Untitled record

Full description

Sister Ar'a's background

As known in the technology, direction finding (DF) can be described as determining the direction in which a received RF signal was transmitted. To make this determination, the DF system receives RF signals at an antenna element. element or

<p dir="rtl">5 More and processes the signals in a receiver. Increasing the number of antenna elements that receive the RF signal and feeding each received signal to the receiver channel for further processing can increase the accuracy of the estimated value. However, many receiving systems contain a limited number of receiver channels through which samples obtained from multiple antenna elements are received and processed simultaneously.</p>

<p dir="rtl">10 General description of the invention</p>

The concepts, systems and methods described here are directed towards direction finding techniques (DF finding) using a switched network architecture to couple a first configuration of antenna elements to a second configuration of fewer existing channels in a radio frequency (RF) receiver. RF signals are processed appropriately to the DF 15 direction finding (DF) processor, which fuses sampled data at phases in the array element configuration. Then, these combined data samples can be used to estimate the direction of a received signal. In some embodiments, the antenna elements are configured as subarrays and the switching network associates different sets of subarrays with RF receive channels during different dwell times. Then, the data collected during each residence time 20 is used to form a sample covariance matrix

8844

-3-

SCMs and sample covariance matrices (SCMs) are multiple spaces combined to provide the values of the aggregate covariance matrix in the total covariance matrix to estimate the direction of a received signal.

The systems and methods described herein may include one or more of the following features 5 alone or in combination with another feature.

According to the concepts described here, the DF system includes a first configuration of array antenna elements coupled to a second variant of RF receive channels by a switched network. In some embodiments, the number of array antenna elements can be greater than the number of receive channels and during a given residence time the switching network is configured to select a number of antenna elements 10 equal to the number of receive channels such that signals from a selected number of antenna elements are coupled

The antenna is equal to the same number of receiving channels. This method ensures that data are collected from the phase centers of a variety of array elements and fed in substantially synchronized fashion to the independent receiving channels of the RF receiver. This is repeated for different configurations of antenna elements during different residence times.

<p dir="rtl">15 In some embodiments, the antenna elements are configured as subarrays. And in response</p>

According to the control signals provided to it, the switching network switches between different subarrays contained in the subarray configuration of an array antenna so that data is collected from different subarrays during different dwell times.

Each subarray may include one or more antenna elements. Significantly, during an initial dwell time, the switching network associates a selected combination or group of subarrays with a set of receive channels where the number of subarrays is equal to the number of receive channels. During a subsequent dwell time, the switching array couples a different combination (or group) of subarrays to the receive channels. This process is repeated for each combination of dwell times and a corresponding combination of subarrays. Therefore, the receiver 25 samples data from a variety of different subarray combinations at a similar variety of different residence times. Data can be collected substantially at the center of the phases of each subarray

8844

-4-

Although it should be realized that it is not necessary that each subarray have the same phase center.

The data collected during each residence time are used to form a composite spatial SCM. Each of the SCMs formed in this way can be combined according to the residence time to form the total covariance matrix.

The information of the compound angles obtained from the total covariance matrix 5 can be extracted and provided to the DF processor, which uses the values obtained from the total covariance matrix to provide an output signal indicating the direction of the received signal.

In one embodiment, the DF processor uses MINDIST technology to provide precise determination of the direction of arrival of RF signals incident on an antenna having randomly placed antenna phases in a computationally feasible manner. In one embodiment, the MINDIST technique compares the elements (i.e., 10 matrix elements) composing the reduced spatial SCM with principal component vectors at a specified phase center location as a function of the frequency and angles of the antenna beams. The main storage components may be stored in one or more tables, combined into groups or otherwise arranged as a function of phase center location, frequency and/or angle of the antenna beams.

In one embodiment, the SCM elements obtained from the total covariance matrix 15 can be determined to have non-zero angles (i.e., angles greater or less than zero) and a portion of these non-zero elements can be used to form principal components of the principal component table. table (or p-table).

By eliminating the zero-valued SCM components and realizing the SCM symmetry, less than half of the SCM components are required for calculations. By extracting half of the SCM elements and using them 20 in subsequent calculations, the overall computation time of the method called MINDIST can be reduced. For each of the principal components, vector data can be calculated and stored (e.g., in tables). In some embodiments, vector data can be pre-computed. Therefore, for each combination of extracted principal components, the corresponding vector data can be identified and used

To configure the table m. Therefore, table M contains values for the main components stored in it. In some 25 embodiments, the extracted principal component values can be classified by frequency and angle data.

When running, the test point can be compared with each of the entries in the table

8844

-5-

To specify the minimum distance point. The MINDIST method can compare distance measurements with previously calculated tables, which makes the methods presented here practically practical.

Arithmetic. The minimum distance point corresponds to the direction of arrival of ergonomic plane waves incident on the array elements. Therefore, the MINDIST technique provides an estimate of the arrival angles of RF signals.

<p dir="rtl">5 In one aspect, in an array antenna having a plurality of subarrays, a method for directing it is included</p>

Direction on signal reception with an array antenna and data acquisition using a variety of subarrays. In one embodiment, each subarray can acquire data during a dwell time selected from a plurality of dwell times. The method further includes creating a combination of spatial SCMs according to the residence time using data corresponding to one or more of the

<p dir="rtl">10 Formation of subarrays and combination of spatial SCMs according to the residence time in a complex manner</p>

To form the total covariance matrix.

In one embodiment, the method further includes pre-computing a principal component table using angle and frequency measurements of the one or more principal components, extracting one or more principal components in the form of composite phases from the total covariance matrix to form a test point,

<p dir="rtl">15 Determine a distance between the test point and each value in the pre-calculated principal components table and determine the minimum distance point based on the distances that were determined between the test point and each value in the principal components table, where the minimum distance point corresponds to the direction of the signals that were received.</p>

In some embodiments, data may be collected at a plurality of stages, where each stage includes two or more switches. For example, data can be received at a first stage

<p dir="rtl">20 From at least one sub-array, data can be supplied to two or more switches in a second stage for the switches. Data can be received in the second phase of the switches starting from the first phase of the switches, and data from at least part of one subarray can be supplied to a standard direction finding module.</p>

In one embodiment, the method includes creating one or more residence times using

<p dir="rtl">25 Data obtained from a plurality of subarrays, where each residence time corresponds to at least one residence time in the plurality of different residence times. In some embodiments,</p>

8844

-6-

The method includes collecting data by a first subarray during a first residence time; Data acquisition by a second subarray during a second dwell time; Form a spatial sample SCM (covariance matrix) according to the residence time for both the first subarray and the second subarray, and combine the spatial SCMs according to the residence time in a complex manner to form the aggregate covariance matrix 5.

In one embodiment, the method includes supplying a first set of data from a first set of subarrays to a permutation matrix and supplying a second set of subarrays directly to a modular trend-finding module. The data may include angle measurements corresponding to the received data with respect to the phase center of one or more array element configurations. In some 10 embodiments, both the residence time SCMs and the total covariance matrix can include angle measurements of a variety of array elements. The method may include determining the phase difference between each of the elements in the total covariance matrix using angular measurements.

In one embodiment, the method includes specifying vector data for each combination of array elements. Vector data can include angle and frequency measurements. In some 15 embodiments, the principal components table may include principal component data that is classified according to frequency and angle measurements for each of the one or more principal components.

In another aspect, a direction-finding system is provided that includes an assortment of array elements intended to receive signals, a direction-finding standard unit, and one or more receiving channels intended to couple the array elements to the direction-finding standard unit. In one embodiment, the number of array elements may be greater than the number of receive channels. The one or more receiving channels include a switching network placed in a signaling path between the array elements and the standard unit to find the direction to switch between the different sets of subarrays in the array elements and collect data for at least one set of subarrays during a selected residence time in the array Residence times provide data to the standard unit to find the trend.

<p dir="rtl">25 In one embodiment, at least one array element can be in each of</p>

Subarrays are effectively subarrays and can have at least one array element in each

8844

-7-

Subarray collections are ineffective. The switching matrix may include a combination of stages, and each combination of stages can include two or more switches. In some embodiments, a first stage of the switches can be coupled to each configuration of array elements to receive data from at least one group of subarrays and can be configured to supply data to one or more switches in a 5-second stage of the switches. The second stage of the shifters may be coupled to the first stage of the shifters to receive data from the first stage of the shifters and may be configured to supply data from a portion of at least one set of subarrays to the standard direction finding unit.

In one embodiment, the array element configuration can include a first set of array elements coupled to the switching network and a second set of array elements directly coupled 10 to the direction finding module.

In one embodiment, the direction modular unit includes a spatial SCM module for receiving data and forming one or more spatial SCMs according to the residence time and total covariance matrix using the data, a table module m for forming a table with components as a function of frequency and angle measurements, and an associated principal component module In the standard unit of SCM 15 and the standard unit of Table M. A principal components module can generate a principal components table that includes one or more principal components classified according to frequency and angle measurements. The direction finding module also includes a distance module combined with a principal component module. The distance standard unit can calculate the distance from the test point to each entry in the principal components table and the minimum distance standard unit 20 is associated with the distance standard unit. The minimum distance measurement unit can determine the minimum distance point on the basis of distances calculated from the test point to each entry in the principal components table.

It should be understood that the elements described in the various embodiments described herein may be combined to form other embodiments not specifically described above. Various elements, where 25 are described within the context of a single embodiment, may also be provided independently or in a suitable combination. Other embodiments, not specifically described herein, are also within the scope of the following claims.

