BEAMFORMING TECHNIQUES IMPLEMENTING THE ITERATIVE ADAPTIVE APPROACH (IAA)

0Associated
Cases 
0Associated
Defendants 
0Accused
Products 
0Forward
Citations 
0
Petitions 
1
Assignment
First Claim
1. A method of performing beamforming, comprising:
 applying matched filtering to each of a plurality of received signals to generate a virtual antenna array output for a multipleinput multipleoutput (MIMO) radar system;
performing, using the virtual antenna array output, delayandsum (DAS) beamforming to generate an initial power spectrum estimation associated with the plurality of received signals; and
generating beamforming data based on an iterative adaptive approach (IAA) that includes a number of iterations using the initial power spectrum estimation as an initial power estimation, and iteratively updating and evaluating the initial power estimation in one of a plurality of different beam directions.
1 Assignment
0 Petitions
Accused Products
Abstract
Techniques are disclosed implementing two alternative approaches for adaptive beamforming for MIMO radar. The first of these includes a “reduced complexity” iterative adaptive approach (RCIAA) algorithm, which uses two steps including a delayandsum beamforming step (DASBF) and an IAA step that is applied to the output generated by the DASBF step. A second technique is described that includes a “beam space” iterative adaptive approach (BSIAA) algorithm, which uses three steps including a delayandsum beamforming step (DASBF), a region of interest (ROI) detection step that is applied to the output generated by the DASBF, and an IAA step that is applied to detected ROIs.
0 Citations
No References
No References
24 Claims
 1. A method of performing beamforming, comprising:
applying matched filtering to each of a plurality of received signals to generate a virtual antenna array output for a multipleinput multipleoutput (MIMO) radar system; performing, using the virtual antenna array output, delayandsum (DAS) beamforming to generate an initial power spectrum estimation associated with the plurality of received signals; and generating beamforming data based on an iterative adaptive approach (IAA) that includes a number of iterations using the initial power spectrum estimation as an initial power estimation, and iteratively updating and evaluating the initial power estimation in one of a plurality of different beam directions.  View Dependent Claims (2, 3, 4, 5, 6)
 7. A nontransitory computerreadable media having instructions stored thereon that, when executed by one or more processors, cause the one or more processors to perform beamforming by:
applying matched filtering to each of a plurality of received signals to generate a virtual antenna array output for a multipleinput multipleoutput (MIMO) radar system; performing, using the virtual antenna array output, delayandsum (DAS) beamforming to generate an initial power spectrum estimation associated with the plurality of received signals; and generating beamforming data based on an iterative adaptive approach (IAA) for a number of iterations using the initial power spectrum estimation as an initial power estimation, and evaluating the initial power estimation in one of a plurality of different beam directions.  View Dependent Claims (8, 9, 10, 11, 12)
 13. A method of performing beamforming, comprising:
applying matched filtering to each of a plurality of received signals to generate a virtual antenna array output for a multipleinput multipleoutput (MIMO) radar system; performing, using the virtual antenna array output, delayandsum (DAS) beamforming to generate an initial power spectrum estimation associated with the plurality of received signals; identifying one or more regions of interest (ROIs) in beam space using the initial power spectrum estimation, the one or more ROIs including one or more beam sectors associated with a detected target based upon a beam power exceeding a detection threshold value; and generating beamforming data based on an iterative adaptive approach by using, for each identified ROI from among the one or more identified ROIs, the initial power spectrum estimation as an initial power estimation in beam space and, for each of a number of iterations for each ROI, iteratively updating the initial power estimation until a power estimation is performed for each active beam sector associated with each of the one or more identified ROIs.  View Dependent Claims (14, 15, 16, 17, 18)
 19. A nontransitory computerreadable media having instructions stored thereon that, when executed by one or more processors, cause the one or more processors to perform beamforming by:
applying matched filtering to each of a plurality of received signals to generate a virtual antenna array output for a multipleinput multipleoutput (MIMO) radar system; performing, using the virtual antenna array output, delayandsum (DAS) beamforming to generate an initial power spectrum estimation associated with the plurality of received signals; identifying one or more regions of interest (ROIs) in beam space using the initial power spectrum estimation, the one or more ROIs including one or more beam sectors associated with a detected target based upon a beam power exceeding a detection threshold value; and generating beamforming data based on an iterative adaptive approach (IAA) approach using, for each identified ROI from among the one or more identified ROIs, the initial power spectrum estimation as an initial power estimation in beam space and, for each of a number of iterations for each ROI, iteratively updating the initial power estimation until a power estimation is performed for each of active beam sector associated with each of the one or more identified ROIs.  View Dependent Claims (20, 21, 22, 23, 24)
1 Specification
Aspects described herein generally relate to radar systems and, more particularly, to techniques implementing the iterative adaptive approach (IAA) for radar beamforming.
Radar systems often require antenna arrays that allow DoA (Direction of Arrival) estimation to be performed. Approaches to performing DoA estimation algorithms typically fall into two categories: classic and adaptive. The classic approach implements a low complexity and low resolution solution, whereas the adaptive approach implements a high complexity and high resolution solution. Each of these techniques has significant drawbacks.
The accompanying drawings, which are incorporated herein and form a part of the specification, illustrate the aspects of the present disclosure and, together with the description, and further serve to explain the principles of the aspects and to enable a person skilled in the pertinent art to make and use the aspects.
The exemplary aspects of the present disclosure will be described with reference to the accompanying drawings. The drawing in which an element first appears is typically indicated by the leftmost digit(s) in the corresponding reference number.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the aspects of the present disclosure. However, it will be apparent to those skilled in the art that the aspects, including structures, systems, and methods, may be practiced without these specific details. The description and representation herein are the common means used by those experienced or skilled in the art to most effectively convey the substance of their work to others skilled in the art. In other instances, wellknown methods, procedures, components, and circuitry have not been described in detail to avoid unnecessarily obscuring aspects of the disclosure.
Multipleinput multipleoutput (MIMO) radar is an advanced type of phased array radar implementing digital receivers and waveform generators distributed across an aperture. MIMO radar signals propagate using closelyspaced antennas within an array to obtain better spatial resolution, Doppler resolution, and dynamic range. In a traditional phased array MIMO radar systems, additional antennas and related hardware are needed to improve spatial resolution. To accomplish this, traditional MIMO radar systems transmit mutually orthogonal signals from multiple transmit antennas (denoted as Ntx), and these waveforms are then extracted from each of the receive antennas (Nrx) using a set of matched filters. For example, if a MIMO radar system has 3 transmit antennas and 4 receive antennas, 12 signals can be extracted from the receiver because of the orthogonality of the transmitted signals. That is, a 12element “virtual” antenna array is created using only 7 antennas by performing digital signal processing on the received signals, thereby obtaining a finer spatial resolution compared with its phased array counterpart.
