JP2005315811A - Characteristic optimization method for adaptive array - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily verify (confirm) whether only a desired wave is extracted or not, and whether a set constrained condition is proper or not, in a simulation stage, as to an adaptive array, and to optimize a directive characteristic of the adaptive array. <P>SOLUTION: A data matrix X of a time-serial data is found at first from a reception signal in each reception part of a reception array, a sub-array matrix X<SB>11</SB>is transformed into a column vector x<SB>11</SB>, to calculate a spatial-averaged correlation matrix R<SB>XX</SB>(m) spatial-averaged with weights, an arrival direction of the signal is estimated based on the correlation matrix, the constrained condition is determined based on the arrival direction of the desired wave or an unnecessary wave by a MUSIC method or the like to determine a complex weighted vector W, each of the sub-array matrix X<SB>11</SB>is transformed into the column vector x<SB>11</SB>, a re-arrayed matrix Y is calculated from the each column vector x<SB>11</SB>and the complex weighted vector W, and an arrival direction re-estimating correlation matrix R<SB>YY</SB>is calculated from the re-arrayed matrix Y. The MUSIC method is applied again to the correlation matrix R<SB>YY</SB>. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention comprises a two-dimensional receiving array in which a plurality of receiving units are arranged two-dimensionally, and optimization for continuously updating the directivity of an adaptive array that controls the directivity pattern of the two-dimensional receiving array It is about the method.

  An array in which a plurality of excitation elements are arranged two-dimensionally and all or a part thereof is excited can obtain a desired directivity pattern, and is utilized in various fields by taking advantage of its characteristics ( Non-patent document 1).

  By the way, for example, one of the major problems when receiving data in GPS (Global Positioning System) is the occurrence of multipath (multi-propagation) resulting from reflection of radio waves from a satellite on the building or the ground. When a multipath occurs, the direct wave and the reflected wave interfere with each other, not only reducing the accuracy of position measurement, but also possibly making measurement impossible at all. For example, at an elevation angle of 20 degrees or less, it becomes difficult to separate and detect signals from GPS satellites due to various unnecessary signals and noise from the ground.

  One of the countermeasures by the hardware is arraying the antenna (see Non-Patent Document 2). By arraying, the output of each element antenna is multiplied by an appropriate complex weight, that is, by appropriately controlling the amplitude and phase of the received signal of each element antenna, directivity is provided only in the direct wave arrival direction. .

  Such an array of antennas makes the reception sensitivity of the antenna almost zero with respect to the direction of arrival of the reflected wave (unwanted wave) without reducing the gain of the direction of arrival of the direct wave (desired wave), that is, null. ) Is technically considered to be the most versatile and effective technique.

Regarding the data processing of the array antenna, the applicant of the present application pays attention to the DCMP (Directionally Constrained Minimization of Power) method in which the constraint condition can be freely set for both the unnecessary wave and the desired wave, and the DCMP method is applied to the two-dimensional array antenna. Has been filed (Japanese Patent Application No. 2003-129596) for an adaptive array and a positioning device to which the above is effectively applied.
Nobuyoshi Kikuma, "Adaptive signal processing by array antenna", Science and Technology Publishers, 1998. Supervised by Takashi Yoshida, "Revised Radar Technology", edited by IEICE, Corona, 1996.

  However, for that purpose, it is necessary to grasp in advance the arrival directions of all incident waves and to set appropriate restraint conditions. As an arrival direction estimation method, a two-dimensional extension of the MUSIC (Multiple Signal Classification) method, which is one of high-accuracy spectrum estimation methods, is effective. That is, the arrival directions of the desired wave and the unnecessary wave are estimated by the MUSIC method, an appropriate constraint condition is set based on the obtained information, and the complex weight is calculated by applying the DCMP method.

  However, one new problem arises here. That is, there is a problem in that it is impossible to verify whether or not only the desired wave is extracted because the unnecessary weight is really suppressed by the complex weight obtained by the DCMP method. This is also a problem in that it is impossible to verify whether or not the set constraint conditions are appropriate, and it is not always possible to obtain optimal directivity characteristics.

  Here, the time series data from each element of the array antenna is multiplied by the complex weight obtained by the DCMP method, and the MUSIC method is applied again as it is. The direction-of-arrival estimation results obtained in this way are completely random. This is natural, and if you remember that the DCMP method is based on the standard of “multiplying the time-series data from each element antenna by the complex weight and minimizing its power sum”, you can immediately understand it. . In other words, the MUSIC method and the DCMP method cannot be repeatedly applied as they are for the above purpose.

