JP2010066179A - Apparatus and method of high resolution signal processing - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce computation load required for high resolution processing for effective use of a computing unit and memory resources. <P>SOLUTION: An apparatus of high resolution signal processing includes: a covariance matrix generation part 20 for Fourier-transforming a reception signal, detecting a desired signal component in a frequency dimension and generating a covariance matrix using the signal component; and a high resolution angle measurement processing part 23 for calculating an arrival angle of the desired signal from the covariance matrix. It further includes: an angle estimating part 24 for estimating the arrival angle a predetermined time later; a covariance matrix estimating part 26 for estimating a covariance matrix the predetermined time later; a covariance matrix coincidence determining part 21 for determining the coincidence of the generated covariance matrix with the estimated covariance matrix; and a processing selecting part 22 for outputting the arrival angle estimated by the angle estimating part without high resolution processing when the two covariance matrixes are determined to coincide with each other and for causing the high resolution angle measurement processing part to carry out high resolution processing to have the arrival angle calculated to be outputted when the two matrixes are determined not to match with each other. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention generates a covariance matrix from a signal received by a receiving element, and calculates a desired specification such as an arrival angle of the received signal, a distance from the source of the received signal, or a relative speed by high resolution processing. The present invention relates to a processing apparatus and a high resolution signal processing method.

  In a signal processing device having a receiving device composed of a plurality of receiving elements, signal processing for each beam direction component of an incoming wave can be performed by forming a plurality of beams in the spatial direction. However, in this case, the spatial resolution is determined by the width of the beam to be formed, and it is necessary to reduce the beam width in order to increase the resolution. In order to reduce the beam width, there is a method of increasing the distance between the receiving elements.

  Here, as a method for calculating the arrival angle of the desired signal, the distance to the signal source, or the relative speed with high resolution, MUSIC (Multiple Signal Classification: see, for example, Non-Patent Document 1) or ESPRIT (Estimation of Signal Parameters via Rotational) A high-resolution processing algorithm such as “Invance Techniques: see Non-Patent Document 2” is known.

  When such a high resolution processing algorithm is used, a plurality of waves can be separated even in a multi-wave environment where information on a plurality of waves from different sources is mixed in the received signal. Processing by MUSIC or ESPRIT is also called super-resolution processing.

  However, high resolution processing involves matrix operations with a heavy processing load such as eigenvalue expansion and inverse matrix calculation. For this reason, in a signal processing apparatus that requires real-time processing with a short time limit or that requires as little computation resources as possible, reduction of the amount of processing becomes an issue.

  In order to solve this problem, there is a method of calculating desired specifications by performing high-resolution processing only in a multi-wave environment, and performing other calculations in other cases (for example, see Patent Document 1). ).

R. O. Schmidt, “Multiple Emitter Location and Signal Parameter Estimation”, IEEE Trans, vol.AP-34, No.3, pp.276-280, March, 1986 R. Roy and T. Kailath, “ESPRIT-Estimation of Signal Parameters via Rotational Invariance Techniques”, IEEE Trans, vol.ASSP-37, pp.894-995, Jul. 1989 JP 2006-091029 A

  However, in the method of Patent Document 1, although the number of times of high resolution processing may be reduced, another type of processing is necessary. Also, the amount of processing when high resolution processing is required cannot be reduced.

  The present invention has been made to solve such a problem, and reduces a calculation load required for high resolution processing, and can effectively use resources of an arithmetic unit and a memory and a high resolution signal processing apparatus. An object is to obtain a resolution signal processing method.

  A high-resolution signal processing apparatus according to the present invention uses a desired signal component detection unit that performs Fourier transform on a received signal to detect a desired signal component in a frequency dimension, and a covariance using the signal component detected by the desired signal component detection unit. A covariance matrix generation unit that generates a matrix, and a high resolution angle measurement processing unit that calculates the arrival angle of a desired signal at high resolution by using the covariance matrix generated by the covariance matrix generation unit as input. A high-resolution signal processing device that calculates an angle of an incoming signal, and an angle prediction unit that predicts an arrival angle of a desired signal after a predetermined time based on an arrival angle of the desired signal at a predetermined time, and an angle prediction unit Generated by the covariance matrix prediction unit that predicts the covariance matrix that will be generated by the covariance matrix generation unit after a predetermined time based on the predicted arrival angle, and the covariance matrix generation unit after the predetermined time A covariance matrix match determination unit that performs a match determination between the covariance matrix that has been predicted by the covariance matrix prediction unit at a predetermined time and a covariance matrix after the predetermined time, and a covariance matrix match determination after the predetermined time If it is determined that the two covariance matrices match each other, the angle of arrival after a predetermined time predicted by the angle prediction unit is output without performing the high resolution processing by the high resolution angle measurement processing unit. If it is determined that the covariance matrix does not match, the high resolution angle measurement processing unit performs high resolution processing on the basis of the covariance matrix generated by the covariance matrix generation unit after a predetermined time. And a process selection unit that calculates and outputs the arrival angle of the desired signal.