Beamforming is used within a MIMO radar system as part of a Direction of Arrival (DoA) algorithm to dynamically identify the DoA for radar signals. But, as mentioned above, both classic and adaptive DoA estimation algorithms have drawbacks. In particular, classic beamforming approaches lack the spatial resolution required for certain applications, such as autonomous driving radar systems for instance. Therefore, adaptive beamforming systems attempt to address the shortcomings of the classic beamforming solutions, but do so by introducing a significant amount of complexity to achieve a higher spatial resolution, which in turn requires significant processing power and added system cost.
More specifically, classic beamforming systems generally rely upon delayandsum (DAS) beamforming techniques to offer a lowcomplexity and low resolution solution by implementing a spatial filter with constant coefficients. The filter coefficients are derived based on a desired filter pattern (spatial frequency response). Adaptive beamforming, on the other hand, functions as a highcomplexity and high resolution beamforming solution by operating at the virtual array output, which again has dimensions equal to the number of transmit antennas multiplied by the number of receive antennas. Such “full virtual array” adaptive beamforming solutions function to adjust the filter response based on received signal statistics, with the goal of the adjustment being a minimization of some criterion. For instance, for MVDR (Minimum Variance Distortionless Response), the goal is a minimization of the beamforming output power under the constraint of unity gain in a desired direction.
However, because practical MIMO radar systems typically implement a large number of antennas (e.g., 30 or more receive and transmit antennas), the size of the virtual array may represent tens up to thousands of virtual channels. Thus, because adaptive beamforming algorithms use the entire virtual array as part of the calculations, as discussed in further detail below, the calculations involving these virtual arrays (e.g. the inversion of large matrices that the virtual array may represent) is very computationally demanding and a numerically unstable operation.
Further to this point, current adaptive beamforming (BF) algorithms may be known as “super resolution algorithms,” and are conventionally used to achieve the high spatial resolution required for radar applications, but are limited in two aspects. First, adaptive beamforming requires that signal processing be implemented to perform an estimation of the covariance (COV) matrix of received signals, and to provide an accurate estimation of the COV matrix, many signal “snapshots” are required. However, in practical applications, the number of snapshots is often limited and, in some cases, only a single snapshot is available. The estimation of the COV matrix with such a low number of snapshots yields an illconditioned COV matrix, which can cause errors and degrade system performance. Second, an inversion of the COV matrix is required as part of the adaptive beamforming algorithm, and this calculation is a highly complex operation. Further complicating this issue, if the COV matrix is also an illconditioned matrix, such a matrix inversion calculation is a numerically unstable operation.
Therefore, to address these issues, the aspects described herein implement two alternative approaches for adaptive beamforming for MIMO radar. The first of these includes what is referred to herein as a “reduced complexity” iterative adaptive approach (RCIAA) algorithm, which is described in further detail in Section 1 of the present disclosure, and is directed to
The RCIAA algorithm discussed in this Section is implemented by using a COV matrix having a reduced size having dimensions of Nrx×Nrx. Again, to perform full virtual array processing, the COV matrix size used for conventional adaptive beamforming algorithms has dimensions of NrxNtx×NrxNtx. This size reduction significantly reduces the computation complexity of adaptive beamforming. In addition, the aspects described herein are also based upon an iterative adaptive approach (IAA) for COV matrix estimation, which estimates a wellconditioned COV matrix even from a single snapshot.
As further discussed below, the aspects described herein implement an adaptive beamforming algorithm using a twostep process. In a first step, conventional delay and sum (DAS) beamforming is applied on matched filter outputs that are received from the same RX antenna. The number of outputs of the DAS beamformers is thus equal to the number of the RX antennas (Nrx). Then, as a second step, adaptive beamforming (e.g. Capon or amplitude and phase estimation of a sinusoid (APES)) is applied to the Nrx outputs from the first step. The main advantage of this approach is that adaptive beamforming is applied to an array of size Nrx instead of Nrx×Ntx, as is the case for a virtual array. As a result, issues related to COV matrix estimation and the calculations associated with COV matrix inversion become less critical. Although this approach results in some loss in degrees of freedom (i.e. number of nulls) in the adaptive beamformer, this tradeoff is acceptable given the processing advantages gained as a result of the reduction in size in the virtual array.
Conventional methods for implementing adaptive beamforming algorithms also include a twostep approach that use Capon or, alternatively, APES as a second step. But to obtain a wellconditioned COV matrix, a number of snapshots L is required to satisfy the condition L>Nrx. But again, as noted above, often only a single snapshot is available. Moreover, the aspects described herein utilize a single snapshot because the range and Doppler processing (which is a type of matched filer) is applied before the beamforming is performed. Therefore, the aspects as further described herein obtain a single virtual array output (single snapshot) for each rangeDoppler bin. Conventional twostep approaches are not capable of performing adaptive beamforming using a single snapshot in this way.
In an aspect, the MIMO radar device 100 may include processing circuitry 102, a memory 104, a transmit antenna array 120 including any suitable number Ntx of transmit antennas and a receive antenna array 130 including any suitable number Nrx or receive antennas. The components shown in
In an aspect, the various components of the MIMO radar device 100 may be identified with functionality further described herein with reference to the example block diagram 150 as shown and described with reference to
As referenced further in this and the other Sections, the term “mmWave frequency,” which the aspects described in this Section as well as others may operate, may include, for example, frequencies and frequency bands above 20 GHz, 24 GHz, 28 GHz, etc., up to an upper frequency. For instance, mmWave frequency bands may include frequencies ranging from 20 GHz to 300 GHz, from 24 GHz to 300 GHz, etc. This may include, for instance, the various bands known to be associated with or otherwise referred to as mmWave frequency bands such as 24 GHz, 28 GHz, 37 GHz, 39 GHz, 40 GHz, 47 GHz, 60 GHz, etc.
To do so, processing circuitry 102 may be configured as any suitable number and/or type of computer processors, which may function to control the MIMO radar device 100 as discussed herein. Processing circuitry 102 may be identified with one or more processors (or suitable portions thereof) implemented by the MIMO radar device 100. As discussed herein, processing circuitry 102 may, for example, be identified with one or more processors implemented by the MIMO radar device 100 such as a host processor, a digital signal processor, one or more microprocessors, microcontrollers, an applicationspecific integrated circuit (ASIC), part (or the entirety of) a fieldprogrammable gate array (FPGA), etc. In any event, aspects include the processing circuitry 102 being configured to carry out instructions to perform arithmetical, logical, and/or input/output (I/O) operations, and/or to control the operation of one or more components of the MIMO radar device 100 to perform various functions associated with the aspects as described herein.
For example, the processing circuitry 102 can include one or more microprocessor cores, memory registers, buffers, clocks, etc., and may generate electronic control signals associated with electronic components to control, tune, and/or modify the operation of one or more components of the MIMO radar device 100 as discussed herein. Moreover, aspects include processing circuitry 102 communicating with and/or controlling functions associated with the memory 104 and/or other components of the MIMO radar device 100. This may include, for example, conditioning signals prior to transmission via the transmit antenna array 120 and processing the reflected radar signals received via the receive antenna array 130.