8844

-8-

Brief explanation of the drawings

The above concepts and features can be fully understood from the following description of the drawings. The drawings help explain and understand the technology revealed. Because it is often impractical or impossible to illustrate and describe each possible embodiment, the figures provided depict one illustrative embodiment

<p dir="rtl">5 Or more. Accordingly, the figures are not intended to limit the scope of the concepts, systems, and techniques described herein. The numbers in the figures indicate the elements.</p>

Figure 1 represents a block diagram of a direction finding (DF) system containing arrays

sub-channels coupled to a single receiver through a switched network;

Figures 2a-2c are illustrations of different combinations of subarrays

<p dir="rtl">10 used during different residence times of the DF system according to Figure 1;</p>

Figure 3a represents a frame diagram of a DF system containing a switching network placed between...

An assortment of array elements;

Figure 3b represents a framework diagram of a DF system containing one or more of...

Subarrays directly coupled to receive channels and a variety of subarrays coupled directly to a few

<p dir="rtl">15 of receiving channels through the switched network;</p>

Figures 4a and 4b are flowcharts of how to perform trend finding using

Data from a variety of different subarrays collected at different residence times; And

Figure 5 represents a framework diagram of one embodiment of a processing system for performing DF processing according to

for the techniques described here.

<p dir="rtl">20 Detailed description:</p>

Now referring to Figure 1, the direction finding system (100)DF includes an array antenna 101 containing a configuration of subarrays 102a-102r. Each of the subarrays 102a-102r may include one or more individual antenna elements (also referred to as “elements” or “radiators”). Thus, each subarray 102a-102r may represent a single antenna element 25 or multiple antenna elements.

8844

-9-

For reasons that will be apparent from the description below, during each of the multiple residence times, the switching network (sometimes referred to herein as a switching array) 110 is coupled to selected sets of a first assortment of subarrays (herein, selected sets of subarrays 102a-102n). ) to a second configuration, lower than the receiving channels (here, receiving channels 108A-108M) 5 for receiver 107. One or more other subarrays (e.g., subarrays 102r-102r) may be coupled directly to individual receive channels (e.g., receive channel 108r) of receiver 107. Receiver 107 processes the signals provided to it as it is commonly known Receiving channels 108a-108r may, for example, amplify, convert and/or demodulate the signals supplied to them.

<p dir="rtl">10 The receiver outputs 107 are coupled to the DF processor 130. The 130 DF processor receives the signals</p>

provided to it and processes the signals to generate an estimate of the direction of arrival (i.e., angles of arrival) of an incident beam of a RF plane wave on the array antenna 101. For example, the DF 130 processor can use the minimum distance (MINDIST) technique to produce a signal DF Output Indicates the direction from which the received signal emanated. The MINDIST technology is described in 15 Joint Examination Application No. 260508/15, entitled “Systems and Methods for Direction Finding Based on Minimum Distance Search for Principal Components,” filed on the same date as this application, which is intended for the assignee of the present application, and has subsequently been Embed it here for full reference.

As shown above, in one embodiment, each of the subarrays 102a-102r may include one or more antenna elements. In some embodiments, each of the subarrays 20 may include two or more array elements. In one embodiment, each of the subarrays may comprise one element. In some embodiments, each of the subarrays may contain the same number of elements. In other embodiments, different subarrays may include a different number of elements.

A number of array elements may be chosen to include, in a given subarray, at least 25, at a total number of elements in the array antenna 101, a number of individual receive channels.

Included in the receiver 107, a desired gain pattern for the DF and/or field of view (FoV) system 100

8844

-10-

For 100 DF system. It should also be appreciated that any type of antenna element may be used to use the array antenna 101.

It should also be appreciated that the switch matrix 110 may be provided from any combination of different types of switches including multi-pole switches or multi-cylinder switches. It may depend

<p dir="rtl">5 The particular type of switch used in the switching array 110 depends at least in part on a particular implementation of the 00 DF system and components of the 100 DF system (e.g., number of array elements, number of receive channels).</p>

Furthermore, it should be appreciated that in the embodiment shown in Figure 1, some of the subarrays (i.e. subarrays 102a-102n) are coupled to the receive channels through

<p dir="rtl">10 switching array 110 while some other subarrays (e.g. subarrays 102p-102r) are coupled directly to individual channels of the receiver 107. As will be shown in conjunction with Figure 3a, in some embodiments, all subarrays may be coupled to receive channels through a switching network It should be appreciated that a certain number of subarrays can be selected to be coupled to the receive channels through a switch (e.g. switch array 110) based on</p>

<p dir="rtl">15 A particular application of the 100 DF system and/or the number of subarrays required to operate the desired system compared to the number of receiving channels 108a-108p.</p>

Referring now to Figures 2a-2c, where similar elements may be provided with their corresponding reference designations, the array antenna 200 uses certain combinations of the five subarrays 204a-204e in various combinations (or combinations) to find the direction.

<p dir="rtl">20 Estimate that in this embodiment shown, the antenna is assumed to provide 200 signals to four receiving channels. Therefore, it is not possible to simultaneously connect all five sub-arrays to a unique receiving channel. Thus, as shown in Figure 2a, during the first dwell time 202a, the first configuration (or combination) of subarrays, here four subarrays 204a-204d, is active for the receiver signals. The fifth configuration of subarrays (herein subarray 204 AH)</p>

<p dir="rtl">25 Inactive. Thus, during the first residence time 202a the received RF signals are coupled by the first configuration of subarrays (i.e. subarrays 204a-204d) through a switch</p>

8844

-11-

(e.g., switch 110 described above in Figure 1) with individual receive channels (e.g., receive channels 108a, 108b, 108c, 108d in Figure 1) so that signals from all selected subarrays 204a-204d are processed simultaneously 1 (For example, in a DF processor such as the 130 DF processor described above in Figure 1) to generate the first dwell space of a covariance matrix

<p dir="rtl">5 Sample 205 (SCM) sample covariance matrixA.</p>

Similarly, referring to Figure 2b, during the second dwell time 202b, a second different configuration (or combination) of subarrays (herein subarrays 204a, 204b, 204d, 204e) is selected to receive the signals. The fifth configuration of subarrays (herein the array sub 204c) is inactive. Therefore, during the second residence time 202b, the received RF signals are coupled

<p dir="rtl">10 By means of the second configuration of subarrays (i.e., subarrays 204a, 204b, 204d, 204e) through a switch to individual receive channels so that signals from each of the selected subarrays can be processed simultaneously to provide the second residence space 205 SCMb.</p>

Similarly, referring to Fig. 2c, during the third residence time 202c, a third different assortment (or combination) of subarrays is selected (here subarrays 204a, 204b,

<p dir="rtl">15 204C, 204H) to receive signals. The fifth configuration is of subarrays (here the array</p>

subarrays 204d) are inactive. Therefore, during the third residence time 202c, the RF signals received by the third configuration of subarrays (i.e., subarrays 204a, 204b, 204c, 204e) are coupled through a switch to individual groups of the four receive channels so that Signals from each of the subarrays are simultaneously selected to provide the third SCM

20 205 EGP.

In this arrangement, data are acquired over a variety of different residence times 202a-202c using three different assemblies (or combinations) of subarrays, i.e. a first configuration including subarrays 204a-204d; a second configuration including subarrays 204a, 204b, 204d, 204E; and a third configuration including subarrays 204A-204C and 204E.

<p dir="rtl">25 Thus, data from a given configuration is made available to subarrays at specific residence times. In a different way, during each of the first, second, and third residence times of 202a-202c,</p>

8844

-12-

The signals are from different combinations (or combinations) of subarrays 204a-204e. Thus, data are sampled from different combinations of subarrays 204a-204e at different time periods.

Although Figures 2a-2c only show 3 residence times and 3 groups of 5 subarrays, it should be appreciated that any number of subarrays and any number of residence times can be used. The number of dwell times used in a given application depends, at least in part, on the number of subarrays and/or receive channels in the DF system.

As shown in Figures 2a-2c, during each of the various residence times 202a-202c, at least one of the subarrays is inactive. Subarrays may refer to non-

<p dir="rtl">10 Active to subarrays that are configured or otherwise coupled to not receiving any signals or may indicate the presence of subarrays that are not associated with the receive channel. As will be described in conjunction with Figure 3b, this can be achieved, for example, by pairing one of the subarrays with a matching termination.</p>

Once the data is collected, the first, second and third residence space 205 SCMs can be combined a-

<p dir="rtl">15 205C in a complex way to generate the aggregate covariance matrix (AGM).</p>

The residence space of SCMs and AGM residences will be described in detail below in combination with Figures 3-4b.

Referring now to Figure 3a, the DF system 300 includes an array antenna 301 comprising a configuration of subarrays 302a-302n coupled through the switching array 310 to a configuration

<p dir="rtl">20 Secondly different from 308 RF receiver channels A-308M to 307 RF receiver. The outputs of the receiving channels 308A-308M are coupled to the inputs of the 330 DF processor. In illustrative embodiments of Figure 3A, the 330 DF processor implements MINDIST DF technology. The array 301, subarrays 302a-302n, switching matrix 310 and receiver channels 308a-308m may be identical or analogous to subarrays 102a-102r, switching matrix 110 and receiver 107 described above.</p>

25 In combination with Figure 1 and array 200 according to Figures 2A-2C.

In some embodiments, the array element array 302a-302n may be an array

8844

-13-

The switching 310 and receive channel configuration 308a-308m are separate from the 330 DF processor. In some embodiments, array element configuration 302a-302n, switching matrix 310 and receive channel configuration 308a-308m may be combined within processor 330 DF.