  Also, the high resolution signal processing method according to the present invention includes a covariance matrix generation step for generating a covariance matrix using a signal component by Fourier-transforming a received signal to detect a desired signal component in the frequency dimension. A high-resolution processing step that calculates the arrival angle of the desired signal, the distance to the source of the incoming signal, or the relative speed of the incoming signal source with its own device at a high resolution with the input covariance matrix as input A high-resolution signal processing method that calculates an arrival angle, distance, or relative speed with high resolution, and based on the arrival angle, distance, or relative speed of the desired signal at a predetermined time, the arrival of the desired signal after the predetermined time A prediction step for predicting an angle, distance, or relative velocity, and a covariance matrix generation step after a predetermined time based on the predicted arrival angle, distance, or relative velocity A covariance matrix prediction step for predicting a covariance matrix that will be formed, a covariance matrix generated by a covariance matrix generation step after a predetermined time, and a predetermined prediction predicted by the covariance matrix prediction step at a predetermined time If it is determined that the two covariance matrices match in the covariance matrix match determination step for performing a match determination with the covariance matrix after the time and the covariance matrix match determination step after a predetermined time, a high resolution process is performed. If the arrival angle, distance, or relative speed after the predetermined time predicted in the prediction step without performing the high-resolution processing by the step is output, and it is determined that the two covariance matrices do not match, By performing high-resolution processing based on the covariance matrix generated in the covariance matrix generation step, the arrival angle, distance, and distance of the desired signal after a predetermined time In which further comprising a processing selection step of outputting to calculate the relative speed.

  According to the present invention, the calculation load required for high resolution processing is reduced by providing a configuration for selectively switching the presence or absence of high resolution processing according to the matching determination processing between the predicted covariance matrix and the actual measurement covariance matrix. It is possible to obtain a high-resolution signal processing apparatus and a high-resolution signal processing method that can reduce and effectively use resources of an arithmetic unit and a memory.

  Hereinafter, preferred embodiments of a high resolution signal processing apparatus and a high resolution signal processing method of the present invention will be described with reference to the drawings.

Embodiment 1 FIG.
FIG. 1 is a configuration diagram of a high-resolution signal processing apparatus according to Embodiment 1 of the present invention. 1 includes a transmission unit 11, a transmission antenna 12, a reception array antenna 13, a reception unit 14, an A / D converter 15, a time Fourier transform unit 16, a spatial Fourier transform unit 17, and a peak detection unit 18. , Velocity calculation unit 19, covariance matrix generation unit 20, covariance matrix match determination unit 21, processing selection unit 22, high resolution processing unit 23, target position prediction unit 24, prediction data storage unit 25, and covariance matrix prediction The unit 26 is configured.

  Further, the transmission unit 11 includes an oscillator 11a, a frequency multiplier 11b, and a low noise amplifier 11c. The receiving unit 14 includes a low noise amplifier 14a, a band pass filter 14b, and a mixer 14c.

  Here, the transmission unit 11 to the speed calculation unit 19 correspond to a desired signal component detection unit. The high resolution processing unit 23 is a general term for a high resolution angle measurement processing unit, a high resolution distance measurement processing unit, and a high resolution speed measurement processing unit, and has a function of obtaining an angle, a distance, and a relative speed with high resolution. is doing. Furthermore, the target position predicting unit 24 corresponds to a generic name of an angle predicting unit, a distance predicting unit, and a speed predicting unit. Based on a measurement result at a predetermined time, desired target positions such as an angle, a distance, and a relative speed are obtained. It has a function to predict the origin.

  Next, the operation of the high resolution signal processing apparatus having the configuration of FIG. 1 will be described. In the first embodiment, a radar apparatus that calculates the arrival angle of a reflected wave by a target of a transmission wave or the distance to the target, which is included in the received signal, using high resolution processing will be described.

