>.

0089As shown in <figref idref="DRAWINGS">FIG. 9</figref>, process <b>900</b> may include receiving observed uplink information from a base station that receives a potentially interfering signal (block <b>910</b>). For example, SON system <b>210</b> may receive observed uplink information from base station <b>220</b>. In some implementations, SON system <b>210</b> may receive the observed uplink information in association with measurement information from base station <b>220</b>, as described in more detail in connection with <figref idref="DRAWINGS">FIG. 5</figref>, above. Base station <b>220</b> may be associated with a source of interference. For example, the source of interference may broadcast an interfering signal that base station <b>220</b> receives, may alter a signal transmitted by mobile device <b>230</b> via the uplink, or the like. The source of interference may include, for example, a coaxial cable egress signal, a bidirectional signal amplifier, a fluorescent light, a higher-order resonance effect associated with a signal, a mobile device <b>230</b> that acts as a relay for communications to/from the network by another mobile device <b>230</b>, or the like.

0090As further shown in <figref idref="DRAWINGS">FIG. 9</figref>, process <b>900</b> may include determining predicted uplink information for the base station based on a predictive model (block <b>920</b>). For example, SON system <b>210</b> may determine predicted uplink information for base station <b>220</b> based on a predictive model. The predictive model may predict uplink information, for a particular base station <b>220</b>, based on measurement information and/or call trace information associated with the particular base station <b>220</b>. SON system <b>210</b> may generate and/or train the predictive model, as described in more detail in connection with block <b>550</b> of <figref idref="DRAWINGS">FIG. 5</figref>, above.

0091As further shown in <figref idref="DRAWINGS">FIG. 9</figref>, process <b>900</b> may include comparing the observed uplink information and the predicted uplink information to identify that the potentially interfering signal is an interfering signal (block <b>930</b>). For example, SON system <b>210</b> may compare the observed uplink information and the predicted uplink information to identify the interfering signal. In some implementations, SON system <b>210</b> may determine whether an interfering signal is present based on whether the observed uplink information differs from the predicted uplink information by a particular threshold. If a difference between the observed uplink information and the predicted uplink information satisfies the particular threshold, then SON system <b>210</b> may determine that base station <b>220</b> has received an interfering signal. If the difference does not satisfy the particular threshold, SON system <b>210</b> may determine that base station <b>220</b> has not received an interfering signal.

0092As further shown in <figref idref="DRAWINGS">FIG. 9</figref>, process <b>900</b> may include characterizing the interfering signal (block <b>940</b>). For example, SON system <b>210</b> may characterize the interfering signal by determining a frequency of the interfering signal, an amplitude of the interfering signal, an arrival time of the interfering signal, whether a source of the interfering signal is stationary or mobile, or the like.

0093SON system <b>210</b> may analyze the interfering signal to characterize the interfering signal. For example, SON system <b>210</b> may analyze sub-band effects to determine a frequency range associated with the interfering signal. As another example, SON system <b>210</b> may analyze errors in physical resource blocks to determine a length of details in the interfering signal. As another example, SON system <b>210</b> may determine an impact, of the interfering signal, in the time domain, to determine the length of details in the interfering signal. As yet another example, SON system <b>210</b> may determine whether the amplitude of the interfering signal is constant or variable. As another example, SON system <b>210</b> may analyze changes in the frequency signature of the interfering signal to determine whether the source of the interfering signal is stationary or mobile. As yet another example, SON system <b>210</b> may analyze individual symbols (e.g., bits, bytes, etc.) of signals received by base station <b>220</b> to determine a time of arrival of the interfering signal.

0094In some implementations, an interfering signal may be received by multiple base stations <b>220</b>. For example, the interfering signal may reach multiple cells and, thus, may be received by multiple base stations <b>220</b>. In such a case, SON system <b>210</b> may characterize the interfering signal for two or more of the multiple base stations <b>220</b>. By characterizing the interfering signal with regard to two or more base stations <b>220</b>, SON system <b>210</b> may improve accuracy of locating a source of the interference, as described in more detail below.

0095As further shown in <figref idref="DRAWINGS">FIG. 9</figref>, process <b>900</b> may include locating a source of the interfering signal (block <b>950</b>). For example, SON system <b>210</b> may locate a source of the interference. To locate the source of the interference, SON system <b>210</b> may analyze measurement information received from one or more base stations <b>220</b>. In some implementations, SON system <b>210</b> may analyze measurement information received from multiple, different base stations <b>220</b>, which improves accuracy of the location.