It should be appreciated that in this illustrative embodiment, all 5 subarrays 302a-302n are coupled to the receive channels 308a-308m through the switching array 310.

In one embodiment, the number of subarrays 302a-302n may be greater than the number of receive channels 308a-308m (i.e. n &lt; m). Thus, the switching matrix 310 can be operated to couple selected groups (groups or combinations) of subarrays 302a-302n to the receive channels Individualism 308a-308m.

<p dir="rtl">10 A set of m subarrays can be specified for use at a specific dwell time. And therefore,</p>

For each residence time, data samples from the preselected set of subarrays M are coupled through the switching matrix 310 to the receive channels M 308A-308M and then to the processor 330 DF.

The specific subarrays used in each set of subarrays from which signals are received may change from one dwell time to the next. Also, in some

<p dir="rtl">15 Embodiments, each subarray may include one or more active array elements and one or more inactive array elements. In some embodiments, to form different combinations of subarrays, the active versus inactive array elements may change from one dwell time to the next dwell times.</p>

For example, in one embodiment, during the first dwell time an array may be operated

<p dir="rtl">20 Switching 310 The subarray is coupled to a reference voltage (e.g., ground) by a matching load such that the subarray terminates, rejects or neglects signals coming from it during the first residence time. Consequently, the subarray in question may be considered inactive such that no Samples data from the subarray in question During the second dwell time, however, the array element in question itself may be considered active so that data from the array element is sampled.</p>

<p dir="rtl">25 the meaning. In one embodiment, the number of active subarrays is chosen based on the number of array elements and the number of receive channels.</p>

8844

-14-

In one embodiment, the signals received by the subarrays 302a-302n may be samples of signals incident on the array element in question at specified points in time (e.g., a snapshot at a specified dwell time). The data may be sampled at or For the phase center of one of the array elements 302a-302n.

<p dir="rtl">5 Data can be expressed as composite values (for example, Q/I data) that represent...</p>

Signal. For example, in some embodiments, the data may correspond to voltage signals represented by composite values representing angles of arrival, amplitude, phase, and/or polarization of the signal, for example. In one embodiment, the data may correspond to a signal value at a predetermined point in time (e.g., a snapshot at a specified residence time) or during a predetermined time period. Data 10 corresponding to the signal configuration may be coupled or otherwise supplied to a processor input 330 DF.

In the illustrative embodiment according to Figure 3a, the DF processor 330 includes a space SCM module 335, a major component module 340, a distance measurement module 350, a minimum distance module 355 and an S-table module 345.

The SCM module 335 receives data fed to it from the receiver 307 and generates 15 a matrix of values (i.e., SCM) using the data (which may be data samples). For example, the SCM module 335 may generate an assortment of dwell SCMs 336a-336n. Associating each of the dwelling SCMs 336a-336n to one or more receiving channels 308a-308m In some embodiments, each dwelling SCM 336a-336n may be linked to the relevant receiving channel 308a-308m and in other embodiments, each dwelling SCM 336a-336n may be associated with channels Reception 20 Multiple 308a-308m. Each residence SCM 336a-336n may be based, at least in part, on data samples taken from a selection of subarrays 302a-302n at a specified residence time (e.g., one of the residence times 202a-202c in Figures 2a-2c) and provided by Switching array 310 to receive channels 308a-308m associated with the selected set of subarrays 302a-302n at a specified residence time and examples of residence SCMs generated 25 using residence times such as those shown in Figures 2a-2c are provided below in Tables 1-3.

Table 1 below represents composite data values of the type sampled

8844

-15-

During the first residence time (e.g., residence time 202a according to Figure 2a) by

A first set of subarrays (for example, subarrays 204a-204e according to

(For Figure 2a)

Table 1

<tr><td><p>0</p></td><td><p dir="rtl">25.2114 + 5.9324</p></td><td><p dir="rtl">1306061-</p><p dir="rtl">Y20.2535</p></td><td><p dir="rtl">-25.3736-</p><p dir="rtl">Y6.0954</p></td><td><p>27.1313</p></td></tr><tr><td><p>0</p></td><td><p dir="rtl">25.8083+y0.1241</p></td><td><p dir="rtl">8.4809+Y22.7156</p></td><td><p>26.879</p></td><td><p dir="rtl">25.3734 + 6.0954</p></td></tr><tr><td><p>0</p></td><td><p dir="rtl">8.3668+Y22.5207</p></td><td><p>23.5265</p></td><td><p dir="rtl">-8.4809-</p><p dir="rtl">Y22.7156</p></td><td><p dir="rtl">13.6061+20.2535</p></td></tr><tr><td><p>0</p></td><td><p>26.5831</p></td><td><p dir="rtl">8.3668-Y22.5207</p></td><td><p dir="rtl">-25.8083-</p><p dir="rtl">Y 0.1241</p></td><td><p dir="rtl">25.2114-Y5.9324</p></td></tr><tr><td><p>0</p></td><td><p>0</p></td><td><p>0</p></td><td><p>0</p></td><td><p>0</p></td></tr>

5 It is worth noting that the data samples for the first residence time are expressed in the figure

Compound, which results in preserving both the amplitude and phase of the received signal from which the matrix values were generated. An input with a value of “zero” represents an inactive subarray (i.e., a subarray that has not been sampled and therefore not supplied to the processor 330 DF). For example, and briefly referring to Figure 2a, Table 1 may represent data from a set of subarrays 204a-10 204e with subarrays 204a-204d active and one subarray 204e inactive.

Therefore, the fifth row and fifth column of the matrix correspond to the fifth subarray 204e and hence, the values of the matrix in the fifth row are equal to R row.

Table 2 below is an illustration of composite data values of the type that were sampled during the second dwell time (for example, dwell time 202b according to Figure 2b).

15 Table 2

<tr><td><p dir="rtl">9.9436+Y24.1408</p></td><td><p dir="rtl">25.2159 + 5.9324</p></td><td><p>0</p></td><td><p dir="rtl">-25.4644-</p><p dir="rtl">Y6.0953</p></td><td><p>27.3245</p></td></tr><tr><td><p dir="rtl">4.0905-25.6354</p></td><td><p dir="rtl">25.6667 + 0.1241</p></td><td><p>0</p></td><td><p>26.9083</p></td><td><p dir="rtl">-25.4644-</p><p dir="rtl">Y6.0953</p></td></tr>

8844

-16-

<tr><td><p>0</p></td><td><p>0</p></td><td><p>0</p></td><td><p>0</p></td><td><p>0</p></td></tr><tr><td><p dir="rtl">3.9979 + 25.4416</p></td><td><p>26.348</p></td><td><p>0</p></td><td><p dir="rtl">-25.6667-</p><p dir="rtl">Y 0.1241</p></td><td><p dir="rtl">25.2159-</p><p dir="rtl">Y5.9324</p></td></tr><tr><td><p>27.0664</p></td><td><p dir="rtl">-3.9979-</p><p dir="rtl">Y25.4416</p></td><td><p>0</p></td><td><p dir="rtl">4.0905 + 25.6354</p></td><td><p dir="rtl">-9.9436-</p><p dir="rtl">Y24.1408</p></td></tr>

Data collected during the second residence time are also expressed in composite form.

Briefly referring to Figure 2b, Table 2 is an illustration of data supplied from a selection of active subarrays (i.e., subarrays 204a, 204b, 204d, 204e) and an inactive subarray (i.e., subarray 204c). Matrix

5 In the third row and column of the matrix, the inactive subarray 204C is equal to zero. Table 3 below represents composite data values captured during the dwell time (for example,

Residence time 202c according to Figure 2c). Table 3

<tr><td><p dir="rtl">9.9754+Y24.1393</p></td><td><p>0</p></td><td><p dir="rtl">13.6792-</p><p dir="rtl">Y20.2522</p></td><td><p dir="rtl">-25.4096-</p><p dir="rtl">Y6.0950</p></td><td><p>27.2336</p></td></tr><tr><td><p dir="rtl">3.9047-25.6338</p></td><td><p>0</p></td><td><p dir="rtl">8.4945 + 22.7141</p></td><td><p>26.8858</p></td><td><p dir="rtl">25.4096 + 6.0950</p></td></tr><tr><td><p dir="rtl">23.6508+Y4.7847</p></td><td><p>0</p></td><td><p>23.6675</p></td><td><p dir="rtl">-8.4945-</p><p dir="rtl">Y22.7141</p></td><td><p dir="rtl">13.6792+20.2522</p></td></tr><tr><td><p>0</p></td><td><p>0</p></td><td><p>0</p></td><td><p>0</p></td><td><p>0</p></td></tr><tr><td><p>27.0452</p></td><td><p>0</p></td><td><p dir="rtl">-23.6508-</p><p dir="rtl">Y4.7847</p></td><td><p dir="rtl">3.9047 + 25.6338</p></td><td><p dir="rtl">-9.9754-Y24.1393</p></td></tr>

In one embodiment, the data samples for the third residence time are 202c according to Figure 2c.