  Inside the transmission unit 11, the frequency multiplier 11b multiplies the frequency of the signal generated by the oscillator 11a to generate a transmission wave. The generated transmission wave is amplified in intensity by the low noise amplifier 11 c and emitted to the front of the antenna via the transmission antenna 12.

  The transmission wave emitted from the transmission antenna 12 is reflected by the target when the target is ahead in the traveling direction of the transmission wave. The reflected wave is received by the receiving array antenna 13. The intensity of the received signal is amplified by the low noise amplifier 14 a in the receiving unit 14. The received signal is frequency band limited by the band pass filter 14b. Thereafter, the signal is mixed with the transmission wave by the mixer 14c, thereby generating a frequency signal having a frequency difference between the reception wave and the transmission wave. Further, the generated frequency signal is converted into a digital signal by the A / D converter 15.

  The digitized signal is Fourier-transformed in the time dimension of the data in the time Fourier transform unit 16, and the dimension of the received signal is transformed from the time dimension to the Doppler frequency dimension.

  The spatial Fourier transform unit 17 forms a desired number of beams by Fourier transform in the spatial direction. The beam forming can increase the integration effect of the received signal.

  The peak detection unit 18 performs threshold processing on the power or amplitude intensity of the signal after the spatial Fourier transform, and detects a peak signal.

  The speed calculation unit 19 calculates the speed of the target from the peak Doppler frequency.

  The covariance matrix generation unit 20 generates a covariance matrix using the signal of the Doppler frequency component detected as a peak. Covariance matrices are generated for the number of detected signal peaks.

  The covariance matrix coincidence determination unit 21 is a covariance matrix (measured covariance matrix) generated by the covariance matrix generation unit 20 and a covariance matrix prediction unit 26 and a covariance matrix stored in the prediction data storage unit 25. The variance matrix (predictive covariance matrix) is compared to determine whether the measured covariance matrix and the predicted covariance matrix match. At this time, a prediction covariance matrix having the same Doppler frequency as that of the actual measurement covariance matrix is used.

  As a method of comparing the two covariance matrices, the difference between the measured covariance matrix and the predicted covariance matrix is taken, and if the Frobenius norm | A | of the difference matrix A is equal to or less than a preset value e, it is regarded as a match. , E may be considered as a disagreement. As another method, a method is also conceivable in which the measured covariance matrix and the predicted covariance matrix are correlated, and when the normalized correlation coefficient C is equal to or greater than 1-e, they are regarded as matching.

  The process selection unit 22 outputs the prediction position stored in the prediction data storage unit 25 as a result for the peak determined to match the actual measurement covariance matrix and the prediction covariance matrix. Also, the processing is transferred to the high resolution processing unit 23 for the peak determined to be inconsistent between the actual measurement covariance matrix and the predicted covariance matrix. In the first process after the start of measurement, since the predicted position does not exist, the process is shifted to the high resolution processing unit 23.

  The high resolution processing unit 23 receives the actual measurement covariance matrix generated by the covariance matrix generation unit 20 as input, obtains the distance and angle by a high resolution algorithm (MUSIC or ESPRIT), and outputs the result.

  The target position prediction unit 24 assumes a motion model of the target, and uses the determined position and speed of the target to calculate the position and speed of the target after a predetermined time (ie, distance, angle, speed). Predict. As an assumed motion model, a constant velocity linear motion or a constant acceleration motion can be considered. The term “after a predetermined time” here means a time after a predetermined interval has elapsed since the predetermined time in a calculation cycle for each predetermined interval.

  The position and speed of the target after a predetermined time predicted by the target position prediction unit 24 are stored in the prediction data storage unit 25.

  A reflected wave received after a predetermined time can be predicted from the predicted position and speed of the target. If the received signal is predicted, the generated covariance matrix can be predicted. Therefore, in order to perform such processing, a covariance matrix prediction unit 26 is provided. The predicted covariance matrix predicted by the covariance matrix prediction unit 26 is stored in the prediction data storage unit 25.

  As described above, according to Embodiment 1, when it is determined that the two matrices match by performing the matching determination process between the predicted covariance matrix and the actual measurement covariance matrix, the high resolution process is performed. First, a prediction result calculated simply assuming a motion model of the target is adopted. As a result, the number of high-resolution processes can be reduced and the processing load can be reduced. On the other hand, when the predicted covariance matrix and the measured covariance matrix do not match, high-resolution processing is performed, so that the target distance and angle can be accurately calculated. As a result, it is possible to realize a high-resolution signal processing apparatus and a high-resolution signal processing method that can reduce the calculation load required for high-resolution processing and can effectively use the resources of the calculator and the memory.