0096SON system <b>210</b> may analyze measurement information, configuration information and/or characterization information to locate a source of the interfering signal. For example, SON system <b>210</b> may compare amplitudes and/or times of arrival associated with different base stations <b>220</b> to locate the source. As another example, SON system <b>210</b> may analyze an angle of arrival of the interfering signal and/or the antenna pattern generated by the interfering signal to locate the source. As yet another example, SON system <b>210</b> may determine whether the frequency of the interfering signal covers multiple frequency bands and/or carrier bands.

0097As another example, SON system <b>210</b> may use characterization information from multiple base stations <b>220</b> to determine whether a particular interfering signal is received by the multiple base stations <b>220</b>. If the particular interfering signal is received by the multiple base stations <b>220</b>, SON system <b>210</b> may analyze measurement information and/or configuration information for the multiple base stations <b>220</b> to generate a set of possible locations for the source, and to attempt to narrow down the location of the source from the set of possible locations. In this way, SON system <b>210</b> locates a source of an interfering signal, which facilitates mitigation of the interfering symbol and improves network performance.

0098As further shown in <figref idref="DRAWINGS">FIG. 9</figref>, process <b>900</b> may include causing an action to be performed to mitigate the interfering signal (block <b>960</b>). For example, SON system <b>210</b> may cause an action to be performed to mitigate the interfering signal. In some implementations, SON system <b>210</b> may determine changed network configuration data to mitigate the interfering signal, and may perform operations described in connection with <figref idref="DRAWINGS">FIGS. 5 and 7</figref>, above. In this way, SON system <b>210</b> may use a predictive model to change network configuration data, which improves network performance and/or simplifies implementation of the changed network configuration data.

0099In some implementations, SON system <b>210</b> may cause an action to be performed. For example, SON system <b>210</b> may cause base station <b>220</b> to hand over mobile devices <b>230</b> (e.g., onto another base station <b>220</b>, another frequency band, another cell, etc.). As another example, SON system <b>210</b> may cause base station <b>220</b> to tilt an antenna downward to reduce a size of a cell served by base station <b>220</b>. As another example, SON system <b>210</b> may perform a voltage standing wave ratio test with regard to base station <b>220</b> to determine an amount of power reflected by antennas of base station <b>220</b>. As another example, SON system <b>210</b> may provide a notification to an entity (e.g., a network administrator, an engineer, a technician to perform a drive test to determine a physical location of the source of interference, etc.). As another example, SON system <b>210</b> may narrow a bandwidth of base station <b>220</b> to exclude the frequency of the interfering signal. As another example, SON system <b>210</b> may remove one or more network resources <b>245</b> from uplink scheduling. As another example, SON system <b>210</b> may change a downlink frequency of a downlink between base station <b>220</b> and one or more mobile devices <b>230</b>. In some implementations, SON system <b>210</b> may perform another type of action. In this way, SON system <b>210</b> characterizes, locates, and mitigates an interfering signal based on call trace information and/or measurement information, which reduces noise at base station <b>220</b> and improves network performance.

0100Although <figref idref="DRAWINGS">FIG. 9</figref> shows example blocks of process <b>900</b>, in some implementations, process <b>900</b> may include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in <figref idref="DRAWINGS">FIG. 9</figref>. Additionally, or alternatively, two or more of the blocks of process <b>900</b> may be performed in parallel.

0101<figref idref="DRAWINGS">FIGS. 10A-10C</figref> are diagrams of an example implementation <b>1000</b> relating to example process <b>900</b> shown in <figref idref="DRAWINGS">FIG. 9</figref>. <figref idref="DRAWINGS">FIGS. 10A-10C</figref> show an example of identifying and mitigating interference based on the predictive model.

0102As shown in <figref idref="DRAWINGS">FIG. 10A</figref>, mobile device <b>230</b> may provide a signal to base station <b>220</b> via an uplink. As shown by reference number <b>1005</b>, base station <b>220</b> may receive an interfering signal from an interference source. As shown by reference number <b>1010</b>, base station <b>220</b> may receive observed uplink information <b>1010</b>, which may differ from uplink information transmitted by mobile device <b>230</b> based on the interfering signal.