<p dir="rtl">10 In complex form, the signal is represented using both real and imaginary parts.</p>

Briefly referring to Figure 2c, Table 3 represents data from a set of five subarrays 204a-204e. The five subarrays (first, second, third and fifth 204a, 204b, 204c, 204e) may be active and the four subarrays 204d may be inactive. Therefore, the four row and column values of data correspond to the four inactive subarrays

8844

-17-

204d which equals zero.

In one embodiment, the SCM 335 may combine each of the SCMs' residence times to form an overall covariance matrix. Residence times of SCMs may be compositely integrated (e.g., quadratic space), so the total covariance matrix also includes composite data.

<p dir="rtl">5 For example, an example of the total covariance matrix in Table 4 is provided below.</p>

Table 4

<tr><td><p dir="rtl">9.9189+y</p><p dir="rtl">48.2801</p></td><td><p dir="rtl">50.4272+y</p><p dir="rtl">11.8648</p></td><td><p dir="rtl">27.2853-</p><p dir="rtl">Y40.5056</p></td><td><p dir="rtl">-76.2475-</p><p dir="rtl">Y18.2857</p></td><td><p>81.6894</p></td></tr><tr><td><p dir="rtl">7.9952-</p><p dir="rtl">Y51.2692</p></td><td><p dir="rtl">51.4751+y</p><p dir="rtl">0.2482</p></td><td><p dir="rtl">16.9755+y</p><p dir="rtl">45.4297</p></td><td><p>80.6731</p></td><td><p dir="rtl">76.2475+y</p><p dir="rtl">18.2857</p></td></tr><tr><td><p dir="rtl">23.6508+y</p><p dir="rtl">4.7847</p></td><td><p dir="rtl">8.3668+y</p><p dir="rtl">22.5207</p></td><td><p>47.1941</p></td><td><p dir="rtl">-16.9755-</p><p dir="rtl">Y45.4297</p></td><td><p dir="rtl">27.2853+y</p><p dir="rtl">40.5056</p></td></tr><tr><td><p dir="rtl">3.9979+y</p><p dir="rtl">25.4416</p></td><td><p>52.9312</p></td><td><p dir="rtl">8.3668-</p><p dir="rtl">Y22.5207</p></td><td><p dir="rtl">-51.4751-</p><p dir="rtl">Y 0.2482</p></td><td><p dir="rtl">50.4272-</p><p dir="rtl">Y11.8648</p></td></tr><tr><td><p>54.1116</p></td><td><p dir="rtl">-3.9979-</p><p dir="rtl">Y25.4416</p></td><td><p dir="rtl">-23.6508-</p><p dir="rtl">Y4.7847</p></td><td><p dir="rtl">7.9952+y</p><p dir="rtl">51.2692</p></td><td><p dir="rtl">-19.9189-</p><p dir="rtl">Y48.2801</p></td></tr>

In one embodiment, the total covariance matrix in Table 4 is a combination of the combined values of the residence times of the SCMs given in Tables 1-3. The SCM module 335 may generate SCM angles that specify angles for each of the SCM entries in the total covariance matrix. on

10 For example, using the real and imaginary parts of the complex number in each entry of the total covariance matrix, the SCM 335 module can calculate complex angles for each entry in the total covariance matrix. Composite angles from all inputs can be used to form SCM angles. For example, an example of SCM angles is provided below in Table 5.

Table 5

<tr><td><p>1.9621</p></td><td><p>0.2311</p></td><td><p>0.978-</p></td><td><p>2.9062-</p></td><td><p>0</p></td></tr>

8844

-18-

<tr><td><p>1.4161-</p></td><td><p>3.1368</p></td><td><p>1.9284-</p></td><td><p>0</p></td><td><p>2.9062</p></td></tr><tr><td><p>2.942</p></td><td><p>1.2151</p></td><td><p>0)</p></td><td><p>1.9284-</p></td><td><p>0.978</p></td></tr><tr><td><p>126</p></td><td><p>0</p></td><td><p>1.2151-</p></td><td><p>3.1368-</p></td><td><p>0.2311-</p></td></tr><tr><td><p>0</p></td><td><p>1261-</p></td><td><p>2.942-</p></td><td><p dir="rtl">17.9952 y 51.2692</p></td><td><p>1.9621-</p></td></tr>

In one embodiment, Table 5 represents an SCM angle generated using data from

The total covariance matrix is in Table 4 Thus, using data from the first, second and third residence times 202-202 according to Figures 2a-2c. The angles in Table 5 are represented in radians. In one embodiment, by combining data from an assortment of subarrays 5 corresponding to an assortment of array elements, an exact SCM angle that is the same or substantially similar to the resulting SCM angle can be generated using data from each assortment of elements at a single residence time.

The SCM module 335 may provide an SCM angle to an input to the principal components module 340. In one embodiment, the SCM module 335 may provide only a portion of the SCM angle input (e.g., so-called principal components as will be described in more detail 10 below) to Standard unit of main components 340.

The principal components module 340 may extract the so-called principal components from the SCM angle. In one embodiment, the principal components may correspond to SCM inputs that have a non-zero angle. A non-zero angle indicates an angle that is greater than or less than zero (and therefore not equal to zero). For example, an SCM angle may have an SCM input that compares 15 the data captured in the array elements to itself (e.g., no=44). Therefore, the angle involved for these inputs may equal zero.

Referring to Table 5 above, the values along the main diagonal of the SCM angle represent the difference between two samples of data taken at the same array element. Therefore, the main diameter values are ideally equal to zero. In some embodiments, SCM inputs that have an angle 20 equal to zero may be removed from further processing or ignored. Thus, SCM inputs having a non-zero angle can be extracted from the SCM.

8844

-19-

In some embodiments, a principal component module 340 may extract only a portion of the non-zero angle SCM input. For example, the SCM angle may include an SCM input that compares two different array elements (i.e., array element m and array element n) to form an SCM input. The first ΔΦ mn and the second SCM input ΦηηιΔ. In this case, it is marked that the first 5 SCM input will be equal in amplitude to the second SCM input (for example, -4411=1()4(). For example, referring to Table 5 above, the value in the first column, second row may match the value in the second column, first row. Thus, only one of these two values, one of the first and second SCM inputs, may be needed for processing.

In one embodiment, the principal component module 340 extracts half of the 10-angle SCM input having a non-zero angle. The modular unit of principal components 340 Table of non-zero SCM inputs may have the extracted angle input SCM having a non-zero angle.

Table M Module 345 may be one or more tables containing the underlying component values (table values M). M table values can be stored and indexed, for example, as a function of angle 15 and frequency. In one embodiment, the angle corresponds to the angle of arrival of a received signal incident on one or more array element combinations 302a-302n. In some embodiments, the angles can be sorted by two-dimensional angles (e.g., azimuth angle, elevation angle).

In an embodiment, the table M can be provided as a precalculated table (e.g., a table with precalculated values stored in it) or a principal components table (e.g., a table 20 containing principal components based on the measured data). A pre-calculated AD table may be generated using the same methods described above, however, a pre-calculated AD table may be generated using previous estimated measurements and/or values. For example, the table m may be precomputed using snapshots previously assembled into one or more array element combinations 302-302n. Data from previous shots may be used to form a pre-calculated SCM angle).

<p dir="rtl">25 Using the pre-calculated SCM angle, the main components can be identified and extracted</p>

8844

-20-

10

SCM. Thus, a previously computed table may be constructed using principal components that were extracted based on data from previous snapshots.

In another embodiment, the table M may be precomputed using estimated array properties (e.g., properties of an array manifold vector). For example, the estimated array properties may be used to form estimated snapshots of one or more array element combinations 302a-302n. It may be The pre-calculated M table is generated using methods as described above based on estimated data (e.g., snapshots) for array elements 302-302n.

In an embodiment, the frequency and angle components can be pre-calculated based on estimated values or previously collected data. For example, in some embodiments, the frequency and angular components may be precomputed using previously collected footage. In other embodiments, the frequency and angle components may be pre-calculated using estimated values performed based on the known phase centers of one or more of the array configuration 302-302n. In some embodiments, using estimated or measured phase center locations, the frequency and angular components may be precomputed using an array manifold vector corresponding to the array element configuration 302a-302n.

<p dir="rtl">15 For example, for the desired range, a table module might perform an analysis</p>

A statistic to measure the amount of energy or information contained in the data stored in a table. This analysis can be performed for all angles within the desired range with respect to a pre-determined azimuth range and a pre-determined elevation range. In an embodiment, principal component analysis may be used to measure the amount of energy and/or information contained in multivariate data, here being the desired range for

<p dir="rtl">20 To the pre-set azimuth range and pre-set elevation range. For example, principal component analysis of each of the array elements over a desired frequency range may be performed using an array manifold vector compatible with the array element configuration 302a-302n. Thus, the standard unit of the table 345 may be vector data (e.g., principal component data) for each of the matrix element combinations 302a-302n. In some embodiments, the vector data may be</p>

<p dir="rtl">25 Component (estimated) based on the phase center positions of the matrix element configuration 302a-302n.</p>

8844

-21-

In some embodiments, the m-table module 345 multiplies the vector data generated by each array element by its own composite conjugate and/or the composite conjugate of vector data from another array element and stores the result in a vector data table.

In an embodiment, the M345 module compares the vector data table with a 5 SCM phase difference matrix to determine the corresponding data vector with SCM angle inputs having a non-zero angle and extract the corresponding vector data. The modular unit of the m-table is 345 final m-table that sorts all SCM entries that have a non-zero angle by their respective vector data (for example, by frequency and angle data). In an embodiment, the final AD table may be a primary component table. Thus, the final AD table may include the extracted principal component data sorted, grouped 10 or ordered by frequency and angle.