Embodiment 2. FIG.
In the second embodiment, a case where a real-time processing control unit (not shown) is further provided in the configuration of the first embodiment will be described. This real-time processing control unit has a function of interrupting the processing in the middle when the high-resolution processing in the high-resolution processing unit 23 based on all the actual measurement covariance matrices cannot be performed within the time limit of the real-time processing. In this case, the prediction result by the target position prediction unit 24 is output. In addition, since there is no prediction result in the first process after the start of measurement, it is assumed that the high resolution process is always executed.

  Further, the real-time processing control unit, when the series of processing shown in the first embodiment can be processed within the time limit of the real-time processing, the peak for which high-resolution processing has not been performed due to matching determination. Also, the high resolution processing unit 23 has a function of causing high resolution processing to be performed. At that time, the high resolution processing unit 23 uses the prediction covariance matrix stored in the prediction data storage unit 25 as an input for the high resolution processing. However, also in this case, when the high resolution processing cannot be performed within the time limit of the real time processing, the processing is interrupted in the middle.

  As described above, according to the second embodiment, in addition to the effect of the first embodiment, the real-time processing control function is further provided, so that high-resolution processing can be executed within a range where real-time processing is possible. Therefore, it is possible to realize a high-resolution signal processing apparatus and a high-resolution signal processing method capable of realizing real-time processing while effectively using computation resources.

Embodiment 3 FIG.
FIG. 2 is a configuration diagram of a high-resolution signal processing apparatus according to Embodiment 3 of the present invention. The high-resolution signal processing apparatus in FIG. 1 includes a reference signal generating unit 10, a receiving array antenna 13, a receiving unit 14, an A / D converter 15, a time Fourier transform unit 16, a spatial Fourier transform unit 17, a peak detection unit 18, In the variance matrix generation unit 20, the covariance matrix match determination unit 21, the process selection unit 22, the high resolution processing unit 23, the prediction data storage unit 25, the covariance matrix prediction unit 26, the calculation result storage unit 27, and the tracking processing unit 28 Composed. Furthermore, the receiving unit 14 includes a low noise amplifier 14a, a band pass filter 14b, and a mixer 14c.

  Here, the reference signal generator 10 to the peak detector 18 correspond to a desired signal component detector. The high resolution processing unit 23 is a general term for a high resolution angle measurement processing unit, a high resolution distance measurement processing unit, and a high resolution speed measurement processing unit, and has a function of obtaining an angle, a distance, and a relative speed with high resolution. is doing. Furthermore, the tracking processing unit 28 corresponds to a generic name of an angle prediction unit, a distance prediction unit, and a speed prediction unit, and calculates desired specifications such as an angle, a distance, and a relative speed after a predetermined time based on a measurement result at a predetermined time. It has a function to predict.

  Next, the operation of the high resolution signal processing apparatus having the configuration of FIG. 2 will be described. The third embodiment relates to an apparatus that receives an incoming wave without emitting a transmission wave from itself, and an apparatus that calculates an angle of arrival of a wave using high resolution processing will be described.

  The intensity of the signal received by the receiving array antenna 13 is amplified by the low noise amplifier 14 a in the receiving unit 14. The received signal is frequency band limited by the band pass filter 14b. Thereafter, the signal is down-converted by mixing with the reference signal of the reference signal generator 10 in the mixer 14 c and converted into a digital signal by the A / D converter 15.

  The digitized signal is Fourier-transformed in the time dimension of the data in the time Fourier transform unit 16, and the dimension of the received signal is transformed from the time dimension to the frequency dimension.

  Spatial Fourier transform unit 17, peak detection unit 18, covariance matrix generation unit 20, and covariance matrix match determination unit 21 perform processing similar to that described in the first embodiment.

  The processing selection unit 22 stores the predicted arrival angle stored in the prediction data storage unit 25 in the calculation result storage unit 27 as a result for the peak determined to match the actual measurement covariance matrix and the prediction covariance matrix. save. Also, the processing is transferred to the high resolution processing unit 23 for the peak determined to be inconsistent between the actual measurement covariance matrix and the predicted covariance matrix. Note that in the first process after the start of measurement, there is no predicted arrival angle, so the process is transferred to the high resolution processing unit 23.

  The high resolution processor 23 receives the measured covariance matrix generated by the covariance matrix generator 20 as an input, obtains the arrival angle of the signal by a high resolution algorithm (MUSIC or ESPRIT), and stores it in the calculation result storage unit 27.

