Direction finding device, program, and direction finding method

Abstract

Provided is an azimuth detection device (100) comprising: a signal reception unit (110) that receives a plurality of incoming signals over a wide band in an antenna array in which a plurality of antennas are arranged at unequal intervals; a signal separation detection unit (121) that separates the plurality of incoming signals, detects each of the plurality of incoming signals, and outputs a frequency component of each of the plurality of incoming signals; a frequency component addition unit (122) that adds a frequency component to each of the plurality of incoming signals while correcting a phase shift due to the frequency difference, using the specific frequency as a reference; and a single azimuth estimation unit (123) that estimates the azimuth of a single incoming signal included in the plurality of incoming signals from the added frequency component.

Description

[Technical Field]

[0001] The present disclosure relates to a direction finding device, a program, and a direction finding method. [Background technology]

[0002] 2. Description of the Related Art Direction finders that use non-uniformly spaced antenna arrays are known as direction finders that estimate the directions of arrival of a plurality of signals contained in a wide frequency band. Even if the signal is included in a wide frequency band, if it is a narrowband signal, which is a signal with a narrow frequency band, the direction of arrival of the signal can be estimated from the phase relationship of the signal received by each antenna of the unequally spaced antenna array, in other words, the phase difference between the antennas, and the signal frequency.

[0003] However, in the case of a wideband signal, which is a signal with a wide frequency band, the phase difference between the signals received by each antenna changes for different frequency components within the signal band, so it is necessary to correct this phase shift before estimating the direction of arrival.

[0004] On the other hand, when estimating the direction of arrival, not only the frequency band width but also the surrounding environment of the direction finder is an important factor to consider when selecting an estimation method. This is because if there is a reflecting object near the antenna array, close-range multipath signals are generated, which has a significant impact on the accuracy of the direction of arrival estimation.

[0005] Even in an environment where nearby multipath signals highly correlated with a direct wave signal occur, technologies have been developed that improve the accuracy of estimating the direction of arrival by reducing the correlation between signals using phase fluctuations due to differences in frequency within the signal band. For example, the estimation technology described in Patent Document 1 can improve the accuracy of estimating the direction of arrival by reducing the correlation between signals regardless of the frequency bandwidth of the signals.

[0006] Specifically, in Patent Document 1, since the phase difference between signals received by each antenna changes depending on the frequency components of the signal, a correlation matrix is ​​generated using a focus matrix that corrects the phase shift due to frequency differences, assuming that the number of arriving signals and an initial solution for their arrival directions are obtained. Then, after reducing the correlation between signals using the correlation matrix, the arrival direction is estimated using a method such as MUSIC (MUltiple SIgnal Classification). [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Application Laid-Open No. 2010-266293 Summary of the Invention [Problem to be solved by the invention]

[0008] However, in a real environment, obtaining the number of incoming signals and the initial solution for their arrival directions, which are prerequisites for generating the correlation matrix, can be a challenge. Also, singular value decomposition is required to calculate the focus matrix for each frequency, and the amount of calculation required becomes an issue, especially when the correlation matrix is ​​repeatedly calculated to improve estimation accuracy.

[0009] Therefore, one or more aspects of the present disclosure aim to receive multiple wideband signals using an unequally spaced antenna array in an environment where multipath signals occur, separate and detect each signal in the time-frequency domain, and then estimate the direction of arrival of a direct wave signal with a small amount of calculation, even if the detected signal contains multipath waves that are difficult to separate in the time-frequency domain. [Means for solving the problem]

[0010] An orientation finding device according to a first aspect of the present disclosure is characterized by comprising: a signal receiving unit that receives a plurality of incoming signals over a wide band using an antenna array in which a plurality of antennas are arranged at unequally spaced intervals; a signal separation and detection unit that separates the plurality of incoming signals, detects each of the plurality of incoming signals, and outputs the frequency components of each of the plurality of incoming signals; a frequency component addition unit that adds the frequency components of each of the plurality of incoming signals while correcting for phase shifts due to frequency differences using a specific frequency as a reference; and a single orientation estimation unit that estimates the orientation of a single incoming signal included in the plurality of incoming signals from the added frequency components.

[0011] A direction finding device according to a second aspect of the present disclosure is characterized by comprising: a signal receiving unit that receives a plurality of incoming signals over a wide band using an antenna array in which a plurality of antennas are arranged at unequally spaced intervals; a signal separation and detection unit that separates the plurality of incoming signals, detects each of the plurality of incoming signals, and outputs the frequency components of each of the plurality of incoming signals; a frequency component addition unit that adds the frequency components of each of the plurality of incoming signals while correcting for phase shifts due to frequency differences using a specific frequency as a reference; a multiple direction candidate calculation unit that calculates a correlation matrix from the frequency components and calculates direction candidates of the plurality of incoming signals; and a maximum strength signal direction estimation unit that estimates the direction of an incoming signal with the maximum signal strength from the added frequency components and the direction candidates of the plurality of incoming signals.