0103As shown in <figref idref="DRAWINGS">FIG. 10B</figref>, SON system <b>210</b> may provide observed uplink information <b>1010</b> to uplink modeler <b>410</b>. As further shown, uplink modeler <b>410</b> may generate/train a predictive model based on observed uplink information <b>1010</b> and other information, and may provide the predictive model to uplink interference detector <b>420</b>. As shown by reference number <b>1015</b>, uplink interference detector <b>420</b> may determine predicted uplink information for base station <b>220</b> based on the predictive model. As shown by reference number <b>1020</b>, uplink interference detector <b>420</b> may detect the interfering signal based on a difference between the observed uplink information and the predicted uplink information. As shown by reference number <b>1025</b>, uplink interference detector <b>420</b> may characterize the interfering signal, and may locate a source of the interfering signal. Here, the interfering signal is associated with a stationary source and a frequency of 1950 megahertz (MHz), and a source of the interfering signal is located in cell <b>40573</b>.

0104As shown in <figref idref="DRAWINGS">FIG. 10C</figref>, and by reference number <b>1030</b>, SON system <b>210</b> may determine mitigating actions to perform to mitigate the interfering signal. Here, SON system <b>210</b> hands over mobile devices <b>230</b> to a neighboring base station <b>220</b>, causes base station <b>220</b> to tilt an antenna downward, and prompts an engineering team to perform a drive test. As shown by reference number <b>1035</b>, SON system <b>210</b> may provide reconfiguration information to base station <b>220</b>, and base station <b>220</b> may hand over mobile devices <b>230</b> and tilt the antenna downward based on the reconfiguration information. As shown by reference number <b>1040</b>, SON system <b>210</b> may prompt a team of engineers to perform a drive test in cell <b>40573</b>. In this way, SON system <b>210</b> identifies, characterizes, and locates an interfering signal based on a predictive model, which permits SON system <b>210</b> to cause mitigating actions to be performed to improve uplink performance.

0105As indicated above, <figref idref="DRAWINGS">FIGS. 10A-10C</figref> are provided merely as an example. Other examples are possible and may differ from what was described with regard to <figref idref="DRAWINGS">FIGS. 10A-10C</figref>.

0106In this way, SON system <b>210</b> predicts outcomes of network reconfiguration based on a predictive model, which improves performance of the network reconfiguration, increases mobile network capacity, and reduces mobile network interference. Further, SON system <b>210</b> locates and mitigates sources of cellular interference, which improves cellular uplink performance.

0107The foregoing disclosure provides illustration and description, but is not intended to be exhaustive or to limit the implementations to the precise form disclosed. Modifications and variations are possible in light of the above disclosure or may be acquired from practice of the implementations.

0108As used herein, the term component is intended to be broadly construed as hardware, firmware, and/or a combination of hardware and software.

0109Some implementations are described herein in connection with thresholds. As used herein, satisfying a threshold may refer to a value being greater than the threshold, more than the threshold, higher than the threshold, greater than or equal to the threshold, less than the threshold, fewer than the threshold, lower than the threshold, less than or equal to the threshold, equal to the threshold, etc.

0110It will be apparent that systems and/or methods, described herein, may be implemented in different forms of hardware, firmware, or a combination of hardware and software. The actual specialized control hardware or software code used to implement these systems and/or methods is not limiting of the implementations. Thus, the operation and behavior of the systems and/or methods were described herein without reference to specific software code—it being understood that software and hardware can be designed to implement the systems and/or methods based on the description herein.

0111Even though particular combinations of features are recited in the claims and/or disclosed in the specification, these combinations are not intended to limit the disclosure of possible implementations. In fact, many of these features may be combined in ways not specifically recited in the claims and/or disclosed in the specification. Although each dependent claim listed below may directly depend on only one claim, the disclosure of possible implementations includes each dependent claim in combination with every other claim in the claim set.

0112No element, act, or instruction used herein should be construed as critical or essential unless explicitly described as such. Also, as used herein, the articles “a” and “an” are intended to include one or more items, and may be used interchangeably with “one or more.” Furthermore, as used herein, the term “set” is intended to include one or more items (e.g., related items, unrelated items, a combination of related items and unrelated items, etc.), and may be used interchangeably with “one or more.” Where only one item is intended, the term “one” or similar language is used. Also, as used herein, the terms “has,” “have,” “having,” or the like are intended to be open-ended terms. Further, the phrase “based on” is intended to mean “based, at least in part, on” unless explicitly stated otherwise.