In some embodiments, the modular unit of the M table 345 may be one or more M tables before the 11-330 DF processor receives signals or data from the matrix element combination 302-302n. In other embodiments, the M-table module 345 can be one or more M-tables simultaneously (e.g., in real time) receiving the DF processor 330 15 signals or data from the array element configuration 302a-302n.

In some embodiments, the number of M-tables configured and/or the size or number of elements in the M-table may vary depending on a particular implementation of the MINDIST method. For example, in some embodiments, the M-table may be configured for the desired angular field of view (e.g., 4 +,22 4+). Thus, the number of elements in table m is at least related to the desired angular field of view.

<p dir="rtl">20 The distance standard unit 350 receives principal component values from the principal component standard unit 340 and receives a table m supplied by the table standard unit m 0345 The distance standard unit 350 calculates the distance between the test point and each of the entries in the table m. In an embodiment, the test point may refer to a point Data measured in real time. For example, a test point can be formed by extracting the principal components as composite phases from</p>

<p dir="rtl">25 The total or angle covariance matrix SCM formed by the modular unit of 335 SCM. Thus, the test point may correspond to data currently received from array elements 302a-302n.</p>

8844

-22-

In an embodiment, a distance between the test point (or real-time data point) of each entry in table M (e.g., a precomputed table M) can be specified to determine a minimum value (e.g., the closest entry) in table M to the test point that Collected In some embodiments, multiple test points may be used for example, a distance may be specified for each configuration

<p dir="rtl">5 of test points, from the test point in question to each entry in table M. Distance is the metric you are trying to improve.</p>

In some embodiments, a distance standard unit 350 may be selected to perform the calculation. For example, without limitation, a distance standard unit 350 may be used as a Mahalanbois distance or a standardized Euclidean distance to perform the calculation. In 10 some embodiments, the Mahalanbois distance or Euclidean distance may be used to calculate the distance between each entry in the non-zero SCM (e.g., test points) and a corresponding entry in Table M. The calculation is described in more detail below in relation to Figures 4a-4b. In the embodiment where the Euclidean distance is used, an inverse covariance matrix can be precomputed and applied to the table entries m in order to reduce the computation time of the MINDIST method.

<p dir="rtl">15 The minimum distance standard unit 355 may receive distances calculated from a standard unit of measurement</p>

Distance 350. In an embodiment, the minimum distance standard unit may specify a value from the calculated distances between test points (or multiple test points) and each entry in table M that represents a minimum distance compared to other calculated distance values. The value may be a minimum distance point that represents the angles of arrival of a signal Assortment of matrix elements 302a-302n.

<p dir="rtl">20 In an embodiment, the minimum distance standard unit 355 may output a signal indicating the distance point</p>

Minimum, such as 360 DF output signal. The minimum distance may refer to the minimum value (e.g., the closest entry) in Table M to the collected test point. In some embodiments, the minimum distance point may correspond to the estimated angle of arrival of the incident signal on one or more array elements 302a-302n Thus, the DF processor 330 may produce an output signal 360 from the estimated arrival angles 25 of a signal incident on one or more array elements 302a-302n.

8844

-23-

In an embodiment, the 330 DF processor may be identical or substantially similar to the 130 DF module described above in connection with Figure 1.

Referring to Figure 3b, an illustrative DF system 300´, which may be similar to the DF systems 100, 300 described above in conjunction with Figures 1 and 3a, includes an array antenna 301´ 5 containing five subarrays 302a´-302e´ and a receiver 307 It has four reception channels 308A-308D. In this illustrative embodiment, switching array 310´ selectively couples subarrays 302a´-302c´ to receive channels 308a´, 308b´ while subarrays 302d´ and 302e´ are directly coupled to receive channels 302 DF´, 308d´, respectively. Therefore, not all subarrays 302a´- 302e´ are associated with a permutation array 310.

<p dir="rtl">10 The number of subarrays associated with receive channels 308 RFa´-308d´ can vary from</p>

through the switching matrix 310 versus being directly coupled to the receive channels 308a´-308d´ (and then to the 330 DF processor) depending at least in part on a variety of factors, including but not limited to, the particular application of the 100 DF system and/or A number of subarrays in the matrix 301´ compared to a number of channels in the receiver 307´.

<p dir="rtl">15 In this illustrative embodiment of Figure 3B, the 300DF´ system includes five subarrays</p>

It consists of antenna elements that form a matrix 301´ and the receiver 307´ has only four channels. Three subarrays 302a, 302b, and 302c, two receive channels 308a and 308b are coupled to a switched network 310, and two subarrays 302d and 302e are coupled directly to receive channels 308c and 308d. In other embodiments, however, it may be desirable or necessary 20 for other combinations of subarrays and/or receive channels to be coupled through a switched network.

In an alternative embodiment, however, it may be desirable or necessary for four subarrays and three receive channels to be coupled to a switching array 310´ and one subarray directly coupled to a single RF receive channel. Alternatively in other embodiments, two of the arrays 25 can be selectively coupled to a single receive channel through a switching array 310´ and three subarrays can be directly coupled to the individual receive channels. Other embodiments are also possible with systems thereof

8844

-24-

Different number of sub-arrays and receive channels, (i.e. other than five sub-arrays and four receive channels).

In the illustrative embodiment of Figure 3b, the switching array 310´ includes two switching stages with a first stage including switches 304a-304c coupled to a second stage including switches 306a-306b. maybe

<p dir="rtl">5 Providing switches as multi-pole switches, multi-crank switches and/or multi-pole switches. For example, as shown in Figure 3B, switches 304a-304c are provided in the first stage as single-pole, three-crank switches having one input coupled to one of the respective subarrays 302a´-302c´ and three outputs. Switches 306A and 306B are also supplied as single-pole three-crank switches. Three inputs of switches 306a and 306b are coupled to the outputs</p>

<p dir="rtl">10 For the switches 304a-304c, the outputs of the second stage switches are coupled to the outputs of the special receiving channels 308a´-308b´.</p>

The switches 304a-304c, 306a and 306b are configured to couple signals from the selected subarrays 302a-302c to the receiving channels 308a-308b. Because there are only two receive channels associated with the output of the switch 310´ (i.e., channels 308a´ and 308b´), the

<p dir="rtl">15 The switching array 310´ couples two of the three subarrays to the two receiver channels (i.e., two subarrays are active) and the third subarray is inactive. In this illustrative embodiment, the subarrays 302a´-302b´ are coupled to the designated channels of the receiver channels 308a´- 308b´ The sub-array 308c´ becomes inactive by coupling it to a reference voltage source (ground) through the matched termination 305.</p>

<p dir="rtl">20 Thus, with such a switch configuration, a first set of subarrays can be coupled,</p>

There are four subarrays 302A´, 302B´, 302C´, 302D´, 302E´, with four receiver channels 308A´-308C´ during residence time. During the next residence time, the switches may be reconfigured so that a second set of subarrays (e.g., subarrays 302a, 302b, 302c, 302d, 302e) are coupled to four receive channels 308a-308d. This can be repeated the operation

<p dir="rtl">25 For as many different residence times as there are different groups (or combinations) of subarrays.</p>

8844

-25-

Although Figure 3B shows the first-stage switches as three-arm switches having one input and three outputs and the second-stage switches as having three inputs and one output, it should be realized that a switching matrix 310´ can be designed using any variety of different types of switches. Multi-pole or multi-crank commutators. The type of switch used may depend, at least in part, on the needs of a particular application and the components of the 300 DF´ system (e.g., number of array elements, number of receive channels). Thus, in some applications, it may be possible or desirable to provide a switching matrix of a single, multi-pole, multi-crank switch. For example, switching array 310´ may be provided as a three-crank switch (rather than being provided as a single-pole, three-crank switch configuration as shown in Figure 3B).

<p dir="rtl">10 Figures 4 and 4a are a series of process flow diagrams illustrating processing</p>

Illustrations that may be implemented in a DF system such as the 100 DF, 300, or 300´ system described above in conjunction with Figures 1, 3a, and 3b) and, more specifically, in a DF processor such as the 130 DF or 330 processors described above in conjunction Refer to the systems illustrative of Figures 1, 3a and 3b. In the description of this treatment, reference may be made to any of the 130DF 15 or 330 processors but it should be understood that a reference to one of them does not limit the invention to that particular element. Rectangular elements (labeled as element 402 in Figure 4a), are referred to here as “processing frames,” and represent computer program instructions or sets of instructions. Diamond-shaped elements (labeled as element 408 in Figure 4a) are Here they are defined as “decision frames”, representing computer program instructions, or sets of instructions, which affect the execution of computer programs.