[0012] A program according to a first aspect of the present disclosure causes a computer to function as a signal separation and detection unit that separates a plurality of incoming signals over a wide band received by an antenna array in which a plurality of antennas are arranged at unequally spaced intervals, detects each of the plurality of incoming signals, and outputs the frequency components of each of the plurality of incoming signals; a frequency component addition unit that adds the frequency components of each of the plurality of incoming signals while correcting for phase shifts due to frequency differences based on a specific frequency; and a single-direction estimation unit that estimates the direction of a single incoming signal included in the plurality of incoming signals from the added frequency components.

[0013] A program according to a second aspect of the present disclosure causes a computer to function as a signal separation and detection unit that separates multiple incoming signals over a wide band received by an antenna array in which multiple antennas are arranged at uneven intervals, detects each of the multiple incoming signals, and outputs the frequency components of each of the multiple incoming signals; a frequency component addition unit that adds the frequency components of each of the multiple incoming signals while correcting phase shifts due to frequency differences based on a specific frequency; a multiple orientation candidate calculation unit that calculates a correlation matrix from the frequency components and calculates orientation candidates of the multiple incoming signals; and a maximum strength signal orientation estimation unit that estimates the orientation of an incoming signal with the maximum signal strength from the added frequency components and the orientation candidates of the multiple incoming signals.

[0014] A direction finding method according to a first aspect of the present disclosure includes receiving a plurality of incoming signals over a wide band using an antenna array in which a plurality of antennas are arranged at unequally spaced intervals, separating the plurality of incoming signals, detecting each of the plurality of incoming signals, outputting frequency components of each of the plurality of incoming signals, adding the frequency components of each of the plurality of incoming signals while correcting for phase shifts due to frequency differences using a specific frequency as a reference, and estimating the direction of a single incoming signal included in the plurality of incoming signals from the added frequency components.

[0015] A direction finding method according to a second aspect of the present disclosure includes receiving a plurality of incoming signals over a wide band using an antenna array in which a plurality of antennas are arranged at unequally spaced intervals, separating the plurality of incoming signals, detecting each of the plurality of incoming signals, outputting the frequency components of each of the plurality of incoming signals, adding the frequency components of each of the plurality of incoming signals while correcting for phase shifts due to frequency differences using a specific frequency as a reference, calculating a correlation matrix from the frequency components, calculating direction candidates of the plurality of incoming signals, and estimating the direction of the incoming signal with the greatest signal strength from the added frequency components and the direction candidates of the plurality of incoming signals. [Effects of the Invention]

[0016] According to one or more aspects of the present disclosure, a plurality of wideband signals are received by an unequally spaced antenna array in an environment where multipath signals occur, and each signal is separated and detected in the time-frequency domain. Even if the detected signals contain multipath waves that are difficult to separate in the time-frequency domain, the direction of arrival of the direct wave signal can be estimated with a small amount of calculation. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a block diagram schematically illustrating the configuration of a direction finding device according to a first embodiment. [Figure 2] 10 is a graph plotting, on a complex plane, an example of frequency components of a signal received by a four-element antenna array and separated and detected by a signal separation and detection unit. [Figure 3] 10 shows a graph plotting on a complex plane an example of frequency components of a signal whose phase shift has been corrected by a frequency component adder before addition. [Figure 4] FIG. 2 is a block diagram illustrating an example of a hardware configuration. [Figure 5] 4 is a flowchart showing the operation of the control unit in the first embodiment. [Figure 6] FIG. 10 is a block diagram showing a schematic configuration of a direction finding device according to a second embodiment. [Figure 7] FIG. 10 is a schematic diagram showing an example of an angular spectrum calculated by receiving two incoming signals and using two eigenvectors. [Figure 8] 10 is a flowchart showing the operation of a control unit in the second embodiment. [Figure 9] FIG. 11 is a block diagram showing a schematic configuration of a direction finding device according to a third embodiment. [Figure 10] 11 is a flowchart showing the operation of a control unit in the third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0018] Embodiment 1 FIG. 1 is a block diagram showing a schematic configuration of a direction finding device 100 according to the first embodiment. The direction finding device 100 includes a signal receiving unit 110 and a control unit 120 .