In an aspect, the memory 104 stores data and/or instructions such that, when the instructions are executed by the processing circuitry 102, the processing circuitry 102 performs various functions as described herein. The memory 104 can be implemented as any wellknown volatile and/or nonvolatile memory, including, for example, readonly memory (ROM), random access memory (RAM), flash memory, a magnetic storage media, an optical disc, erasable programmable read only memory (EPROM), programmable read only memory (PROM), etc. The memory 104 can be nonremovable, removable, or a combination of both.
For example, the memory 104 may be implemented as a nontransitory computer readable medium storing one or more executable instructions such as, for example, logic, algorithms, code, etc. As further discussed below, the instructions, logic, code, etc., stored in the memory 104 are represented by the various modules as shown in
In an aspect, the executable instructions stored in matched filter control circuitry 105 may facilitate, in conjunction with the processing circuitry 102, the correlation of a known delayed signal, or template, with an unknown signal to detect the presence of the template in the unknown signal. This known signal may be transmitted as part of the radar signals transmitted via the transmit array 120, for example. Thus, the matched filter control circuitry 105 may function to detect known signals as part of one or more radar signal processing operations. For instance, the matched filter control circuitry may detect known signals received via the receive antenna array 130 in accordance with any suitable techniques, including known techniques, that may be performed as part of known radar system processing operations. This may include, for instance, the use of pulse compression or other matched filtering to exploit the Doppler effect to identify moving targets and/or the range to identified targets. The functionality of matched filters is generally understood with respect to radar signal processing operations, and thus the operation of the matched filter control circuitry 105 is not further described in detail herein.
Referring now to
As shown in
As discussed above, traditional adaptive beamforming systems utilize the entire virtual array as part of a full virtual array processing, which is significantly complex and is processor intensive given the large virtual array size. In contrast to this approach, the aspects described herein advantageously utilize a lower complexity twostep approach. A summary of the algorithm including both of these steps is shown in Tables 2A2B and further discussed below. As part of this approach, the first step includes the use of a number of delay and sum (DAS) circuitry blocks equal to the number Nrx of receive antennas. Thus, in the example block diagram 150, the MIMO radar device 100 includes DAS circuitry blocks 155.1155. Nrx, with each DAS circuitry block 155 being coupled to the outputs of the set of matched filters associated with each receive antenna within the receive antenna array 130 as shown in
The DAS circuitry blocks 155.1155.Nrx may be implemented with any suitable type of hardware components configured to timealign the outputs of each of the matched filters. These component may include, for instance, any suitable type of delay elements, summer circuits, correlators, etc., that ensure the outputs of each matched filter are substantially timealigned with one another and summed together to constructively add as part of a beamforming operation. The DAS circuitry blocks 155.1155.Nrx may be implemented in accordance with nay suitable technique to achieve this functionality, including techniques known to be utilized for such operations in accordance with beamforming systems. In an aspect, the DAS circuitry blocks 155.1155.Nrx may be tuned, controlled, monitored, and/or each respective received signal processed via the delay and sum algorithm 107 as shown in
In an aspect, the outputs provided by the DAS circuitry blocks 155.1155.Nrx constitute an initial power spectrum estimation for the radar signals received via the receive antenna array 130. Thus, the DAS circuitry blocks 155.1155.Nrx function to distill the initial size of the virtual array, which is conventionally used for adaptive beamforming algorithms and has a size of Ntx×Nrx, to an initial power spectrum estimation having a size equal to the number of receive antennas Nrx.
In an aspect, the reducedcomplexity iterative adaptive approach (RCIAA) algorithm, which is shown in further detail below in Tables 2A2B, is implemented as a combination of the two steps described herein. Thus, aspects include the RCIAA algorithm being performed via execution of instructions stored in the RCIAA algorithm module 111 by the processing circuitry 102 as shown in
Thus, the aspects of the RCIAA algorithm as described implements two steps. In a first step, as shown and described in further detail below with reference to Table 2A, an initial DAS spectrum estimation is performed using a DAS beamforming algorithm.
Then, in a second step, which is shown and described in greater detail with respect to Table 2B below, an algorithm is implemented in accordance with what is known as the iterative adaptive approach (IAA), which operates on the aforementioned initial power spectrum estimation. Table 1 below provides a general form of the IAA algorithm. The IAA is a known spectral estimation technique that is based on a weighted least squares minimization. However, the aspects described herein are not limited to this particular implementation of the IAA technique, and other variations may be implemented without departing from the spirit and scope of the disclosure.
Now turning to Table 2B and the second step of the RCIAA algorithm, the initial power spectrum estimation obtained in the first step (DAS beamforming—Table 2A) is represented as the virtual steering matrix A. In an aspect, the second step of the RCIAA algorithm operates on this virtual steering matrix A in accordance with an application of the iterative adaptive approach (IAA).
With respect to the Table 2A below, the RCIAA algorithm uses the following notations:
Spectrum grid: {θ_{i}}_{0}^{N}^{grid}^{−1 }
TX steering vector: a_{t}(θ_{i})
RX steering vector: a_{r}(θ_{i})
Window for spectral estimation: w
Power in direction θ_{i}: p_{i }
Again, as shown in Table 2A below, the first step includes an initial DAS spectrum estimation that defines an initial TX steering vector a_{t}_{i}, an initial receive steering vector a_{r}_{i}, an initial virtual steering vector a_{v}_{i}, a DAS power estimation with windowing defined as p_{i}, and an operation associated with building a virtual steering matrix A(:, i). Furthermore, to demonstrate the level of complexity of the RCIAA algorithm aspects described herein, several example parameters are shown that result in a measured complexity in terms of MAC.
As shown in further detail in Table 2B above, the IAA covariance (COV) matrix estimation is represented as R_{iaa}, whereas the inverse of the IAA covariance matrix estimation is represented as R_{iaa}^{−1}. In an aspect, the inverse of the COV matrix estimation may be calculated using Cholesky decomposition, although this is by way of example and not limitation, and any suitable techniques may be implemented to calculate the inverse of the covariance matrix estimation R_{iaa}^{−1}, in accordance with various aspects. Thus, this portion of the IAA algorithm forms an “outer loop” in which a number of iterations are performed with respect to the calculations of the power estimation of the virtual steering matrix A (5 in this example as iter_{num}=5).