  Accordingly, an object of the present invention is to make it possible to easily verify at the simulation stage whether or not only the desired wave has been extracted, and whether or not the set constraint conditions are appropriate. It is to provide a method of optimizing.

The adaptive array directivity optimization method of the present invention obtains a data matrix X of time-series data from received signals at each receiving unit of a receiving array in which a plurality of receiving units that receive incoming wave signals are arranged, and converts the subarray matrix X ₁₁ data matrix column vector x ₁₁ calculates a correlation matrix R _XX11, the arrival direction of the correlation matrix space average correlation matrix was weighted spatial averaging of R _xx (m) signal based on Estimating
A complex weight vector W is determined by determining a constraint condition based on the arrival direction of the desired wave and / or unnecessary wave of the signal, each of the subarray matrices X ₁₁ is set to a column vector x ₁₁ , and each column vector and the complex weight are determined. A rearrangement matrix Y obtained by rearranging the inner product y ₁₁ of the vector transposed conjugate W ^H with the subarrayed arrangement order is obtained, a direction-of-arrival reestimation correlation matrix R _YY is obtained from the rearrangement matrix Y, and the spatial average correlation Replacing the matrix R _xx (m) with the arrival direction re-estimation correlation matrix R _YY and re-applying the step of estimating the arrival direction of the signal.

  As the receiving array, the receiving units may be arranged one-dimensionally or two-dimensionally. Further, the receiving unit may be arranged so as to be orthogonal to the cross in a plane.

  The adaptive array directivity optimizing method according to the present invention also includes an integrated delta range (obtained by receiving an incoming signal from a transmission source at each of the plurality of receiving units and tracking the carrier phase of the received signal by the reception). ADR) is obtained, and the value of the integrated delta range is used as the received signal at the receiving unit.

  According to the adaptive array characteristic optimizing method of the present invention, the data matrix in consideration of the complex weight is generated, and the MUSIC method is applied to the data row example. It is possible to verify whether the data is extracted and whether the set constraint condition is appropriate, thereby optimizing the characteristics of the adaptive array.

  According to the adaptive array characteristic optimizing method of the present invention, the directivity in the one-dimensional azimuth direction with respect to the straight line can be optimized by arranging the plurality of receiving units in a one-dimensional array. In addition, by arranging the plurality of receiving units in a two-dimensional plane, the directivity in the two-dimensional azimuth direction with respect to the plane can be optimized.

  According to the adaptive array characteristic optimization method of the present invention, the plurality of receiving units are arranged so as to be orthogonal in a cross shape on a plane, thereby mitigating interference caused by adjacent of the plurality of receiving units, It can also be applied to an adaptive array that is small, light, and low in cost.

  Further, according to the adaptive array characteristic optimization method of the present invention, the integrated delta range (ADR) obtained by tracking the carrier phase of the received signal is obtained, and the value of the integrated delta range is obtained as the received signal at the receiving unit. By so doing, the present invention can be applied to an apparatus in which each receiving unit independently obtains the integrated delta range ADR and estimates the arrival direction of the signal.

An adaptive array to which the adaptive array characteristic optimization method of the present invention is applied and a receiving apparatus using the adaptive array will be described with reference to the drawings.
FIG. 1 is a block diagram showing the overall configuration. Here, 1 is an array antenna in which a plurality of element antennas 1a, 1b... 1m are arranged one-dimensionally or two-dimensionally. A portion indicated by 2 is a complex weight giving unit that gives a complex weight to the received signals of the element antennas 1a to 1m. 21a, 21b,... 21m are amplitude adjusters that adjust the amplitude of the received signal of each element antenna, and 22a, 22b,... 22m are phase adjusters that adjust the phase of the signal of each element antenna. The reception signals of the respective element antennas that have been subjected to the amplitude adjustment and the phase adjustment are added by the adding unit 3 to be obtained as one reception signal Pout.

  The reception signal processing unit 4 receives the reception signals of the respective element antennas 1a, 1b,... 1m, obtains complex weights to be given to the reception signals of the respective element antennas by the method described later, and Corresponding adjustment amounts are given to the amplitude adjuster 21 and the phase adjuster 22, respectively.

  The receiver 5 inputs a received signal received under the predetermined directivity pattern and performs processing for signal reception.

  FIG. 2 shows a two-dimensional arrangement example of the plurality of element antennas. In the case of (A), a total of 25 element antennas of 5 × 5 are arranged at each intersection of the orthogonal matrix. In the example shown in (B), when a 5 × 5 matrix is considered, a total of nine element antennas are arranged in the third row × third example.