[0019] The signal receiving unit 110 receives a plurality of incoming signals over a wide band using an antenna array in which a plurality of antennas are arranged at uneven intervals.

[0020] First, the phase of the incoming signal received by the non-uniformly spaced antenna array will be explained. A first antenna, which is one of the antennas included in the plurality of antennas, is set as a reference antenna, and the phase of the frequency component of the frequency kΔf (k is a natural number satisfying k=1, . . . , K) of the signal arriving from the direction θ and received by the first antenna is defined as ψ k Then, the element spacing between the first antenna is d m,1 The phase φ of the frequency component of the signal received at the mth antenna (m is a natural number satisfying m=1, . . . , M) is m,k,θ is expressed by the following equation (1).

number

[0021] The first term on the right side of equation (1) corresponds to the phase difference between antennas, and from this first term, we can see that it changes linearly with frequency kΔf, and that in the case of a wideband signal, the phase difference between antennas changes at different frequencies. Also, from the first term on the right side of equation (1), we can see that the phase difference between antennas changes linearly with the element spacing d m,1 and the sine of the signal arrival direction, sin θ, also change linearly. The phase ψ of the frequency component of frequency kΔf k It should be noted that this varies depending on the reception timing and signal specifications.

[0022] The control unit 120 estimates the direction of the incoming signal received by the signal receiving unit 110 . The control unit 120 includes a signal separation detection unit 121 , a frequency component addition unit 122 , and a single direction estimation unit 123 .

[0023] The signal separation and detection unit 121 separates the multiple incoming signals, detects each of the multiple incoming signals, and outputs the frequency components of each of the multiple incoming signals. For example, the signal separation detector 121 separates the incoming signals received by the signal receiver 110 in the time-frequency domain and detects the separated signals. The signal separation and detection unit 121 in this embodiment can use the signal separation and detection method described in the following document 1.

[0024] Reference 1: Takafumi Nagano, Wataru Tsujita, "Real-time pulse detection method from wideband signals," IEICE Technical Report, SANE2021-3, pp.11-14, May 2021

[0025] The method described in Document 1 divides a wide frequency band into multiple subbands using a filter bank, and assigns a signal band detected in the frequency domain to a signal detected in the time domain in each subband, repeating this process for each processing unit time, thereby separating multiple signals in the time-frequency domain and detecting the separated signals. Here, since a set of frequency components of the signal is calculated when the signal band is detected, this set of frequency components can be used in the frequency component adder 122.

[0026] It is assumed that the signal separation detection unit 121 of the first embodiment separates signals transmitted from different transmission sources in the time-frequency domain, assigns a signal band of a signal transmitted from one transmission source to the detected signals, and calculates a set of frequency components corresponding to the signal band. The signal transmitted from one transmission source may contain multipath signals in some cases.

[0027] Furthermore, the signal separation detection unit 121 may use another method as long as it can separate multiple signals and detect the separated signals. Even if the method does not calculate the set of frequency components of the detected signals, the set of frequency components may be calculated after signal detection.

[0028] The frequency component adder 122 adds the frequency components of each of the multiple incoming signals while correcting for phase shifts due to frequency differences, using a specific frequency as a reference. For example, the frequency component adder 122 adds the frequency components of the signals separated and detected by the signal separation detector 121 while correcting for phase shifts due to frequency differences using a specific frequency as a reference. A specific explanation will be given below. Note that although the signals separated and detected by the signal separation detector 121 may contain multipath signals, processing is performed assuming that the direct wave signal is the signal with the greatest intensity and that the separated and detected signals contain only the direct wave signal.

[0029] The above equation (1) is based on a specific reference frequency k ref Using Δf, it can be transformed into the following equation (2).

number

[0030] The second and third terms on the right side of equation (2) are the phase shift due to the frequency difference. More specifically, the second term is the phase shift due to the difference in wavelength, and the third term is the phase shift due to the phase difference between frequency components that differ for each frequency due to the signal characteristics.

[0031] When adding frequency components, it is necessary to correct the phase shift between these two components. The phase observation value W1 shown in the following equation (3) of the frequency component of the frequency kΔf of the signal arriving from the direction θ and received by the mth antenna is obtained from the above equation (2) by ref Using the phase observation value W2 of the frequency component of Δf, it is expressed by the following equation (5).

number

[0032] If the matrix on the left side of the second term on the right side of equation (5) is A, then the pseudo-inverse matrix A of A is + =(A T A) -1 A T Using the least squares solution (Δφ k ,ψ k ) T can be obtained by the following equation (6).

number

[0033] The frequency component z of the signal received by the mth antenna at frequency kΔf m,k The method of adding while correcting the phase shift is to use the least squares solution (Δφ k ,ψ k ) T Using this, it is expressed by the following equation (7).

number

[0034] By adding frequency components while correcting the phase shift in this manner, the frequency component adder 122 can improve the S / N (Signal / Noise) ratio of the frequency components and reduce the influence of multipath signals if they are included.