In an aspect, the IAA algorithm operates using a second “inner loop” that is nested within the outer loop as noted above. In this inner nested loop, the power in a direction as defined by θ_{i }is iteratively calculated for each i value until a maximum number of spectrum samples is reached (200 in this example as Ngrid=200). Thus, the IAA algorithm functions to iteratively update and evaluate the power over a number of samples for each iteration of a different direction θ_{i }until an appropriate weighted beamforming combination is calculated. This weighted beamforming combination is then output as shown in
In other words, by implementing the IAA technique in the second step of the RCIAA algorithm as described above, the COV matrix is estimated iteratively, when it is modeled as R=A diag(p)A^{H }(see line 4 in Table 1), where A is a matrix of the steering vectors and P=diag(p) is a diagonal matrix of powers from all angles of arrival (AoA) (i.e. a spectrum estimation). Once the COV matrix is estimated, the power spectrum in the inner loop is estimated (see lines 58 in Table 1). Then, once a better spectrum estimation is obtained, the COV matrix is estimated again, and so on. Again, aspects include the initial spectrum estimation as shown in Table 1, lines 12, being calculated via the DAS beamforming algorithm as shown in
Thus, using the example values as shown in Tables 2A2B above, an overall complexity value in terms of MAC is calculated for the overall RCIAA algorithm of approximately 5 M. The complexity and performance of the RCIAA algorithm is shown in further detail below in Table 3 compared to other conventional techniques to calculate beamforming data.
The first algorithm listed is the DAS BF applied to the entire virtual array as shown in
The second algorithm in Table 3 above is an IAA algorithm applied to the entire virtual array as shown in
The third algorithm is the RCIAA to which the aspects of the present disclosure are directed, which is summarized in Tables 2A and 2B above along with a corresponding complexity analysis in the righthand column. As can be seen from the results of the complexity analysis, the RCIAA only introduces a slight amount of complexity compared to nonadaptive solutions to achieve an adaptable beamforming solution (5M vs 4M) and is significantly less complex compared to the IAA virtual array solution (5M vs 55M).
In summary, the aspects described in this Section provide super resolution spectrum estimation similar to IAA techniques applied to the entire virtual array. The resolution of the adaptive beamforming solution achieved via the RCIAA algorithm is significantly better than the nonadaptive DAS beamforming solutions as shown by the graphs in
The method flow 500 may begin by applying (block 502) matched filtering to each of a set of received signals to generate a virtual antenna array output. This may include, for instance, the use of the matched filters as shown and described with reference to
The method flow 500 may continue by performing (block 504) DAS beamforming to the virtual antenna array output to generate an initial power spectrum estimation with respect to the received signals. This may include, for instance, the use of the DAS beamforming algorithm as shown and described with reference to the DAS beamforming circuitry blocks 155.1155.Nrx as shown in
The method flow 500 may continue by determining (block 506) an initial COV matrix estimation using the initial power spectrum estimation with respect to the received signals. This may include, for instance, the use of the IIA beamforming algorithm as shown and described with reference to the IAA algorithm circuitry block 109 as shown in
The method flow 500 may continue assuming that the maximum number of iterations has not been reached (block 508—NO). In other words, the IAA algorithm may proceed as shown in Table 2A and continue to perform processing operations as discussed herein for a suitable number of iterations (e.g. 5).
Assuming that the maximum number of iterations is not reached, the method flow 500 may continue to calculate (block 510) power in one direction by evaluating a number of power spectrum samples (e.g., 200). As an example, power calculation may be performed in accordance with the inner loop of the IAA algorithm as shown in Table 2B. Continuing this example, the method flow 500 may continue to iteratively calculate, within the outer loop, the power in one direction until the last sample is reached (block 512—YES). When this occurs, the method flow 500 may proceed to update the COV matrix estimation using the calculated power in one direction, and then repeat this process for additional directions until the maximum number of iterations (e.g., 5) has been reached (block 508—YES). Once this occurs, the IAA algorithm may generate (block 514) beamforming data using the most recent calculations of the COV matrix and the power calculations in each direction.
The BSIAA algorithm discussed in this Section, similar to the RCIAA algorithm, also provides a lower complexity solution with respect to existing adaptive beamforming techniques, as the algorithm does not apply the IAA algorithm to the entire virtual array. However, the BSIAA implements an additional step as further described below. The first step of the BSIAA algorithm is similar or identical to the first step of the RCIAA algorithm discussed in Section 1. In particular, the BSIAA algorithm described in this Section may also estimate the initial spectrum using DAS beamforming. However, instead of applying the IAA algorithm to the output of the DAS beamforming step, an intermediate step is introduced that detects regions of interest (ROIs) as one or more angular regions where it is suspected that one or more targets are located. As the third step, a local adaptive beamforming algorithm is executed on each of the detected ROIs, significantly improving the resolution on the ROIs. The local adaptive beamforming algorithm is based on combining the IAA algorithm and beam space processing, as further discussed below.
Thus, the BSIAA algorithm aspects also provide advantages with respect to the classic and adaptive beamforming techniques described above. In particular, the BSIAA beamforming algorithm provides a high resolution output that is equal (or nearly equal) to the advanced adaptive beamforming methods (such as full virtual array IAA adaptive beamforming. Moreover, the BSIAA algorithm aspects described in this Section may also be implemented with a reduced complexity compared to existing adaptive beamforming solutions. As further discussed below, this is due to the additive beamforming being applied only to regions that are suspected to contain targets, whereas existing superresolution approaches require that the adaptive beamforming calculations be applied at each point of the angle axis grid. Furthermore, the use of beam space processing enables a reduction in the size of COV matrix, which significantly reduces the complexity of its inversion.
In an aspect, the MIMO radar device 600 may include processing circuitry 602, a memory 604, a transmit antenna array 620 including any suitable number Ntx of transmit antennas and a receive antenna array 630 including any suitable number Nrx or receive antennas. These components may be identical to or operate in a substantially similar manner as the processing circuitry 102, the memory 104, the transmit antenna array 120, and the receive antenna array 130, respectively, as described in Section 1 with reference to the MIMO radar device 100 as shown in
Moreover, like the MIMO radar device 100, the MIMO radar device 600 may also include matched filter control circuitry 605, which may function in a manner that is identical to or substantially similar as the matched filter control circuitry 105 of the MIMO radar device 100 as described in Section 1, and may detect known signals as part of one or more radar signal processing operations. The MIMO radar device 600 also stores executable code, logic, instructions, etc., to facilitate the execution of the BSIAA algorithm, as further discussed in this Section. Thus, the MIMO radar device 600 includes several modules stored in the memory 604 that constitute a different executed function associated with the overall operation of the BSIAA algorithm, as further described below.