<p dir="rtl">20 Computer program instructions represented by processing frameworks. Alternatively, processing and decision frameworks may represent steps or operations performed by functionally equivalent circuits such as a digital signal processor circuit or an application specific integrated circuit (ASIC). Process flow charts do not depict the syntax of any particular programming language. Moreover, process flow charts show the functional information that a person of ordinary technical skill would require to manufacture circuits or 25 Configure Computer Software to perform the processing required for a particular device. It should be noted that many routine program elements, such as initializing loops and variables and using temporary variables, are not shown.</p>

8844

-26-

It will be understood by those of ordinary skill in the technique that unless otherwise stated, the specific sequence of frameworks described is illustrative only and may be changed without departing from the spirit of the concepts, structures and techniques described. Thus, unless otherwise stated the frameworks described below are unordered meaning that, where possible, the functions represented by the frameworks may be implemented in any appropriate order or

<p dir="rtl">5 desired.</p>

Referring now to Figures 4a and 4b, an illustrative method 400 of performing direct identification using an assortment of spatial SCMs according to residence time and a total covariance matrix begins within processing 402, where a first set of subarrays can be selected from the assortment of sets

10

15

20

25

Subarrays. RF signals can be received by a selection of subarrays from one or a variety of different RF sources. For example, as noted above, in some embodiments the RF signals may be identical to a type of emergency beacon signal used in a variety of different applications, including but not limited to airborne or ground search and rescue applications. the site.

Within processing 404, during a first residence time, data is collected across the first selected set of subarrays. That is, the signals received by the first group of subarrays are simultaneously coupled to an equal number of receive channels (i.e. one receive channel for each effective subarray) that process the signals and provide data from each subarray to the DF processor (for example a 330 DF processor In Figure 3b(.

Data from the received signal can be captured across a subarray at an instantaneous point in time (such as a snapshot during each dwell time) or over a pre-determined time period. Data is captured at a different set of dwell times for each set of subarrays. Thus, each set Subarrays capture data at different residence times (e.g., different snapshots). In some embodiments, the data may be captured at predetermined residence times.

Processing occurs to the processing frame 406 where the data may be represented as composite values (e.g., I/Q in-phase/quadrature data) stored in the SCM space according to the dwell time of the first set of subarrays. This may be Composite values are represented

8844

-01-

For some or all of the angle of arrival, amplitude, phase, and/or polarization of the received signal. Data may be stored, processed or analyzed in a more complex manner.

In some embodiments, multiple samples may be taken from a single signal. In other embodiments, multiple samples may be taken from different signal combinations. In one embodiment, the data may correspond to

<p dir="rtl">5 Different signals. It should be realized that the number of data samples taken may vary depending on a variety of factors, including but not limited to, the number of array elements, the number of subarrays used and the requirements of a particular application. In some embodiments, the number of data samples taken may be based, at least in part, on the signal-to-noise ratio of the received RF signal. For example, for RF signals with a signal ratio</p>

10 For low noise, it may be desirable to collect more data samples to increase the accuracy of the technique

DF processing compared to processing signals with a high signal-to-noise ratio.

In frame 406, the SCM has a dwell-time space using signals received during the first dwell time from the first set of subarrays. The dwell time of the SCM may be configured, for example, by a DF processor which may be the same or similar to the 330 DF processor.

<p dir="rtl">15 Which is described in conjunction with Figures 3a and b. In one embodiment, the dwell time of the dwell SCMs can be configured by a modular SCM module 335 for a DF processor 330. The residence time of the given SCMs may include data sampled at the antenna of the phase centers of the array elements in the subarray group of interest during the relevant residence time. Data can be compared with other data from a given subarray set. In an embodiment, the size of the residence time of the SCM can be proportional to a number</p>

20 Total of array subgroups, thus including entries for both active array elements and passive array elements. However, no data is received from the subarrays

Ineffective and therefore a zero value is entered in the relevant array entry for the inactive subarrays. A sample residence time for SCM is provided below:

<tr><td></td><td><p>^11</p></td><td><p>512</p></td><td><p>513 ...</p></td><td><p dir="rtl">5٩٤٦</p></td></tr><tr><td></td><td></td><td><p>522</p></td><td><p>523 ...</p></td><td><p>5 23</p></td></tr><tr><td><p>59 =</p></td><td><p>531</p></td><td><p>532</p></td><td><p>533 ...</p></td><td><p>S3N</p></td></tr><tr><td></td><td><p dir="rtl">1 7,j</p></td><td><p>=12</p></td><td><p>SN3...</p></td><td><p>s:n.</p></td></tr>

8844

-28-

Where Sxy represents a comparison of sample data taken in subarray X to sample data taken in subarray No. In an embodiment, the sample residence time for the SCM may be the same or substantially similar to Tables 1-3 described above in conjunction with Figures. 3a-3b. The SCM residence time sample values for a matrix represent composite values (for example, amplitude and phase may be expressed as 5 41283). In the SCM dwell time above, the values in the first column of the SCM dwell time represent the difference between a sample of data taken at the first subarray to samples taken at each of the other subarrays. The values along the main diagonal of the SCM residence time represent the difference between two data samples taken at the same subarray. Therefore, the main diameter values are ideally equal to zero. The SCM dwell time column (p) contains values of 10 corresponding to the difference between the sample data taken at p subarray and the sample data taken at each of the other subarrays in the subarray group. Therefore, it should be realized that the dwell time of an SCM typically includes a number of rows and/or columns corresponding to the number of rows and/or columns in a subarray array. In some embodiments, the number of rows and/or columns in the dwell time of the SCM can be based at least on the total number of subarrays in the 15 subarray array (i.e., both active array elements and passive array elements).

Under Resolution 408, a decision is made as to whether to collect data through additional sets of subarrays. In an embodiment, the number of subarrays used in a particular application can vary and can depend, at least in part, on the number of total elements in the array antenna, a number of receive channels, a desired gain pattern of the DF system and/or the field of view of the DF system 20 as As mentioned above, each different set of subarrays receives signals during a different dwell time. Thus, the decision framework 408 and processing frameworks 410 , 412 implement a loop during which data is collected and an SCM is formed for each subplate group during different residence times.

In an embodiment, method 400 may include a feedback mechanism for continuously collecting data from different sets of subarrays until a desired number of 25 sets of subarrays have been sampled. For example, if the answer to decision frame 408 is yes, the data is collected into additional subarrays and processing continues to the processing frame

8844

-29-

410, to select the next different set of subarrays from the assortment of subarrays into which the data is collected.

Then, within processing 412, during a subsequent dwell time, data may be collected by a new selection of subarrays. Processing occurring within 5 Processing 412 may be identical or substantially similar to processing occurring within Processing 404, however the data collected will be from a different set of subarrays at a different residence time than within Processing Frame 404.

Once the data is collected in processing frame 412, processing continues again to frame 406 to configure the spatial SCM according to the residence time of the most recently selected set of 10 subarrays. Thus, the method 400 can be repeated continuously over frames 406, 408, 410, 412 until the desired number of subarray sets have been sampled. Once the predetermined number of subarray sets has been sampled, the response to decision frame 408

It does not and continues processing to processing frame 414.

In processing frame 414, the residence time of the corresponding SCM for each is synthesized

<p dir="rtl">15 Subarrays group to form a total covariance matrix. In an embodiment, the SCM module 335 of the processor 330 DF may combine each of the dwell times of the SCMs to form a total covariance matrix. The residence time of the SCMs may be combined in composite form (e.g., quadratic space) to form the total covariance matrix (i.e., the composite values of the residence time of the SCMs are combined in composite form to form the total covariance matrix). The total covariance matrix may be the same as or</p>

<p dir="rtl">20 Substantially similar to the total covariance matrix described above in Table 4 in conjunction with Figures 3a-3b.</p>

In one embodiment, the MINDIST technique is applied to data stored in a total covariance matrix. The MINDIST technique is described in detail in pending application No. 260508/15, filed at a later date, entitled “Systems and methods for finding direction based on...

<p dir="rtl">25 Find Minimum Distance to Principal Components", which the order is assigned to the assignee</p>

8844

-30-

About the current demand. In an embodiment, the MINDIST technology may include processing described in Processing Frameworks416-424.

Under processing 416, according to the MUST technique, the SCM angle may be formed using the total covariance matrix. The SCM angle can be formed based on the phase values of 5 elements in the total covariance matrix. In an embodiment, the SCM angle can be configured by a 130 DF modular unit of Figure 1 or a 330 DF processor of Figures 3a-3b. For example, in one embodiment, the SCM corner may be formed by a modular module of 335 SCM for a 330DF processor. The SCM angle may be the same or substantially similar to the SCM angle in Table 5 described above in relation to Figures 3a-3b.

<p dir="rtl">10 In the embodiment, the angle SCM includes measuring the angles of each entry in the total covariance matrix. The SCM angle may represent the angular measure of the combination of array elements in each of the sampled array subgroups. Goniometry may be used to determine or identify any phase difference between each of the entries in the total covariance matrix or angle SCM. In an embodiment, the value of each entry in the angle SCM may represent a comparison between data samples taken in</p>

<p dir="rtl">15 The phase centers of each of the array elements used in the subarray combination. In some embodiments, the input angle value represents a phase difference between array elements represented by the combination of array subgroups. For example, the input corresponding to a sample comparison between a first and a second array element contains an angle corresponding to the phase difference between the first and second array element. For example, an embodiment of the SCM angle is provided below:</p><table border="1"><tbody><tr><td><p>-5+1 •</p></td><td><p>ΔΦ21</p><p>ΔΦ31</p></td><td><p>-Δ^12</p><p>ΔΦ31</p></td><td><p>-Δ^23</p></td><td><p dir="rtl">٠٠</p><p>- -1</p><p>- —ΔΦ3Ν</p></td><td></td></tr><tr><td></td><td></td><td><p>+:2</p></td><td><p>ΔΦΝ3</p></td><td><p dir="rtl">0 ٠</p></td><td><p>2()</p></td></tr></tbody></table>

Where 52 represents an angle determined by comparing the phase measurement in array element X with a phase measurement in array element y. Thus, ΔΦ21, corresponds to the phase difference between the first element and the second element. In one embodiment, the SCM angle in Table 7 may be the same or substantially similar to the SCM angle in Table 5 described above with reference to Figures 3-3.