[0035] The specific reference frequency k refΔf may be the frequency having the largest frequency component at which the error in the calculation of equation (6) is statistically small, or may be the center frequency of the signal band or any other arbitrary frequency.

[0036] FIG. 2 shows a graph plotting, on a complex plane, an example of frequency components of a signal received by a four-element antenna array and separated and detected by signal separation and detection section 121. FIG. 3 shows a graph plotting, on a complex plane, an example of frequency components of a signal whose phase shift has been corrected by frequency component adder 122 before addition. In these figures, four points connected by lines represent frequency components of the same frequency of the signal received by the four elements.

[0037] The single direction estimation unit 123 estimates the direction of a single incoming signal included in the multiple incoming signals from the added frequency components. For example, the single direction estimation unit 123 estimates the direction of a single incoming signal from the frequency components added by the frequency component addition unit 122 by digital beam forming (hereinafter referred to as DBF).

[0038] Specifically, the single-direction estimator 123 calculates the array output Z(θ) of the following equation (8) using a beamformer method so as not to calculate the product component of the direct wave signal and the multipath signal in processing equivalent to calculating a phase difference, such as the product of complex conjugates, and detects the direction in which the amplitude is maximum.The single-direction estimator 123 then outputs the detected direction as the direction of arrival.In other words, the single-direction estimator 123 estimates the direction of a single arriving signal using the beamformer method.

number

[0039] In calculating the array output, if the element spacing of the non-uniform antenna array is an integer ratio, a fast Fourier transform (FFT) that can be performed at high speed may be used.

[0040] q m , q1:q2:···:qM =d 1,1 :d 2,1 :···:d M,1 The single direction estimation unit 123 calculates the element q m +1, the length q having the value W3 shown in the following equation (9) M By generating a vector of +1 and applying an FFT of length L, the array output for direction W4 shown in the following equation (10) can be quickly calculated. Note that Δd is expressed by the following equation (11), and λ is expressed by the following equation (12). Note that when calculating the array output, the single-direction estimation unit 123 may multiply the output of each element by a weight to improve directivity.

number

[0041] The direction where the amplitude of the array output is maximum is determined by the peak of the amplitude of the array output (θ peak ,Z(θ peak Alternatively, a curve passing through the peak and its surrounding points may be calculated, and the peak position may be estimated as a real number. For example, the single-orientation estimation unit 123 calculates a parabola passing through three points, including the peak and two points on either side of the peak, and estimates the horizontal axis position of the focus of the parabola as the peak position. The curve may not be a parabola, but may be any curve having a maximum point.

[0042] The direction finder 100 described above can be configured with hardware such as that shown in FIG. As shown in FIG. 4, the direction finder 100 includes an antenna 1, a filter 2, a mixer 3, a high-frequency signal generating circuit 4, an amplifier 5, an A / D converter 6, and a signal processing circuit 7.

[0043] Each mixer 3 has the function of filtering out unwanted signals after signal mixing. The A / D converter 6 and the signal processing circuit 7 are connected to a common reference signal generating circuit and operate in synchronization with each other.

[0044] The signal receiving section 110 shown in FIG. 1 is a section that receives an incoming signal, and is made up of an antenna 1 , a filter 2 , a mixer 3 , a high frequency signal generating circuit 4 , an amplifier 5 and an A / D converter 6 .

[0045] A plurality of incoming signals over a wide band are received by an antenna array in which a plurality of antennas 1 are arranged at uneven intervals. The received signals are then input to a mixer 3 after filtering out undesired components by a filter 2.

[0046] In the mixer 3, the signal filtered by the filter 2 is mixed with the high-frequency signal generated by the high-frequency signal generating circuit 4, whereby the signal is frequency converted and a signal from which undesired components have been filtered out is output.

[0047] The signal output from the mixer 3 is amplified by an amplifier 5 and converted into a digital signal by an A / D converter 6 .

[0048] The control unit 120 shown in FIG. Although not shown, the signal processing circuit 7 may be configured with, for example, a memory and a processor such as a CPU (Central Processing Unit) that executes a program stored in the memory. Such a program may be provided via a network or may be provided by being recorded on a recording medium. That is, such a program may be provided as, for example, a program product. In other words, the control unit 120 can be configured by a so-called computer.