In an aspect, the beam space iterative adaptive approach (BSIAA) algorithm is implemented as a combination of the three steps described herein. A summary of the BSIAA algorithm including each of these steps is shown in Tables 4A4C and further discussed below. Thus, aspects include the BSIAA algorithm being performed via execution of instructions stored in the BSIAA algorithm module 611 by the processing circuitry 602 as shown in
Referring now to
Thus, as shown in
The DAS circuitry blocks 655.1655.Nrx may also be implemented identical to or substantially similar as the DAS circuitry blocks 155.1155.Nrx, as shown in
In any event, this first step of the BSIAA algorithm, which includes the initial spectral estimation, is reproduced again below in Table 4A for clarity and includes some additional notations that are relevant to the BSIAA algorithm. As shown in Table 4A below, to reduce the sidelobes of the beam pattern, aspects include using a windowing function (e.g. a Hamming or Chebyshev window) as part of the DAS beamforming step. This windowing function may also be used as part of the step 1 DAS beamforming algorithm in Section 1 with respect to
As shown in Table 4A above, the first step in the BSIAA algorithm functions to perform DAS beamforming to provide an initial spectrum estimation, which is similar to the DAS beamforming algorithm as shown in Table 2A. However, the DAS beamforming algorithm shown in Table 4A, which is used as part of the BSIAA algorithm, implements beam space processing to do so. To provide additional clarity, an explanation of beam space processing is presented in further detail below with respect to the inset 680 in
As shown in inset 680 of
With the following notations understood as defined above, the beam space transformation used in accordance with the beam space processing as shown in inset 680 of
With respect to the use of Capon beamforming, in the element
Moreover, COV matrix R has a size equal to N×N. In the beam space,
where R_{bs}=B_{bs }R B_{bs}^{H }is the COV matrix in the BS of size N_{bs}×N_{bs }and a_{bs}=B_{bs }a.
The array symmetry of the uniform linear array (ULA) is generally used to improve the COV matrix estimation using a technique known as forwardbackward (FB) averaging. The application of this technique on the COV matrix in the beam space is given by Rbsfb={Rbs}. An additional advantage of the FB averaging is that the COV matrix becomes real, and therefore this reduces the number of complex multiplications required in further calculations. As an example of beam space transformation,
Thus, the first step of the BSIAA algorithm implements beam space processing to provide the initial spectral estimation using the virtual array. The following assumptions are also provided with respect to the DAS beamforming algorithm shown in Table 4A:
1. The number index in ROI is 20% of the grid, N_{ROI}=40;
2. The number of active sectors N_{act sect}=5;
3. All indices in the active sector are in the ROI A number of ROI indices (samples of the grid) per sector: N_{samp_sect}=200/20=10.
Again, the DAS beamforming step in Table 4A may implement windowing to reduce the sidelobes of the beam pattern. However, windowing functions may cause a widening of the main lobe of the beam, which yields reduced spatial resolution. Therefore, to improve the resolution, aspects include detecting regions of interest in the second step, which indicates regions where targets are suspected to be found, and then applying local adaptive beamforming in the third step to further improve the resolution in these detected regions.
As discussed in further detail below, the second step of the BSIAA algorithm thus uses the initial power spectrum estimation provided by the DAS beamforming (step 1) to identify one or more regions of interest (ROIs) in beam space. An example algorithm associated with this second step of the BSIAA algorithm is shown below in Table 4B.
In various aspects, the ROI detection may be performed using any number of suitable techniques, such as one of the classic detection methods including orderedstatistic constant falsealarm rate (OSCFAR) or cell averaging constant falsealarm rate (CACFAR). As OSCFAR and CACFAR are known techniques, additional details regarding this portion of the ROI detection algorithm as shown in Table 4B are not further described herein. In the example ROI detection algorithm shown in Table 4B, the ROI detection algorithm implements OSCFAR to detect peaks in the spectrum. The ROI detection algorithm is further discussed below with reference to
In an aspect, the ROI detection algorithm implements a detection threshold, which is denoted in
In an aspect, portions of the spectrum having a power that exceeds the threshold value Thr define a region 702 as shown in
As further discussed below with reference to Table 4C, aspects include the BSIAA algorithm advantageously executing, as the third step, the IAA algorithm only on the active beam sectors as indicated in each of the detected ROIs using beam space processing, as discussed above and further described in Table 4C below, to reduce time and complexity required to identify the desired beamforming pattern and DoA. An example algorithm associated with this third step of the BSIAA algorithm is shown below in Table 4C.
As shown in Table 4C above, within the IAA (outer) loop, the COV matrix R_{iaa }is estimated iteratively (similarly to line 4 in Table 1 with respect to R). In other words, the IAA COV matrix estimation is based on samples in the ROI, with diagonal loading.
Next, for each active beam sector (i.e., the beam sectors associated with each detected ROI), a beam transform is defined as B_{bs}, which is a set of sector matrices of size N_{bs}×N, where N_{bs }is the size of the beam sector. The rows of DFT matrix (N×N) are then used to produce the B_{bs }matrix using beam space processing techniques as discussed above. Thus, the BS COV matrix for processing each active beam sector is given by R_{iaa bs}=Re{B_{bs }R_{iaa }B_{bs}^{H}}. Again, the operator Re{ } is used as a result of the forwardbackward averaging. The size of this matrix R_{iaa bs }is N_{bs}×N_{bs }instead of N×N in the original COV matrix R_{iaa}. Consequently, the inversion of the matrix R_{iaa bs }to R_{iaa bs}^{−1 }advantageously requires O(N_{bs}^{3}) operations, instead of O(N^{3}) operations. In addition, R_{iaa bs }is a real matrix, which further reduces the complexity of additional calculations. The inverse COV matrix R_{iaa bs}^{−1 }may then be multiplied by the matrix R_{yy bs }for each active beam sector as noted above in Table 4A by the following expression: R_{iaa bs}^{−1 }R_{yy bs }R_{iaa bs}^{−1}.
Therefore, as shown in Table 4C above, the R_{iaa bs }matrix is used to estimate powers in specific beam sectors (similar to the inner loop in Table 1 that operates on the matrix
The COV matrix estimation technique described above with respect to Table 4C is but one example of COV matrix estimation that may be implemented in accordance with the BSIAA algorithm aspects discussed in this Section. Aspects include the use of any suitable type of COV matrix estimation technique once the ROIs are detected in step 2 of the BSIAA algorithm as noted above. For instance, in an aspect, the estimation of the COV matrix R_{iaa }as shown in Table 4C above may alternatively be computed as shown below in Equation 1.
R_{iaa}=A_{p}diag(p·*roi_{mask})A_{p}^{H}+A_{notp}diag(p·*not(roi_{mask}))A_{notp}^{H} Eqn. 1
This alternative approach for COV matrix estimation is more accurate, but has higher complexity. Moreover, the 2nd matrix of the sum in Eqn. 1 (A_{notp}) should be computed only once since it doesn'"'"'t change with the iterations. Therefore, the desired application, cost, and implemented hardware may drive the decision regarding a particular implementation of COV estimation.
Thus, using the example values as shown in Tables 4A4C above, an overall complexity value in terms of MAC is calculated for the BSIAA algorithm of approximately 9.5 M. The complexity and performance of the BSIAA algorithm is shown in further detail below in Table 5 compared to other conventional techniques to calculate beamforming data.