  The received signal processing unit 4 in FIG. 1 estimates the arrival direction of a signal by the MUSIC method. In the MUSIC method, a correlation matrix is first created from time series output data of each element of the array antenna. Then, the eigenvalue and eigenvector are calculated. The magnitudes of these eigenvalues are discriminated and classified into eigenvectors belonging to the signal space and eigenvectors belonging to the noise space, and the direction of arrival is estimated using the fact that they are orthogonal to each other. That is, when the eigenvector matrix belonging to the noise space is multiplied by the signal arrival direction vector in the denominator and the angle of the signal arrival direction is scanned, the denominator value becomes zero when the angle matches the actual signal signal arrival direction. A steep peak is obtained by approaching. The direction of arrival of the signal is estimated using this action.

  However, since the desired wave and the multipath unnecessary wave are generally highly correlated, it is difficult to distinguish the eigenvalues of the eigenvectors of the signal and noise. Therefore, the array antenna obtained as described above is divided into subarrays, and a pre-processing called a spatial average (or moving average) of a correlation matrix for each subarray is performed.

  If the arrival directions of the desired wave and the unnecessary wave are obtained by applying the above-described MUSIC method, the constraint condition for applying the DCMP method can be determined. As a result, it is possible to form a combined directivity that has sensitivity only in the desired wave arrival direction and forms a null in the unnecessary wave arrival direction.

  A specific processing procedure when the arrangement of the array antenna is two-dimensional has already been filed in Japanese Patent Application No. 2003-129497. Thereby, the complex weight to be multiplied with the output data of each element antenna can be calculated.

  However, as described above, even if the time series output data from each element antenna is multiplied by such a complex weight and the MUSIC method is applied again, the direction estimation result obtained there is quite random. Become. The MUSIC method and the DCMP method have different purposes of use, and the MUSIC method and the DCMP method cannot be repeatedly applied as they are.

A procedure for solving this problem will be described with reference to FIGS.
3A is a block diagram showing the adaptive array characteristic optimization method processing procedure according to the embodiment of the present invention, FIG. 4 is a diagram showing data at each stage of arithmetic processing, and FIG. 5 is a diagram of the adaptive array. It is a flowchart which shows the process sequence of the characteristic optimization method. As shown in these figures, the array antenna is first divided into subarrays, the correlation matrix of each subarray is averaged, and the arrival direction of the signal is estimated by the MUSIC method. The arrival directions of the desired wave and the unnecessary wave are given as constraint conditions, and complex weights for obtaining a desired directivity pattern are determined by the DCMP method. Each element antenna is subarrayed into a plurality of arrays, each subarray is multiplied by the complex weight, and reconfigured according to the arrangement of the subarrays. This makes it possible to re-estimate the signal arrival direction by the MUSIC method or the like.

  FIG. 3B is a block diagram illustrating a state in which a signal is received with a desired directivity pattern. If the result of the re-estimation of the signal arrival direction is a desired result, the reception signal is multiplied by the complex weight as shown in FIG.

FIG. 4 shows an example in which the output data matrix X of a 5 × 5 square array antenna is divided into subarrays of size 3 × 3. First, as shown in FIG. 4A, it is assumed that a data matrix X from each element antenna is obtained. At this time, the 3 × 3 subarray matrix (only one of them, X ₁₁ is shown in the figure) of all combinations included in the 5 × 5 matrix is rearranged as shown in the figure. To a column vector x, and a partial correlation matrix R _xx is calculated based on the column vector x. Then, the correlation matrix is weighted and the spatial average correlation matrix R _XX (m) = xx ^H (m) is calculated, and the MUSIC method is applied. Here, H represents transposition conjugate. As a result, the arrival directions (the zenith angle θ and the azimuth angle φ) of the desired wave and the unnecessary wave are obtained. Since the constraint condition of the DCMP method can be determined based on the information, the DCMP method is applied to the spatially averaged correlation matrix. As a result, a complex weight vector having the number of elements equal to the number of elements of the vector x is obtained. This is the application of the conventional MUSIC method and DCMP method.

Next, sub-array This is conjugate transpose W ^H of the complex weight vector W, for all combinations of 3 × 3 sub-array matrices in the original data matrix, and a column vector in a manner as shown in FIG. 4 (b) The inner product y with the matrix x is calculated. FIG. 4B shows the case of X ₁₁ which is one of these combinations, and the result of the inner product calculation is y ₁₁ .