[0049] Furthermore, the signal processing circuit 7 can also be configured, for example, as a single circuit, a composite circuit, a processor operated by a program, a parallel processor operated by a program, an ASIC (Application Specific Integrated Circuit), or an FPGA (Field Programmable Gate Array). As described above, the signal processing circuit 7 can be realized by a processing circuit network.

[0050] FIG. 5 is a flowchart showing the operation of the control unit 120 in the first embodiment. First, the signal separation detection unit 121 divides a wide frequency band into a plurality of subbands using a filter bank, and assigns a signal band detected in the frequency domain to a signal detected in the time domain in each subband, repeating this process for each processing unit time, thereby separating the plurality of signals and detecting the separated signals (S10). When detecting the signal band, the signal separation detection unit 121 calculates a set of frequency components of the signal.

[0051] Next, the frequency component adder 122 adds the frequency components of the signal while correcting the phase shift due to the frequency difference with a specific frequency as a reference (S11). The phase shift correction can be performed with a small amount of calculation by using a pseudo-inverse matrix calculated in advance.

[0052] Next, the single direction estimation unit 123 performs DBF processing on the added frequency components (S12). Next, the single direction estimation unit 123 detects the direction in which the amplitude is maximum from the output of the DBF processing (S13). Then, the single direction estimation unit 123 estimates the direction in which the amplitude is maximum as the direction of arrival (S14).

[0053] As described above, according to the direction finding device 100 of the first embodiment, the frequency component adder 122 adds frequency components while correcting the phase shift due to the frequency difference using a specific frequency as a reference, so that signals other than the maximum intensity signal are suppressed, and the single direction estimation unit 123 can accurately estimate the arrival direction of the maximum intensity signal.

[0054] Embodiment 2 FIG. 6 is a block diagram showing a schematic configuration of a direction finding device 200 according to the second embodiment. The direction finding device 200 includes a signal receiving unit 110 and a control unit 220 . The signal receiving unit 110 of the direction finding device 200 according to the second embodiment is similar to the signal receiving unit 110 of the direction finding device 100 according to the first embodiment.

[0055] The control unit 220 estimates the direction of the incoming signal received by the signal receiving unit 110 . The control unit 220 includes a signal separation detection unit 121 , a frequency component addition unit 122 , a multiple direction candidate calculation unit 224 , and a maximum strength signal direction estimation unit 225 . The signal separation detection unit 121 and the frequency component addition unit 122 of the control unit 220 in the second embodiment are the same as the signal separation detection unit 121 and the frequency component addition unit 122 of the control unit 120 in the first embodiment.

[0056] The multiple direction candidate calculation unit 224 calculates a correlation matrix from the frequency components calculated by the signal separation detection unit 121, and calculates direction candidates of multiple incoming signals.

[0057] For example, the multiple direction candidate calculation unit 224 calculates the frequency component z of the frequency kΔf of the signal received by the mth antenna. m,k The correlation matrix is ​​calculated from the matrix Z shown in the following equation (13), which has the following elements, using the following equation (14).

number

[0058] Next, the multiple direction candidate calculation unit 224 performs eigenvalue decomposition on the calculated correlation matrix, and calculates an angular spectrum by the MUSIC (MUltiple SIgnal Classification) method, which is one of the subspace methods that are existing methods for estimating the direction of arrival of narrowband signals, using eigenvectors corresponding to a predetermined number of eigenvalues ​​in ascending order from the largest. In other words, the multiple direction candidate calculation unit 224 first calculates an angular spectrum to be used in calculating multiple direction candidates using the MUSIC method.

[0059] The predetermined number of eigenvectors is set to two as a standard because, when nearby multipath signals are included, it is difficult to estimate the number of incoming signals, and even if the number of incoming signals can be estimated and the same number of eigenvectors are used to calculate the angular spectrum, the peak of the direct wave signal of the strongest signal is not necessarily the largest, and further, peaks may occur in directions other than the actual arrival direction, so the multiple direction candidate calculation unit 224 calculates multiple direction candidates including the arrival direction of the direct wave signal, whether or not multipath signals are included.

[0060] FIG. 7 shows an example of the angular spectrum calculated by receiving two incoming signals and using two eigenvectors. As shown in FIG. 7, many peaks can be seen in the angular spectrum.

[0061] In an environment where many multipath signals occur, the number of eigenvectors may be increased. Also, singular value decomposition may be used instead of eigenvalue decomposition.