The first algorithm and second algorithms in Table 5 above (the DAS virtual array and the IAA virtual array) are the same as those discussed in Section 1 above and are reproduced from Table 3. The third algorithm is the BSIAA to which the aspects of the present disclosure as discussed in Section 2 are directed, which is summarized in Tables 4A4C above along with a corresponding complexity analysis. As can be seen from the results of the complexity analysis, the BSIAA introduces some additional complexity compared to nonadaptive solutions to achieve an adaptable beamforming solution (9.5M vs 4M), but is significantly less complex compared to the IAA virtual array solution (9.5M vs 55M).
In summary, the BSIAA algorithm aspects described in this Section provide similar accuracy in the peak estimation as an IAA algorithm applied on the full virtual array. In the regions in which peaks were not detected, the BSIAA performs at least as well as nonadaptive DAS beamforming solutions. Thus, the BSIAA aspects achieve a similar resolution compared to full virtual array IAA solutions, but with a significant reduction in complexity to do so (e.g. a factor of 510 depending on the array and algorithmic parameters).
The method flow 1100 may begin by applying (block 1102) matched filtering to each of a set of received signals to generate a virtual antenna array output. This may include, for instance, the use of the matched filters as shown and described with reference to
The method flow 1100 may continue by performing (block 1104) DAS beamforming to the virtual antenna array output to generate an initial power spectrum estimation with respect to the received signals. This may include, for instance, the use of the DAS beamforming algorithm as shown and described with reference to the DAS beamforming circuitry blocks 655.1655.Nrx as shown in
The method flow 1100 may continue by detecting (block 1106) ROIs using the initial power spectrum estimation with respect to the received signals. This may include, for instance, the use of the ROI detection algorithm as shown and described with reference to the ROI detection algorithm circuitry block 608 as shown in
The method flow 1100 may continue by determining (block 1108) an initial COV matrix estimation using the initial power spectrum estimation with respect to the detected ROIs. This may include, for instance, the use of the IIA beamforming algorithm as shown and described with reference to the IAA algorithm circuitry block 609 as shown in
The method flow 100 may continue assuming that the maximum number of iterations has not been reached (block 1110—NO). In other words, the IAA algorithm may proceed as shown in Table 6C and continue to perform processing operations as discussed herein for a suitable number of iterations (e.g. 5).
Assuming that the maximum number of iterations is not reached, the method flow 1100 may continue to calculate (block 1112) power for the first beam sector within the first detected ROI (assuming more than one ROI is detected) by evaluating a number of power spectrum samples (e.g., 200). As an example, these power calculations may be performed in accordance with the two inner loops of the IAA algorithm as shown in Table 6C. Continuing this example, the method flow 1100 may continue to iteratively calculate, within the outer loop, the power for each identified ROI in beam space until the power has been estimated for all beam sectors in a detected ROI (block 1114—YES). Once the power is estimated for each beam sector within an ROI, then the method flow 11100 may include incrementing to the next ROI (block 1108) to update the initial power spectrum estimation (i.e. the COV matrix).
Thus, for each of the number of iterations, the method 100 functions to iteratively update the initial power estimation until a power estimation is performed for each of the active beam sectors associated with each of the one or more identified ROIs to generate the beamforming data (block 1118).
The following examples pertain to further aspects.
Example 1 is a method of performing beamforming, comprising: applying matched filtering to each of a plurality of received signals to generate a virtual antenna array output for a multipleinput multipleoutput (MIMO) radar system; performing, using the virtual antenna array output, delayandsum (DAS) beamforming to generate an initial power spectrum estimation associated with the plurality of received signals; and generating beamforming data based on an iterative adaptive approach (IAA) that includes a number of iterations using the initial power spectrum estimation as an initial power estimation, and iteratively updating and evaluating the initial power estimation in one of a plurality of different beam directions.
In Example 2, the subject matter of Example 1, wherein the MIMO radar system includes a number of transmit antennas Ntx, and a number of receive antennas Nrx, and wherein the virtual antenna array output has a size equal to a multiplication of the number of transmit antennas and the number or receive antennas (Ntx)(Nrx).
In Example 3, the subject matter of any combination of Examples 12, wherein the initial power spectrum estimation comprises a size equal to the number of receive antennas Nrx.
In Example 4, the subject matter of any combination of Examples 13, wherein generating the beamforming data includes initially estimating a covariance (COV) matrix using the initial power spectrum estimation, and iteratively reestimating the COV matrix as the initial power estimation is iteratively updated and evaluated.
In Example 5, the subject matter of any combination of Examples 14, wherein generating the beamforming data includes, for each of the number of iterations, updating and evaluating the initial power estimation in one of the plurality of different beam directions over a predetermined number of spectrum samples.
In Example 6, the subject matter of any combination of Examples 15, wherein generating the beamforming data includes repeating updating and evaluating the initial power estimation for each one of the plurality of different beam directions for each of the plurality of different beam directions.
Example 7 is a nontransitory computerreadable media having instructions stored thereon that, when executed by one or more processors, cause the one or more processors to perform beamforming by: applying matched filtering to each of a plurality of received signals to generate a virtual antenna array output for a multipleinput multipleoutput (MIMO) radar system; performing, using the virtual antenna array output, delayandsum (DAS) beamforming to generate an initial power spectrum estimation associated with the plurality of received signals; and generating beamforming data based on an iterative adaptive approach (IAA) for a number of iterations using the initial power spectrum estimation as an initial power estimation, and evaluating the initial power estimation in one of a plurality of different beam directions.
In Example 8, the subject matter of Example 7, wherein the MIMO radar system includes a number of transmit antennas Ntx, and a number of receive antennas Nrx, and wherein the virtual antenna array output has a size equal to a multiplication of the number of transmit antennas and the number of receive antennas (Ntx)(Nrx).
In Example 9, the subject matter of any combination of Examples 78, wherein the initial power spectrum estimation comprises a size equal to the number of receive antennas Nrx.
In Example 10, the subject matter of any combination of Examples 79, wherein the instructions further cause the one or more processors to generate the beamforming data by initially estimating a covariance (COV) matrix using the initial power spectrum estimation, and iteratively reestimating the COV matrix as the initial power estimation is iteratively updated and evaluated.
In Example 11, the subject matter of any combination of Examples 710, wherein the instructions further cause the one or more processors to generate the beamforming data by, for each of the number of iterations, updating and evaluating the initial power estimation in one of the plurality of different beam directions over a predetermined number of spectrum samples.
In Example 12, the subject matter of any combination of Examples 711, wherein the instructions further cause the one or more processors to generate the beamforming data by repeating the updating and evaluating of the initial power estimation for each one of the plurality of different beam directions for each of the plurality of different beam directions.
Example 13 is a method of performing beamforming, comprising: applying matched filtering to each of a plurality of received signals to generate a virtual antenna array output for a multipleinput multipleoutput (MIMO) radar system; performing, using the virtual antenna array output, delayandsum (DAS) beamforming to generate an initial power spectrum estimation associated with the plurality of received signals; identifying one or more regions of interest (ROIs) in beam space using the initial power spectrum estimation, the one or more ROIs including one or more beam sectors associated with a detected target based upon a beam power exceeding a detection threshold value; and generating beamforming data based on an iterative adaptive approach by using, for each identified ROI from among the one or more identified ROIs, the initial power spectrum estimation as an initial power estimation in beam space and, for each of a number of iterations for each ROI, iteratively updating the initial power estimation until a power estimation is performed for each active beam sector associated with each of the one or more identified ROIs.