  Next, as shown in FIG. 4C, the inner product calculation results are rearranged in the order in which the respective sub-array matrices are extracted from the original data matrix. In this way, a 3 × 3 rearrangement matrix Y in which each element is multiplied by the complex weight vector is calculated.

Then, as shown in FIG. 4 (d), the rearrangement matrix Y is converted into a column vector y having 9 elements, and a 9 × 9 correlation matrix (hereinafter referred to as “arrival direction re-estimation”). It is called “correlation matrix.”) R _YY is calculated. The MUSIC method is applied to this arrival direction re-estimation correlation matrix R _YY . That is, the step of replacing the spatial average correlation matrix R _xx (m) with the arrival direction re-estimation correlation matrix R _YY and re-applying the signal arrival direction is applied again.

The spatial average correlation matrix R _xx (m), which is a calculation target when the complex weight W is determined by the DCMP method, is obtained first. That is, even when the loop shown in FIG. 5 is repeated, R _xx (m) is constant.

  In this way, the MUSIC method can be applied again in consideration of complex weights. Then, from the result obtained by applying the MUSIC method again, the desired wave extraction state and the unnecessary wave suppression state are confirmed. If there is no problem, the optimization process may be terminated. If the desired wave extraction status and unnecessary wave suppression status are incomplete, the above constraint conditions are corrected and the same confirmation is performed by applying the MUSIC method again to obtain the desired optimum directivity pattern. Finding the constraint conditions that can be found.

  As is clear from the above description, the method of the present invention is premised on performing a spatial average of the subarray correlation matrix. However, since the desired wave and the unnecessary wave (especially multipath) are highly correlated with each other, this spatial averaging is substantially essential, and in most cases this assumption is acceptable.

  Although not explicitly shown in FIG. 4, each element of the matrix X, Y and the vector x is sampled N-point discrete data from each element antenna, and is a three-dimensional data array and two-dimensional data, respectively. Is an array.

Next, a simulation result is shown.
First, an arbitrary plurality of sine waves are incident on a 5 × 5 square array antenna in the (θ, φ) coordinate system, and these input waves are added and synthesized as a function of time on each element antenna. . In addition, the simulation was performed assuming one wave each of the desired wave and the unnecessary wave.

FIG. 10 shows the relationship between the zenith angle θ and the azimuth angle φ. The arrival directions of the desired wave (S) and the unwanted wave (U) are (θ, φ) coordinates where the zenith angle is θ (0 ≦ θ ≦ 90 °) and the azimuth angle is φ (−180 ≦ φ ≦ 180 °). And (30 °, 45 °) and (45 °, −60 °), and simply adding the incident waves as a function of time t on each element antenna, and σ = Gaussian noise having a standard deviation of 10% was superimposed on each element independently. The relative intensity Ps: Pu of the desired wave and the unnecessary wave was set to 1: 1 as a dB value. The carrier frequency is f = 1.5 GHz, the sampling frequency is fs = 15 GHz, the total number of data points is N = 200, the element spacing is d = c / 2f in both vertical and horizontal directions (c is the speed of light), the number of element antennas is 5 × 5, and the number of subarrays The size was 3 × 3. FIG. 11A shows the result of applying the MUSIC method to the matrix R _XX (m) after the spatial averaging process. S is a desired wave, U is an unnecessary wave, and the direction is correctly estimated. On the other hand, (b) in the figure shows the result of calculating the antenna directivity by applying the DCMP method with the constraint values of 1 and 0 imposed on the desired wave and unnecessary wave directions, respectively. . A deep null is correctly formed in the unnecessary wave direction. (C) of the figure shows a result of calculating the data matrix by the above-described method using the complex weight obtained here and applying the MUSIC method again. It can be seen that the unnecessary wave U is removed and only the desired wave S is correctly extracted.

FIG. 12 shows the time series data Y ₁₁ , Y ₁₂ , Y ₁₃ ,... Y ₃₃ of each element of the 3 × 3 matrix in order. In the figure, the waveform of the desired incident wave with σ = 0% is also overwritten and displayed. From this result, it can be seen that the relative phase relationship between the element waveforms between the matrix elements is maintained. That is, it can be seen that the second MUSIC method is successfully applied.