[0062] Next, a method for selecting direction candidates of an incoming signal from the calculated angular spectrum in the multiple direction candidate calculation unit 224 will be described.

[0063] First, the multiple direction candidate calculation unit 224 selects a peak having a large peak value in the calculated angle spectrum using a predetermined method. The peaks selected as orientation candidates are V n (n is a natural number satisfying n=1, . . . , N), it may be from the maximum peak to the peak immediately preceding where the peak value ratio W5 shown in the following equation (15) changes the most, or from the maximum peak to the peak immediately preceding where the peak value changes the most. In other words, the multiple direction candidate calculation unit 224 calculates the directions θ corresponding to the peaks in the order determined by a predetermined method from the maximum peak. n(n=1, ,i) is output as a candidate direction of the arriving signal.

number

[0064] The maximum strength signal direction estimation unit 225 estimates the direction of the incoming signal with the maximum signal strength from the frequency components added by the frequency component addition unit 122 and the multiple direction candidates of the incoming signals calculated by the multiple direction candidate calculation unit 224.

[0065] For example, the maximum signal direction estimator 225 uses the above equation (8) to determine the direction candidate θ of the incoming signal. n The array output Z(θ n ) and calculate the direction θ where the amplitude is maximum. n is output as the direction of arrival.

[0066] FIG. 8 is a flowchart showing the operation of the control unit 220 in the second embodiment. First, the signal separation detection unit 121 divides a wide frequency band into a plurality of subbands using a filter bank, and assigns a signal band detected in the frequency domain to a signal detected in the time domain in each subband, repeating this process for each processing unit time, thereby separating the plurality of signals and detecting the separated signals (S20). When detecting the signal band, the signal separation detection unit 121 calculates a set of frequency components of the signal.

[0067] Next, the frequency component adder 122 adds the frequency components of the signal while correcting the phase shift due to the frequency difference with a specific frequency as a reference (S21). The phase shift correction can be performed with a small amount of calculation by using a pseudo-inverse matrix calculated in advance.

[0068] On the other hand, the multiple direction candidate calculation unit 224 calculates a correlation matrix from the frequency components calculated by the signal separation detection unit 121 (S22).

[0069] Next, the multiple direction candidate calculation unit 224 performs eigenvalue decomposition on the calculated correlation matrix (S23).

[0070] Next, the multiple direction candidate calculation unit 224 uses the resolved eigenvalues ​​and eigenvectors corresponding to a predetermined number of eigenvalues ​​in ascending order of size to calculate an angular spectrum by the MUSIC method (S24).

[0071] Next, the multiple direction candidate calculation unit 224 selects direction candidates of the incoming signal from among the peaks of the calculated angular spectrum (S25).

[0072] Then, the maximum signal strength direction estimation unit 225 estimates the direction of the incoming signal with the maximum signal strength from the frequency components added by the frequency component addition unit 122 and the multiple incoming signal direction candidates calculated by the multiple direction candidate calculation unit 224 (S26).

[0073] As described above, according to the direction finding device 200 of the second embodiment, the multiple direction candidate calculation unit 224 calculates the angular spectrum using the eigenvectors calculated from the correlation matrix to calculate direction candidates of the arriving signal, so that direction candidates of the arriving signal can be calculated even when there are nearby multipath signals. Also, since the maximum intensity signal direction estimation unit 225 selects the direction in which the amplitude of the array output by DBF is maximum from among the direction candidates of the arriving signal, the direction of the direct wave signal of the maximum intensity signal can be estimated even when the peak of the direct wave signal of the maximum intensity signal is not maximum in the angular spectrum.

[0074] Embodiment 3 FIG. 9 is a block diagram showing a schematic configuration of a direction finding device 300 according to the third embodiment. The direction finding device 300 includes a signal receiving unit 110 and a control unit 320 . The signal receiving unit 110 of the direction finding device 300 according to the third embodiment is similar to the signal receiving unit 110 of the direction finding device 100 according to the first embodiment.

[0075] The control unit 320 estimates the direction of the incoming signal received by the signal receiving unit 110 . The control unit 320 includes a signal separation detection unit 121 , a frequency component addition unit 122 , a single direction estimation unit 123 , a multiple direction candidate calculation unit 224 , a maximum strength signal direction estimation unit 225 , and a signal direction selection unit 326 . The signal separation detection unit 121, frequency component addition unit 122, and single direction estimation unit 123 of the control unit 320 in the third embodiment are the same as the signal separation detection unit 121, frequency component addition unit 122, and single direction estimation unit 123 of the control unit 120 in the first embodiment. The multiple direction candidate calculation unit 224 and maximum intensity signal direction estimation unit 225 of the control unit 320 in the third embodiment are similar to the multiple direction candidate calculation unit 224 and maximum intensity signal direction estimation unit 225 of the control unit 220 in the second embodiment.