  The tracking processing unit 28 predicts the signal arrival angle after a predetermined time by tracking processing using the angle measurement result stored in the calculation result storage unit 27. The signal arrival angle predicted by the tracking processing unit 28 is stored in the prediction data storage unit 25.

  The covariance matrix prediction unit 26 predicts a covariance matrix that will be generated by processing after a predetermined time based on the prediction angle stored in the prediction data storage unit 25, and stores it in the prediction data storage unit 25. To do.

  As above. According to Embodiment 3, the storage unit for storing the angle measurement result is provided, and the angle after a predetermined time is predicted by performing tracking processing using the stored data, so that the prediction accuracy of the covariance matrix is increased. It is possible.

  In the first to third embodiments described above, each component of the reference signal generation unit 10, the transmission unit 11, the transmission antenna 12, the reception array antenna 13, the reception unit 14, and the A / D converter 15 is in the radio wave region. Shown as being conscious of the use of. However, the present invention is not limited to such a configuration. Instead, a so-called camera such as light or infrared rays or an electromagnetic wave observation device of other frequencies may be used, and the same effect can be obtained. .

It is a block diagram of the high-resolution signal processing apparatus in Embodiment 1 of this invention. It is a block diagram of the high-resolution signal processing apparatus in Embodiment 3 of this invention.

Explanation of symbols

  10 Reference signal generator, 11 Transmitter, 11a Transmitter, 11b Frequency multiplier, 11c Low noise amplifier, 12 Transmit antenna, 13 Receive array antenna, 14 Receiver, 14a Low noise amplifier, 14b Band pass filter, 14c Mixer, 15 A / D converter, 16 time Fourier transform unit, 17 spatial Fourier transform unit, 18 peak detection unit, 19 speed calculation unit, 20 covariance matrix generation unit, 21 covariance matrix match determination unit, 22 process selection unit, 23 High resolution processing unit (high resolution angle measurement processing unit, high resolution distance measurement processing unit, high resolution speed measurement processing unit), 24 target position prediction unit (angle prediction unit, distance prediction unit, and speed prediction unit), 25 prediction Data storage unit, 26 covariance matrix prediction unit, 27 calculation result storage unit, 28 tracking processing unit (angle prediction unit, distance prediction unit, and speed prediction unit) Part).

Claims (5)