8844

-31-

However with reference to frame 416, in one embodiment, each non-zero element of the SCM angle may be determined. In one embodiment, each non-zero element of the SCM angle may be determined by the modulator 130 DF of Figure 1 or the processor 330 DF of Figures 3a-3b. For example, in one embodiment, each non-zero component of the SCM angle can be identified by

<p dir="rtl">5 335 SCM standard module of the 330 DF processor.</p>

Referring to the SCM angle above, it should be noted that the values along the main array correspond diagonally to the phase difference taken between the phases of one subarray and are therefore zero.

It should also be noted that the angle measure between the first and second array element may 10 be included twice in the table (for example, in the first column, ΔΦ21, and in the second column, ΔΦ21,

ΑΦ12-)0 i.e. 400-=00. Since ΔΦιηη=-ΔΦηπι, there is some redundancy in some phase differences and thus only one of these values may be needed for calculations. Therefore, in some embodiments, it is necessary to extract only a portion of the non-zero input from the SCM angle. Thus, a reduced number of elements to be analyzed can be realized to reduce the overall calculation duration of the technique.

<p dir="rtl">15 MINDIST by extracting the non-zero input diagonal of the SCM angle.</p>

In frame 418, a principal components table (Table M) may be calculated using the angle and frequency measurements of one or more principal components. In some embodiments, the M table may be pre-calculated. In one embodiment, the table M may be generated by the unit 130 DF of Figure 1 or the processor 330 DF of Figures 3a-3b. For example, in one embodiment, it may

20 The AD table is generated by the AD table module 435 of the DF 330 processor.

In one embodiment, the frequency and angle components of table M can be pre-calculated in tables and can be extracted from table M. For example, the frequency and angle components can be calculated in advance based on previously collected data and/or using estimated data. In one embodiment, the estimated data may be based on a known phase center location of a suitable array element or data

<p dir="rtl">25 Measured in laboratory settings using, for example, a computer model of the array element (or array antenna). In some embodiments, the frequency and angle components may be calculated before one or more</p>

8844

-32-

Of the array elements that receive a signal or before a sample of data is captured in the array elements in question. Thus, the frequency and angle components may be extracted from tables that were previously calculated during the execution of the MINDIST technique. In one embodiment, the tables may be searched in tables (or indexes) and the corresponding vector data may be extracted after identifying the principal components.

<p dir="rtl">5 In other embodiments, the frequency and angle components may be computed simultaneously or closely together such that one or more array elements receive signals or samples of data captured in the one or more array elements.</p>

In one embodiment, the frequency and angle components of a desired range may be calculated. For example, a table can be created by performing a principal components analysis of all angles within the desired range of 10 for a predefined azimuth and elevation range. For example, referring to Equation 1 below:

2,=-12=,3-(6,9 Equation R

As shown in Equation 1, using a range defined by the azimuth range [22] and elevation range [22a], the frequency and angle components can be calculated using an array manifold vector. In one 15 embodiments, an array manifold vector corresponds to the array elements being parsed. Therefore, an array manifold vector (0,8,9*) can be represented by Equation 2 below:

where

Speed of light Speed of light Cght

Element position =ρ 20 vector transpose operator =τ

Element Index 2

U = vector toward the normalized line of sight as a function of (0, φ)

Azimuth angle 0 = azimuth angle

25 *= Elevation angle

8844

-33-

The frequency and angle components can be calculated for each array element over a frequency range

Desired using an array manifold vector. In one embodiment, vector data can be generated for each element of the array. In one embodiment, the vector data may be generated by the 130 DF module of Figure 1 or the 330 DF processor of Figures 3a-3b. For example, in one embodiment, the vector data may be generated by a table module of the processor DF (330.

In one embodiment, a vector data table can be generated by comparing vector data from each element of the array. As an example, one embodiment of a vector spreadsheet is shown below:

Γ fl aaa dla

I, no, no, no, no

<p>^٠Ala = Lalak -]</p>

10

Where T represents the vector data for the array element * and T represents the complex conjugate of the vector data for the array element In one embodiment, to create a vector spreadsheet, the resulting vector data for each array element may be multiplied by its own composite conjugate and/or the composite conjugate of vector data from another array element.

<p dir="rtl">15 In some embodiments, the vector data table can be precomputed and thus used as a lookup table to pull certain entries corresponding to the extracted principal component. In one embodiment, the SCM angle (Table 7) can be used to select appropriate entries in the vector data table (Table 8) to be extracted. For example, the non-zero elements in the SCM corner correspond to the extracted principal components and, in some embodiments, the entries in table data can be extracted</p>

<p dir="rtl">20 The corresponding vector of principal components is extracted to create a table m. Thus, the number of entries in table m may match the number of extracted principal components (e.g., one entry for each extracted principal component) and the entries extracted from the vector spreadsheet.</p>

Table M may sort each SCM element with a nonzero angle by its vector data (for example, by frequency and angle data). Therefore, it may include

8844

-34-

The final AD table contains the extracted principal component data sorted by frequency and angle. And is provided

One embodiment of Table M is in Table 9 below:

Where (,8,/) represents the vector data of the SCM element compared to the array element »

<p dir="rtl">5 With array element y. For example, 3=33 represents the angle value of the principal component of an SCM element comparing array element to array element 7, [f] represents frequency, e represents a first angle value (for example, azimuth angle) and φ represents a second angle value (for example, elevation angle ). M represents the number of entries in the table M, which may refer to the number of principal components extracted and the entries extracted from the vector spreadsheet, as shown above. Table M may include all data</p>

<p dir="rtl">10 Vector for each of the extracted principal components.</p>

In some embodiments, the number and/or size (i.e., number of entries in Table M) of Tables M that are generated may vary according to a particular implementation of the MINDIST technology. For example, in some embodiments, Table M may be generated for the desired angular field of view ( For example, (1,2 4). Therefore, the number of entries in table M may vary according to the angular field of view

<p dir="rtl">15 Required. Furthermore, the number of dislocations in table M may correspond to the number of array elements in an antenna array and/or the number of principal components extracted from the SCM 0. For example, in some embodiments, for an array of array elements K, tem2* may be extracted from the SCM , where * represents the number of principal components to be used (i.e., principal components to be extracted) for a given application of the MINDIST technology.</p>

<p dir="rtl">20 In some embodiments, the table m can be divided into different subspaces (0, φ)p. Each of the subspaces may be processed independently of each other, or two or more different (0, 0) subspaces may be processed together. In one embodiment, each of the different subspaces (0, 0) p can be processed on different systems (e.g., different processors p), and can be processed</p>

8844

-35-

At the same time or at a very close time. Therefore, the computation time can be reduced by 1/p. The results from each system, each of the different processors, can be compared to determine the minimum distance point.

In some embodiments, one or more tables (e.g., a table of non-zero SCM elements, a vector data table) may be calculated before one or more array elements receiving a signal and/or a sample of data being taken from a signal received at one or More than array elements For example, tables may be precomputed using previously collected data for an array and/or estimated data for an array (for example, estimated snapshots at one or more array elements). In other embodiments, tables (e.g., non-zero SCM element table, vector data table) can be created simultaneously for one or more array elements that receive a signal and/or a sample of 10 data is taken from a signal received at one or more array elements.

In frame 420, the principal components can be extracted from the total covariance matrix or the SCM to form a test point. In one embodiment, the main components may be extracted by the standard unit 130 DF of Figure 1 or the processor 330 DF of Figures 3a-3b. For example, in one embodiment, the principal components may be extracted by the principal component module 15 340 of the processor 330 DF.

In an embodiment, the principal components in Table M may correspond to desired values from the total covariance matrix or the SCM. The main components in Table M can be sorted according to their respective frequency, angle and/or phase components. Thus, the corresponding principal components in the total or angular covariance matrix of the SCM can be identified using their respective angle and frequency measurements.

<p dir="rtl">20 In one embodiment, the principal components may correspond to the inputs in the SCM by angles</p>

Non-zero. For example, referring to Table 7, phase difference measurements that are greater than or less than zero (i.e., not equal to zero) can be identified in the SCM angles. Phase difference measurements that are greater than or less than zero can be extracted from the SCM angles It represents the main components.

In one embodiment, half of the non-zero inputs can be extracted from the SCM angles.

<p dir="rtl">25 In other embodiments, the number of principal components extracted may vary depending on a variety of factors including, but not limited to, the frequency of the signal(s) being</p>

8844

-36-

Analyze it. For example, in one embodiment, as the frequency increases, the variance (for example, power) may be distributed across the principal components. In some embodiments, the frequency signal may have a smaller wavelength and therefore more information may be contained in more centers. Antenna phase Alternatively, in other embodiments, the low-frequency signal may have longer wavelengths

<p dir="rtl">5 There is less need to focus on developing the MINDIST technique. Therefore, more principal components may be needed to calculate the minimum distance point at a high frequency rather than a low frequency.</p>

In frame 422, a distance can be specified between the test point and each value in the principal components table. In one embodiment, the distance measurement may be specified by the 130 DF module of Figure 1 or the 330 DF processor of Figures 3a-3b. For example, in one embodiment,

<p dir="rtl">10 The distance measurement can be specified through the standard distance measurement unit 350 of the DF 330 processor.</p>

The test point may refer to collected data (e.g. data snapshot) that was taken while the DF system was in operation (e.g. radar system, antenna system). The test point may be chosen from real-time data collected from snapshots at One or more of the array elements 102a-102n of Fig. 1, one or more of the array elements

<p dir="rtl">15 302a-302n of Fig. 3a and/or 302a`-302e` of Fig. 3b. For example, it can</p>

Forming the test point by extracting the principal components as complex phases from the total or nodal covariance matrix (SCM) generated by the standard 335 SCM module. Thus, the test point may correspond to data currently received from array elements 302a-302n.