In Example 14, the subject matter of Example 13, wherein the act of identifying the one or more ROIs includes an orderedstatistic constant falsealarm rate (OSCFAR) detection.
In Example 15, the subject matter of any combination of Examples 1314, wherein each one of the one or more ROIs is based upon a region in beam space formed around a portion of a respective beam having a power that exceeds the detection threshold value.
In Example 16, the subject matter of any combination of Examples 1315, wherein each one of the one or more ROIs has a respective beam width that is based upon a beam pattern generated as a result of the DAS beamforming.
In Example 17, the subject matter of any combination of Examples 1316, wherein the MIMO radar system includes a number of transmit antennas Ntx, and a number of receive antennas Nrx, and wherein the virtual antenna array output has a size equal to a multiplication of the number of receive antennas by the number of transmit antennas (Ntx)(Nrx).
In Example 18, the subject matter of any combination of Examples 1317, wherein the initial power spectrum estimation comprises a size equal to the number of receive antennas Nrx.
Example 19 is a nontransitory computerreadable media having instructions stored thereon that, when executed by one or more processors, cause the one or more processors to perform beamforming by: applying matched filtering to each of a plurality of received signals to generate a virtual antenna array output for a multipleinput multipleoutput (MIMO) radar system; performing, using the virtual antenna array output, delayandsum (DAS) beamforming to generate an initial power spectrum estimation associated with the plurality of received signals; identifying one or more regions of interest (ROIs) in beam space using the initial power spectrum estimation, the one or more ROIs including one or more beam sectors associated with a detected target based upon a beam power exceeding a detection threshold value; and generating beamforming data based on an iterative adaptive approach (IAA) approach using, for each identified ROI from among the one or more identified ROIs, the initial power spectrum estimation as an initial power estimation in beam space and, for each of a number of iterations for each ROI, iteratively updating the initial power estimation until a power estimation is performed for each of active beam sector associated with each of the one or more identified ROIs.
In Example 20, the subject matter of Example 19, wherein the instructions further cause the one or more processors to identify the one or more ROIs using an orderedstatistic constant falsealarm rate (OSCFAR) detection.
In Example 21, the subject matter of any combination of Examples 1920, wherein the instructions further cause the one or more processors to identify the one or more ROIs based upon a region in beam space formed around a portion of a respective beam having a power that exceeds the detection threshold value.
In Example 22, the subject matter of any combination of Examples 1921, wherein the instructions further cause the one or more processors to identify the one or more ROIs having a respective beam width that is based upon a beam pattern generated as a result of the DAS beamforming.
In Example 23, the subject matter of any combination of Examples 1922, wherein the MIMO radar system includes a number of transmit antennas Ntx, and a number of receive antennas Nrx, and wherein the virtual antenna array output has a size equal to multiplication of the number of receive antennas by the number of transmit antennas (Ntx)(Nrx).
In Example 24, the subject matter of any combination of Examples 1923, wherein the initial power spectrum estimation comprises a size equal to the number of receive antennas Nrx.
Example 25 is a method of performing beamforming, comprising: applying matched filtering means to each of a plurality of received signals to generate a virtual antenna array output for a multipleinput multipleoutput (MIMO) radar system means; performing, using the virtual antenna array output, delayandsum (DAS) beamforming means to generate an initial power spectrum estimation associated with the plurality of received signals; and generating beamforming data based on an iterative adaptive approach (IAA) that includes a number of iterations using the initial power spectrum estimation as an initial power estimation, and iteratively updating and evaluating the initial power estimation in one of a plurality of different beam directions.
In Example 26, the subject matter of Example 25, wherein the MIMO radar system means includes a number of transmit antennas Ntx, and a number of receive antennas Nrx, and wherein the virtual antenna array output has a size equal to a multiplication of the number of transmit antennas and the number or receive antennas (Ntx)(Nrx).
In Example 27, the subject matter of any combination of Examples 2526, wherein the initial power spectrum estimation comprises a size equal to the number of receive antennas Nrx.
In Example 28, the subject matter of any combination of Examples 2527, wherein generating the beamforming data includes initially estimating a covariance (COV) matrix using the initial power spectrum estimation, and iteratively reestimating the COV matrix as the initial power estimation is iteratively updated and evaluated.
In Example 29, the subject matter of any combination of Examples 2528, wherein generating the beamforming data includes, for each of the number of iterations, updating and evaluating the initial power estimation in one of the plurality of different beam directions over a predetermined number of spectrum samples.
In Example 30, the subject matter of any combination of Examples 2529, wherein generating the beamforming data includes repeating updating and evaluating the initial power estimation for each one of the plurality of different beam directions for each of the plurality of different beam directions.
Example 31 is a nontransitory computerreadable means having instructions stored thereon that, when executed by one or more processor means, cause the one or more processor means to perform beamforming by: applying matched filtering means to each of a plurality of received signals to generate a virtual antenna array output for a multipleinput multipleoutput (MIMO) radar system means; performing, using the virtual antenna array output, delayandsum (DAS) beamforming means to generate an initial power spectrum estimation associated with the plurality of received signals; and generating beamforming data based on an iterative adaptive approach (IAA) for a number of iterations using the initial power spectrum estimation as an initial power estimation, and evaluating the initial power estimation in one of a plurality of different beam directions.
In Example 32, the subject matter of Example 31, wherein the MIMO radar system means includes a number of transmit antennas Ntx, and a number of receive antennas Nrx, and wherein the virtual antenna array output has a size equal to a multiplication of the number of transmit antennas and the number of receive antennas (Ntx)(Nrx).
In Example 33, the subject matter of any combination of Examples 3132, wherein the initial power spectrum estimation comprises a size equal to the number of receive antennas Nrx.
In Example 34, the subject matter of any combination of Examples 3133, wherein the instructions further cause the one or more processor means to generate the beamforming data by initially estimating a covariance (COV) matrix using the initial power spectrum estimation, and iteratively reestimating the COV matrix as the initial power estimation is iteratively updated and evaluated.
In Example 35, the subject matter of any combination of Examples 3134, wherein the instructions further cause the one or more processor means to generate the beamforming data by, for each of the number of iterations, updating and evaluating the initial power estimation in one of the plurality of different beam directions over a predetermined number of spectrum samples.
In Example 36, the subject matter of any combination of Examples 3135, wherein the instructions further cause the one or more processor means to generate the beamforming data by repeating the updating and evaluating of the initial power estimation for each one of the plurality of different beam directions for each of the plurality of different beam directions.