[0076] The signal direction selection unit 326 uses a predetermined method to select the direction of the incoming signal from among the directions estimated by the single direction estimation unit 123 and the maximum signal strength direction estimation unit 225. In other words, the signal direction selection unit 326 selects either the direction of a single incoming signal estimated by DBF or the direction estimated as the direction of the incoming signal with the maximum signal strength from among multiple direction candidates of the incoming signals.

[0077] In the maximum intensity signal direction estimation unit 225 of the second embodiment, when the signal band is wide or when the signal band is narrow but the S / N ratio is high, the probability that a direct wave signal is included in the direction candidates of the multiple incoming signals calculated by the multiple direction candidate calculation unit 224 increases, and therefore the accuracy of the calculated direction is higher than that of the single direction estimation unit 123.

[0078] Therefore, the signal direction selection unit 326 in the third embodiment selects the direction estimated by the maximum signal strength direction estimation unit 225 when the signal bandwidth is wider than a predetermined value or when the S / N ratio is greater than a predetermined value, and selects the direction estimated by the single direction estimation unit 123 in all other cases.

[0079] FIG. 10 is a flowchart showing the operation of the control unit 320 in the third embodiment. First, the signal separation detection unit 121 divides a wide frequency band into a plurality of subbands using a filter bank, and assigns a signal band detected in the frequency domain to a signal detected in the time domain in each subband, repeating this process for each processing unit time, thereby separating the plurality of signals and detecting the separated signals (S30). When detecting the signal band, the signal separation detection unit 121 calculates a set of frequency components of the signal.

[0080] Next, the frequency component adder 122 adds the frequency components of the signal while correcting the phase shift due to the frequency difference with a specific frequency as a reference (S31). The phase shift correction can be performed with a small amount of calculation by using a pseudo-inverse matrix calculated in advance.

[0081] Next, the single direction estimation unit 123 performs DBF processing on the added frequency components (S32). Next, the single direction estimation unit 123 detects the direction in which the amplitude is maximum from the output of the DBF processing (S33). The single direction estimation unit 123 then estimates the direction in which the amplitude is maximum as the direction of arrival (S34). The estimated direction is provided to the signal direction selection unit 326.

[0082] On the other hand, the multiple direction candidate calculation unit 224 calculates a correlation matrix from the frequency components calculated by the signal separation detection unit 121 (S35).

[0083] Next, the multiple direction candidate calculation unit 224 performs eigenvalue decomposition on the calculated correlation matrix (S36).

[0084] Next, the multiple direction candidate calculation unit 224 uses the resolved eigenvalues ​​and eigenvectors corresponding to a predetermined number of eigenvalues ​​in ascending order of size to calculate an angular spectrum by the MUSIC method (S37).

[0085] Next, the multiple direction candidate calculation unit 224 selects direction candidates of the incoming signal from among the peaks of the calculated angular spectrum (S38).

[0086] The maximum signal strength direction estimation unit 225 then estimates the direction of the incoming signal with the maximum signal strength from the frequency components added by the frequency component addition unit 122 and the multiple direction candidates of the incoming signals calculated by the multiple direction candidate calculation unit 224 (S39). The estimated direction is provided to the signal direction selection unit 326.

[0087] The signal direction selection unit 326 selects the direction estimated by the maximum signal strength direction estimation unit 225 when the signal bandwidth is wider than a predetermined value or when the S / N ratio is greater than a predetermined value, and otherwise selects and outputs the direction estimated by the single direction estimation unit 123 (S40).

[0088] As described above, according to the direction finding device 300 of the third embodiment, when the direction estimated by the maximum strength signal direction estimation unit 225 is highly accurate, the signal direction selection unit 326 selects and outputs the direction estimated by the maximum strength signal direction estimation unit 225; otherwise, it selects and outputs the direction estimated by the single direction estimation unit 123. This makes it possible to accurately estimate the direction of the direct wave signal, which is the maximum strength signal. [Explanation of symbols]

[0089] 100, 200, 300 Direction finding device, 110 signal receiving unit, 120, 220, 320 control unit, 121 signal separation detection unit, 122 frequency component addition unit, 123 single direction estimation unit, 224 multiple direction candidate calculation unit, 225 maximum strength signal direction estimation unit, 326 signal direction selection unit.