  Next, a simulation was performed under the same conditions with the setting of one desired wave and two unnecessary waves. The arrival direction of newly added unnecessary waves was set to (60 °, 120 °), and the relative intensity was set to the same level as other unnecessary waves. The result is shown in FIG. FIG. 13A shows the direction estimation result by the MUSIC method, and FIG. 13B shows the direction of the desired wave arrival by the MUSIC method again by applying the DCMP method with the constraint of removing two unnecessary waves. It is an estimated result. Also in this case, it can be seen that the two unnecessary waves U1 and U2 are removed and only the desired wave S is correctly extracted.

  The estimation accuracy of the radio wave arrival direction depends mainly on the spectrum estimation accuracy of the MUSIC method, although it depends on the directivity null shape of the DCMP method. That is, the estimation accuracy improves as the number of time series data points increases. Regarding the power of the desired wave and the unnecessary wave, the null is formed better as the relative power level of the unnecessary wave with respect to the desired wave is larger from the principle of the DCMP method. Therefore, normally, it is sufficient to set a constraint condition that only the desired wave is set to 1. However, if the power level of the desired wave is slightly higher than that of the unnecessary wave, 1 is set to 1 for the desired wave and the unnecessary wave is set. On the other hand, if a constraint condition is imposed on both, such as 0, good results are often obtained.

  Next, the simulation was performed by changing the array antenna shape from a 5 × 5 25-element square shape to a 9-element cross shape. Other conditions were the same as in FIG. FIG. 14A shows the first estimation result by the MUSIC method, and FIG. 14B shows the final result from which unnecessary waves are removed. Although the spectrum is slightly broadened, the peak position is estimated almost correctly as in the case of 25 elements of 5 × 5.

Next, an example in which the integrated delta range (ADR) obtained inside the GPS receiver is used as a reception signal at the reception unit of the reception array will be described with reference to FIGS.
FIG. 6 is a block diagram showing the configuration of the positioning device. 6, 11a, 11b,... 11i are a plurality of element antennas as shown in FIG. A portion indicated by 2 is a complex weight giving unit that gives a complex weight to the received signals of the element antennas 11a to 11i. 21i is an amplitude adjuster for adjusting the amplitude of the received signal of each element antenna, and 22a, 22b... 22i are phase adjusters for adjusting the phase of the signal of each element antenna. The adder 3 adds the reception signals of the respective element antennas that have been subjected to the amplitude adjustment and the phase adjustment, and gives them to the GPS receiver 7 as one reception signal.

  FIG. 7 is a diagram showing the configuration of the signal arrival direction estimation apparatus. (B) is a block diagram showing the overall configuration, and (A) is a block diagram showing the configuration of a GPS receiver. As shown in (A), the GPS receiver sequentially converts a down converter 12 that converts a signal from the GPS antenna 11 into a predetermined intermediate frequency signal, an A / D converter 13 that converts the signal into a digital data string, and the digital data. The signal processing unit 14 inputs and performs signal processing and outputs data including ADR, which will be described later, and a positioning calculation unit 15 that controls the signal processing unit 14 and performs positioning calculation. The signal processing unit 14 includes a plurality of channels in order to simultaneously receive signals from a plurality of GPS satellites. The down converter 12 uses the reference frequency signal as a local signal, and converts the input signal from the GPS antenna 11 into an intermediate frequency signal. The A / D converter 13 converts the signal into digital data having a predetermined number of bits.

  As shown in FIG. 7B, the signal arrival direction estimation device includes a plurality of GPS antennas 11a, 11b... 11i, a plurality of GPS receivers 5a, 5b... 5i, and a DSP (digital signal processor). A data processing unit 6 is provided. The data processing unit 6 reads the ADR output from each of the GPS receivers 5a, 5b,... 5i, and estimates the arrival direction of the signal by data processing.

  FIG. 8 is a block diagram showing a configuration of the signal processing unit 14 shown in FIG. FIG. 9 is a block diagram showing the relationship between the carrier NCO 71 and ADR provided in the signal processing unit 14.

  In the signal processing unit 14, the carrier NCO 71 receives control data from the positioning calculation unit 15 and generates carrier signals (I signal, Q signal) having a phase of 0 ° and 90 ° at a predetermined frequency. The code generator 73 generates three C / A codes (E, P, L) having a predetermined code phase shift, and the code NCO 72 numerically controls the code phase. Multipliers 74, 75, and 76 multiply the IF signal with three C / A codes (E, P, and L) having a predetermined code phase shift. Multipliers 77 and 78 multiply the multiplication result of multiplier 74 by an I signal and a Q signal, respectively. Multipliers 79 and 80 multiply the multiplication results of multipliers 75 and 76 by the I signal, respectively.