Example 37 is a method of performing beamforming, comprising: applying matched filtering means to each of a plurality of received signals to generate a virtual antenna array output for a multipleinput multipleoutput (MIMO) radar system means; performing, using the virtual antenna array output, delayandsum (DAS) beamforming means to generate an initial power spectrum estimation associated with the plurality of received signals; identifying one or more regions of interest (ROIs) in beam space using the initial power spectrum estimation, the one or more ROIs including one or more beam sectors associated with a detected target based upon a beam power exceeding a detection threshold value; and generating beamforming data based on an iterative adaptive approach by using, for each identified ROI from among the one or more identified ROIs, the initial power spectrum estimation as an initial power estimation in beam space and, for each of a number of iterations for each ROI, iteratively updating the initial power estimation until a power estimation is performed for each active beam sector associated with each of the one or more identified ROIs.
In Example 38, the subject matter of Example 37, wherein the act of identifying the one or more ROIs includes an orderedstatistic constant falsealarm rate (OSCFAR) detection.
In Example 39, the subject matter of any combination of Examples 3738, wherein each one of the one or more ROIs is based upon a region in beam space formed around a portion of a respective beam having a power that exceeds the detection threshold value.
In Example 40, the subject matter of any combination of Examples 3739, wherein each one of the one or more ROIs has a respective beam width that is based upon a beam pattern generated as a result of the DAS beamforming.
In Example 41, the subject matter of any combination of Examples 3740, wherein the MIMO radar system means includes a number of transmit antennas Ntx, and a number of receive antennas Nrx, and wherein the virtual antenna array output has a size equal to a multiplication of the number of receive antennas by the number of transmit antennas (Ntx)(Nrx).
In Example 42, the subject matter of any combination of Examples 3741, wherein the initial power spectrum estimation comprises a size equal to the number of receive antennas Nrx.
Example 43 is a nontransitory computerreadable media means having instructions stored thereon that, when executed by one or more processor means, cause the one or more processor means to perform beamforming by: applying matched filtering means to each of a plurality of received signals to generate a virtual antenna array output for a multipleinput multipleoutput (MIMO) radar system; performing, using the virtual antenna array output, delayandsum (DAS) beamforming means to generate an initial power spectrum estimation associated with the plurality of received signals; identifying one or more regions of interest (ROIs) in beam space using the initial power spectrum estimation, the one or more ROIs including one or more beam sectors associated with a detected target based upon a beam power exceeding a detection threshold value; and generating beamforming data based on an iterative adaptive approach (IAA) approach using, for each identified ROI from among the one or more identified ROIs, the initial power spectrum estimation as an initial power estimation in beam space and, for each of a number of iterations for each ROI, iteratively updating the initial power estimation until a power estimation is performed for each of active beam sector associated with each of the one or more identified ROIs.
In Example 44, the subject matter of Example 43, wherein the instructions further cause the one or more processor means to identify the one or more ROIs using an orderedstatistic constant falsealarm rate (OSCFAR) detection.
In Example 45, the subject matter of any combination of Examples 4344, wherein the instructions further cause the one or more processor means to identify the one or more ROIs based upon a region in beam space formed around a portion of a respective beam having a power that exceeds the detection threshold value.
In Example 46, the subject matter of any combination of Examples 4345, wherein the instructions further cause the one or more processor means to identify the one or more ROIs having a respective beam width that is based upon a beam pattern generated as a result of the DAS beamforming.
In Example 47, the subject matter of any combination of Examples 4346, wherein the MIMO radar system means includes a number of transmit antennas Ntx, and a number of receive antennas Nrx, and wherein the virtual antenna array output has a size equal to multiplication of the number of receive antennas by the number of transmit antennas (Ntx)(Nrx).
In Example 48, the subject matter of any combination of Examples 4347, wherein the initial power spectrum estimation comprises a size equal to the number of receive antennas Nrx.
An apparatus as shown and described.
A method as shown and described.
The aforementioned description of the specific aspects will so fully reveal the general nature of the disclosure that others can, by applying knowledge within the skill of the art, readily modify and/or adapt for various applications such specific aspects, without undue experimentation, and without departing from the general concept of the present disclosure. Therefore, such adaptations and modifications are intended to be within the meaning and range of equivalents of the disclosed aspects, based on the teaching and guidance presented herein. It is to be understood that the phraseology or terminology herein is for the purpose of description and not of limitation, such that the terminology or phraseology of the present specification is to be interpreted by the skilled artisan in light of the teachings and guidance.
References in the specification to “one aspect,” “an aspect,” “an exemplary aspect,” etc., indicate that the aspect described may include a particular feature, structure, or characteristic, but every aspect may not necessarily include the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same aspect. Further, when a particular feature, structure, or characteristic is described in connection with an aspect, it is submitted that it is within the knowledge of one skilled in the art to affect such feature, structure, or characteristic in connection with other aspects whether or not explicitly described.
The exemplary aspects described herein are provided for illustrative purposes, and are not limiting. Other exemplary aspects are possible, and modifications may be made to the exemplary aspects. Therefore, the specification is not meant to limit the disclosure. Rather, the scope of the disclosure is defined only in accordance with the following claims and their equivalents.
Aspects may be implemented in hardware (e.g., circuits), firmware, software, or any combination thereof. Aspects may also be implemented as instructions stored on a machinereadable medium, which may be read and executed by one or more processors. A machinereadable medium may include any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computing device). For example, a machinereadable medium may include read only memory (ROM); random access memory (RAM); magnetic disk storage media; optical storage media; flash memory devices; electrical, optical, acoustical or other forms of propagated signals (e.g., carrier waves, infrared signals, digital signals, etc.), and others. Further, firmware, software, routines, instructions may be described herein as performing certain actions. However, it should be appreciated that such descriptions are merely for convenience and that such actions in fact results from computing devices, processors, controllers, or other devices executing the firmware, software, routines, instructions, etc. Further, any of the implementation variations may be carried out by a general purpose computer.
For the purposes of this discussion, the term “processing circuitry” or “processor circuitry” shall be understood to be circuit(s), processor(s), logic, or a combination thereof. For example, a circuit can include an analog circuit, a digital circuit, state machine logic, other structural electronic hardware, or a combination thereof. A processor can include a microprocessor, a digital signal processor (DSP), or other hardware processor. The processor can be “hardcoded” with instructions to perform corresponding function(s) according to aspects described herein. Alternatively, the processor can access an internal and/or external memory to retrieve instructions stored in the memory, which when executed by the processor, perform the corresponding function(s) associated with the processor, and/or one or more functions and/or operations related to the operation of a component having the processor included therein.
In one or more of the exemplary aspects described herein, processing circuitry can include memory that stores data and/or instructions. The memory can be any wellknown volatile and/or nonvolatile memory, including, for example, readonly memory (ROM), random access memory (RAM), flash memory, a magnetic storage media, an optical disc, erasable programmable read only memory (EPROM), and programmable read only memory (PROM). The memory can be nonremovable, removable, or a combination of both.