In some embodiments, multiple test points may be used. Data can be measured from

<p dir="rtl">20 The test point against the values in Table M to determine the minimum distance point (or the closest entry in Table M to the combined test point). In some embodiments, multiple test points can be used. For example, in one embodiment, a distance between each test point can be calculated For each entry in Table M, in one embodiment, a test point may be randomly selected from the set of snapshots (e.g., real-time data points) collected during operation.</p>

<p dir="rtl">25 In other embodiments, the test point may be predetermined.</p>

8844

-31-

<img file="SA8844B1_D0001.tif" />

In one embodiment, for each of the entries (values) in the tables m, the distance from the respective entry to the test point can be calculated. In some embodiments, a distance measure may be chosen to perform the calculation. For example, but not limited to, the Mahalanbois distance (Equation 3 below) or Euclidean distance (Equation 4 below) to perform the calculation.

5 (55-2(2 - : 32) Equation 3

= -0

Equation 4

Where = represents the entry data of a table M with rank F and T represents the data of the test point. In an embodiment where the Euclidean distance is used, an inverse covariance matrix can be formed from the data points in table M for each principal component. The inverse covariance matrix can be applied to measure 10 distance to calculate the Euclidean distance between the test point and each of the entries in the M table. In one embodiment, the inverse covariance matrix can be pre-computed and applied to table M to reduce the computation time of the MINDIST technique.

At frame 424, the minimum distance point corresponding to the direction of the received signal can be determined. In one embodiment, the minimum distance point may be determined by the 130 DF module of Figure 1 15 or the 330 DF processor of Figures 3-3. For example, in one embodiment, the minimum distance point may be specified by a minimum distance standard unit 355 from a 330 DF processor.

In one embodiment, each of the measured distances for each of the entries in the M table can be compared to determine the minimum distance point. For example, in one embodiment, Equation 5, as shown below, may be used to solve the minimum distance point.

20 (8,9)0/,) 2=(8,2) 5

The point with the minimum distance to the test point can be determined, compared to the other entries in Table M. The value may be the minimum distance point representing the angle of arrival of a signal on one or more array elements.

8844

-38-

In some embodiments, an estimate of the expected minimum point for a particular frequency may be generated. For example, in some embodiments, a predicted minimum point can be estimated using data and/or pre-collected estimated values for array properties. All entries in Table M can be compared to the estimated values to determine which entry in Table M is the closest estimate of the frequency.

<p dir="rtl">5 In one embodiment, an output signal indicating the minimum distance point may be generated, such as a signal</p>

DF came out. In one embodiment, the output signal may be generated by a 130 DF module of Figure 1 or a 330 DF processor of Figures 3a-3b. The DF output signal may be an estimated angle of arrival of an incident signal on one or more array elements. In some embodiments, the estimated arrival angles of the signal may be a two-dimensional estimate.

<p dir="rtl">10 Referring now to Figure 5, computer 500 includes a processor 502, volatile memory</p>

504 volatile memory 506 non-volatile memory (e.g., hard disk), graphical user interface 508 (GUI) graphical user interface (e.g., mouse cursor, keyboard, display, e.g. Example) and a computer disk 520 computer disk. The non-volatile memory stores 506 computer instructions

<p dir="rtl">15 512 computer instructions, 516 operating system and data 518 data. And in</p>

In one embodiment, the data 518 may include aggregated data corresponding to signals received at one or more array elements. The data may include complex Q/I data representing the signal. For example, in some embodiments, the data may include complex voltage signals representing the angle, amplitude, phase, and/or polarity of the signal. In one embodiment, the data may include:

<p dir="rtl">20 Measuring the angles of a signal relative to the phase center of the array element in question that receives the signal. In one embodiment, the data may be a snapshot of the signal at a predetermined time interval or over a predetermined time period.</p>

In some embodiments, the non-volatile memory 506 includes a lookup table that stores and organizes data corresponding to the collected data, as well as any tables (e.g., 25 m tables, dwell time SCM, aggregated SCMs, angular SCM, phase difference matrices SCM , vector spreadsheets) or matrices created using sample data. In one

8844

-39-

Examples: Computer instructions 512 are executed by processor 502 outside of volatile memory 504 to perform all or part of the method (or operation) 400 of Figures 4a and 4b.

In one embodiment, the computer 500 may be the same or substantially similar to both components of the modular unit 130 DF and/or processor 330 DF, e.g., the modular unit

<p dir="rtl">5 Spatial SCM 335, principal component standard unit 340, distance measurement unit 350, minimum distance standard unit 355 and table m standard unit 345. The computer 500 may perform all of the same functions and be configured to receive and generate the same data as a whole component of the DF standard unit 130 and /or DF processor 330 as described herein, such as space sample covariance matrix (SCM) module 335, principal component module 340, module</p>

<p dir="rtl">10 Standard unit for measuring distance 350, standard unit for minimum distance 355 and standard unit for table m 345. For example, computer 500 may be configured to perform real-time direction determination, capture data corresponding to signals incident on one or more array elements, create tables, and and/or matrices (e.g., M tables, SCM residence time, summed SCMs, angular SCM, SCM phase difference matrices, vector data tables) to determine the direction of</p>

<p dir="rtl">15 Signal arrival.</p>

Method 400 is not limited to using the hardware and software of Figure 5; They may find applicability in any computing or processing environment and with any type of machine or group of machines capable of running a computer program. Method 400 may be implemented in hardware, software, or a combination of the two. Method 400 can be applied to computer programs that are executed on computers/

<p dir="rtl">20 Programmable machines, each of which includes a processor, storage medium, or other processor-readable material (including volatile and non-volatile memory and/or storage elements), at least one input device, and one or more output devices. Program code may be applied to data entered using an input device to implement method 400 and to generate output information.</p>

<p dir="rtl">25 The system may be implemented, at least in part, via a computer software product (e.g.,</p>

In a machine-readable storage device), for execution by it, or to control the operation of a data processing device

8844

-40-

10

15

20

25

(For example, a programmable processor, a computer, or multiple computers). Each such program may be implemented in a high-level procedural or object-oriented programming language to communicate with the computer system. However, the Program in assembly or machine language. The language may be a compiled or interpreted language and may be arranged in any form, including as a stand-alone program or as a standard unit, component, subroutine, or other unit suitable for use in the computing environment. Alternatively, the system may be implemented, at least in part, as firmware.

A computer program can be arranged to be executed on a single computer, on multiple computers at a single location, or distributed across multiple locations and linked by a communications network. A computer program may be stored on a storage medium or device (for example, CD-ROM, hard disk, magnetic diskette) that can be read by a general computer or for a special purpose. Programmable to configure and operate a computer when the storage medium or device is readable by a computer to implement method 400. Method 400 may also be implemented as a machine-readable storage medium, formed using a computer program, where upon execution, instructions in the computer program cause the computer to operate in accordance with method 400.

Method 400 may be implemented by one or more programmable processors that execute one or more computer programs to perform system functions. All or part of the system may be implemented as a special purpose logic circuitry (for example, a field programmable gate array (FPGA)) and/or an application-specific integrated circuit (ASIC). (.

A number of incarnations of Sister Arraa were described. However, it will be understood that various modifications can be made without departing from the spirit and scope of the invention. Elements of various embodiments described herein may be combined to form other embodiments not specifically described above. Other embodiments not specifically described herein are also within the scope of the following protections.

Bookmark drawings

8844

-41-

Figure 2

Residence time 1

Residence time 2

Z residence time 3

<p dir="rtl">5 205a Calculate Cxx1</p>

205b Calculate Cxx2

205C Calculate Cxx3

206 Calculate Cxx

Figure 4

<p dir="rtl">10 A beginning</p>

402 Select a first set of subarrays from a selection of subarrays.

404 During the first dwell time, collect data using the first selected set of arrays

sub.

406 Construct a sample residence time spatial covariance matrix (SCM) for the selected set of

<p dir="rtl">15 Subarrays.</p>

408 Are there other sets of subarrays?

B Yes

410 Choose the next different set of subarrays from the selection of subarray sets.

20 412 During the second dwell time, collect data using the selected set of arrays

sub.

C No

414 Integrate the spatial SCMs during the residence time to form a total covariance matrix.

416 Formation of SCM corners using the total covariance matrix.

<p dir="rtl">25 418 Pre-calculation of the initial component table using angle and frequency measurements for one or more of</p>

Primary components.

8844

-42-

420 Extract one or more elementary components as composite phases from the angular SCM to form a point

the test.

422 Determine the distance between the test point and each value in the pre-calculated initial component table.

424 Set the minimum distance point corresponding to the direction of the received signal.

<p dir="rtl">5 D the end.</p>

8844

-43-