Claims

1. a signal receiving unit that receives a plurality of incoming signals over a wide band using an antenna array in which a plurality of antennas are arranged at uneven intervals; a signal separation and detection unit that separates the plurality of incoming signals, detects each of the plurality of incoming signals, and outputs a frequency component of each of the plurality of incoming signals; a frequency component adding unit that adds the frequency components of each of the plurality of incoming signals while correcting a phase shift due to a frequency difference based on a specific frequency; a single direction estimation unit that estimates the direction of a single incoming signal included in the plurality of incoming signals from the added frequency components. A direction finding device characterized by:

2. a multiple direction candidate calculation unit that calculates a correlation matrix from the frequency components and calculates direction candidates of the multiple arrival signals; a maximum signal strength direction estimation unit that estimates the direction of the incoming signal with the maximum signal strength from the added frequency components and the plurality of incoming signal strength direction candidates; a signal direction selection unit that selects either the direction of the single incoming signal or the direction of the incoming signal with the maximum signal strength.

2. The direction finder according to claim 1, wherein:

3. The multiple direction candidate calculation unit estimates the direction of the incoming signal with the maximum signal strength using a MUSIC (Multiple Signal Classification) method.

3. The direction finder according to claim 2, wherein:

4. The single direction estimation unit estimates the direction of the single incoming signal using a beamformer method.

4. The direction finder according to claim 1, wherein:

5. a signal receiving unit that receives a plurality of incoming signals over a wide band using an antenna array in which a plurality of antennas are arranged at uneven intervals; a signal separation and detection unit that separates the plurality of incoming signals, detects each of the plurality of incoming signals, and outputs a frequency component of each of the plurality of incoming signals; a frequency component adding unit that adds the frequency components of each of the plurality of incoming signals while correcting a phase shift due to a frequency difference based on a specific frequency; a multiple direction candidate calculation unit that calculates a correlation matrix from the frequency components and calculates direction candidates of the multiple arrival signals; a maximum signal strength direction estimation unit that estimates the direction of the incoming signal with the maximum signal strength from the added frequency components and the direction candidates of the plurality of incoming signals. A direction finding device characterized by:

6. The multiple direction candidate calculation unit estimates the direction of the incoming signal with the maximum signal strength using a MUSIC (Multiple Signal Classification) method.

6. The direction finder according to claim 5, wherein:

7. Computer, a signal separation and detection unit that separates a plurality of incoming signals over a wide band received by an antenna array in which a plurality of antennas are arranged at uneven intervals, detects each of the plurality of incoming signals, and outputs a frequency component of each of the plurality of incoming signals; a frequency component adder that adds the frequency components of each of the plurality of incoming signals while correcting a phase shift due to a frequency difference based on a specific frequency; and a single direction estimation unit that estimates the direction of a single incoming signal included in the plurality of incoming signals from the added frequency components; A program characterized by.

8. Computer, a signal separation and detection unit that separates a plurality of incoming signals over a wide band received by an antenna array in which a plurality of antennas are arranged at uneven intervals, detects each of the plurality of incoming signals, and outputs a frequency component of each of the plurality of incoming signals; a frequency component adding unit that adds the frequency components of each of the plurality of incoming signals while correcting a phase shift due to a frequency difference based on a specific frequency; a multiple direction candidate calculation unit that calculates a correlation matrix from the frequency components and calculates direction candidates of the multiple incoming signals; and and functioning as a maximum strength signal direction estimation unit that estimates the direction of the incoming signal with the maximum signal strength from the added frequency components and the direction candidates of the plurality of incoming signals. A program characterized by.

9. An antenna array with multiple antennas arranged at uneven intervals receives multiple incoming signals over a wide band, Separating the plurality of incoming signals, detecting each of the plurality of incoming signals, and outputting a frequency component of each of the plurality of incoming signals; adding the frequency components of each of the plurality of incoming signals while correcting for phase shifts due to frequency differences based on a specific frequency; and estimating the direction of a single incoming signal included in the plurality of incoming signals from the added frequency components. A direction finding method characterized by:

10. An antenna array with multiple antennas arranged at uneven intervals receives multiple incoming signals over a wide band, Separating the plurality of incoming signals, detecting each of the plurality of incoming signals, and outputting a frequency component of each of the plurality of incoming signals; adding the frequency components of each of the plurality of incoming signals while correcting for phase shifts due to frequency differences based on a specific frequency; Calculating a correlation matrix from the frequency components and calculating direction candidates of the plurality of arrival signals; estimating the direction of the incoming signal with the maximum signal strength from the added frequency components and the direction candidates of the plurality of incoming signals; A direction finding method characterized by:

Images