A desired signal component detection unit for Fourier-transforming the received signal and detecting a desired signal component in the frequency dimension;
A covariance matrix generation unit that generates a covariance matrix using the signal component detected by the desired signal component detection unit;
A high-resolution angle measurement processing unit that calculates the arrival angle of a desired signal with high resolution using the covariance matrix generated by the covariance matrix generation unit as an input, and is configured to calculate the angle of a signal arriving with high resolution. A resolution signal processing apparatus comprising:
An angle prediction unit that predicts the arrival angle of the desired signal after a predetermined time based on the arrival angle of the desired signal at a predetermined time;
A covariance matrix prediction unit that predicts a covariance matrix that will be generated by the covariance matrix generation unit after the predetermined time based on the angle of arrival predicted by the angle prediction unit;
A covariance matrix that performs a coincidence determination between the covariance matrix generated by the covariance matrix generation unit after a predetermined time and the covariance matrix after the predetermined time predicted by the covariance matrix prediction unit at the predetermined time A match determination unit;
After a predetermined time, when the covariance matrix coincidence determination unit determines that the two covariance matrices match, the angle prediction unit predicts the high resolution angle measurement unit without performing high resolution processing. If the angle of arrival after the predetermined time is output and it is determined that the two covariance matrices do not match, based on the covariance matrix generated by the covariance matrix generation unit after the predetermined time. A high-resolution signal processing further comprising: a processing selection unit that calculates and outputs the arrival angle of the desired signal after the predetermined time by causing the high-resolution angle measurement processing unit to perform high-resolution processing. apparatus. A desired signal component detection unit for Fourier-transforming the received signal and detecting a desired signal component in the frequency dimension;
A covariance matrix generation unit that generates a covariance matrix using the signal component detected by the desired signal component detection unit;
A high-resolution ranging processor that calculates the distance to the source of a signal that arrives at a high resolution using the covariance matrix generated by the covariance matrix generator as an input, and generates a signal that arrives at a high resolution A high-resolution signal processing device that calculates a distance to a source,
A distance predicting unit that predicts a distance to a desired signal generation source after a predetermined time based on a distance to the desired signal generation source at a predetermined time;
A covariance matrix prediction unit that predicts a covariance matrix that will be generated by the covariance matrix generation unit after the predetermined time based on the distance predicted by the distance prediction unit;
A covariance matrix that performs a coincidence determination between the covariance matrix generated by the covariance matrix generation unit after a predetermined time and the covariance matrix after the predetermined time predicted by the covariance matrix prediction unit at the predetermined time A match determination unit;
After a predetermined time, when the covariance matrix coincidence determination unit determines that the two covariance matrices match, the distance prediction unit predicts the high resolution range processing unit without performing high resolution processing. The distance to the generation source of the desired signal after the predetermined time is output, and when the two covariance matrices are determined to be inconsistent, the covariance matrix generation unit is generated after the predetermined time. A processing selection unit that calculates and outputs a distance to the source of the desired signal after the predetermined time by causing the high-resolution ranging processing unit to perform high-resolution processing based on the covariance matrix. A high-resolution signal processing apparatus. A desired signal component detection unit for Fourier-transforming the received signal and detecting a desired signal component in the frequency dimension;
A covariance matrix generation unit that generates a covariance matrix using the signal component detected by the desired signal component detection unit;
A high-resolution speed-measurement processing unit that calculates the relative speed of the source of the signal arriving at high resolution with the covariance matrix generated by the covariance matrix generation unit as input. A high-resolution signal processing device that calculates the relative speed of the signal generation source with the own device,
A speed prediction unit that predicts a relative speed of the desired signal generation source after a predetermined time based on a relative speed of the desired signal generation source at a predetermined time;
A covariance matrix prediction unit that predicts a covariance matrix that will be generated by the covariance matrix generation unit after the predetermined time based on the relative speed predicted by the speed prediction unit;
A covariance matrix that performs a coincidence determination between the covariance matrix generated by the covariance matrix generation unit after a predetermined time and the covariance matrix after the predetermined time predicted by the covariance matrix prediction unit at the predetermined time A match determination unit;
After a predetermined time, when the covariance matrix match determination unit determines that the two covariance matrices match, the speed prediction unit predicts the high resolution speed measurement processing unit without performing high resolution processing. And outputting the relative velocity after the predetermined time and determining that the two covariance matrices do not match, based on the covariance matrix generated by the covariance matrix generation unit after the predetermined time. And a processing selection unit that causes the high-resolution speed measurement processing unit to perform high-resolution processing to calculate and output the relative speed of the desired signal generation source after the predetermined time. A high-resolution signal processing device. In the high resolution signal processing device according to any one of claims 1 to 3,
If high-resolution processing cannot be completed for all covariance matrices generated by the covariance matrix generation unit within the time limit for real-time processing, the processing is aborted, while the covariance matrix matches When all the processing is completed within the time limit without performing high resolution processing due to the coincidence determination by the determination unit, high resolution processing of the covariance matrix that has not been subjected to high resolution processing is limited to the real-time processing. A high-resolution signal processing apparatus, further comprising a real-time processing control unit that performs the remaining time within the time. A covariance matrix generating step of detecting a desired signal component in a frequency dimension by Fourier-transforming the received signal and generating a covariance matrix using the signal component;
High resolution processing step of calculating the arrival angle of the desired signal, the distance to the source of the arriving signal, or the relative speed of the arriving signal source with its own device with the generated covariance matrix as input A high-resolution signal processing method for calculating the angle of arrival, the distance, or the relative velocity with high resolution,
A prediction step of predicting the arrival angle, distance, or relative speed of the desired signal after a predetermined time based on the arrival angle, distance, or the relative speed of the desired signal at a predetermined time;
A covariance matrix prediction step for predicting a covariance matrix that will be generated by the covariance matrix generation step after the predetermined time based on the predicted arrival angle, the distance, or the relative velocity;
A covariance matrix that performs coincidence determination between the covariance matrix generated in the covariance matrix generation step after a predetermined time and the covariance matrix after the predetermined time predicted in the covariance matrix prediction step at the predetermined time A match determination step;
After a predetermined time, when it is determined that the two covariance matrices match in the covariance matrix match determination step, the predetermined prediction predicted in the prediction step without performing the high resolution processing in the high resolution processing step. The arrival angle after the time, the distance, or the relative velocity is output, and when the two covariance matrices are determined to be inconsistent, the covariance matrix generation step generated after the predetermined time And a processing selection step of calculating and outputting an arrival angle, a distance, or a relative speed of the desired signal after the predetermined time by performing high resolution processing based on a covariance matrix. Processing method.

Images