  The PI integrator 81 and the PQ integrator 82 obtain the correlation value between the carrier signal generated by the carrier NCO 71 and the carrier component of the IF signal by accumulating the output values of the multipliers 77 and 78, and the result is stored in the register 86. , 87. Further, the EI integrator 83 and the LI integrator 84 integrate the output values of the multipliers 79 and 80 to obtain a correlation value between the two codes generated by the code generator 73 and the codes of the IF signal. The adder 85 obtains the difference between the integrated value of the EI integrator 83 and the integrated value of the LI integrator 84 and inputs the value to the register 88.

  The positioning calculation unit 15 obtains the C / A code phase and the carrier phase from the correlation results obtained in the registers 86, 87, 88 and performs tracking.

  As shown in FIG. 9, the carrier NCO 71 is a combination of a D-type flip-flop 711 for a predetermined bit and an adder 712 having a predetermined bit width. The output of the D-type flip-flop 711 is the output of the carrier NCO 71 and corresponds to the decimal value of ADR. The most significant bit of the flip-flop 711 has a weight of π radians, the next bit is π / 2 radians, the next is π / 4 radians, and the weight is decreased by ½. Therefore, the phase changes by 2π radians when the output value of the flip-flop 711 changes once. The ADR counter 89 is a counter by software processing, detects a carry from the most significant bit of the flip-flop 711, and counts the number thereof to obtain an integer value of ADR.

  With such a configuration, the number of clocks required for one cycle of the flip-flop 711 varies depending on the value of the set frequency for the adder 712. Therefore, the frequency of the output signal of the carrier NCO 71 can be set based on the set frequency value. Further, since the cycle in which the change of the decimal value of the ADR makes a round corresponds to one cycle of the carrier signal that is phase-locked, 1 of the carrier signal in which the output value of the D-type flip-flop 711 is synchronized with the reference clock signal. The value read at the timing corresponding to the period is equal to the carrier phase information of the received signal.

  Each GPS receiver 5a-5i shown in FIG. 7 outputs the integer value and decimal value of this ADR, for example, every second as carrier phase information of the received signal.

  (A) of FIG. 15 arranges nine GPS receivers 5a to 5i shown in FIG. 6 in the shape of a cross-shaped array antenna, and 20 points (1 point / second, 20 seconds) obtained discretely therefrom. This is a result of applying the MUSIC method to the ADR data of the minute data. The measurement date and time, the measurement location, and the satellite number are respectively GPS time, February 24, 2001, 135 degrees east longitude, 34 degrees north latitude, satellite numbers # SV7, 666181928 sec. It is. Here, the MUSIC method is applied on the assumption that the number of incident waves is two, that is, the desired wave S and the unnecessary wave U1, but small peaks U2 and U3 appear in addition to them. (B) of the figure is a result of applying the DCMP method and estimating the direction of arrival again using the MUSIC method for the purpose of suppressing such unnecessary peaks. Thus, it was confirmed that the present invention can also be applied to the ADR obtained by each receiving unit of the receiving array.

  In the example shown above, a receiving array in which the receiving units are arranged so as to cross in a cross shape on a plane or a receiving array arranged in a two-dimensional manner is used, but a plurality of receiving units are arranged in a straight line. The same applies to the receiving array.

  In the example described above, the MUSIC method is used to estimate the direction of arrival of the signal, but other methods such as a beamformer method, a Capon method, a linear prediction method, a minimum norm method, an ESPRIT method, or a MODE method are used. You can also This is because all of them can be similarly optimized by the arrival direction estimation algorithm.

The block diagram which shows the structure of the adaptive array which concerns on each embodiment, and a receiver provided with the same The figure which shows the example of arrangement | positioning of each element antenna of a two-dimensional array antenna Block diagram showing processing procedure of adaptive array characteristics optimization method Diagram showing data at each stage of arithmetic processing Flow chart showing processing procedure of adaptive array characteristic optimization method Block diagram showing the configuration of the positioning device Block diagram showing the configuration of the GPS receiver Block diagram showing the configuration of the signal processing unit of the same device The figure which shows the relationship between a carrier NCO and an ADR counter The figure which shows the coordinate system of the two-dimensional array antenna and the radio wave arrival direction The figure which shows the result of applying the MUSIC method to the matrix after the spatial averaging process The figure which shows the result which displayed the time series data of each element of the matrix after the complex weight multiplication in order The figure which shows the result of having performed the simulation with the setting of 1 desired wave and 2 unnecessary waves The figure which shows the result of having made the array antenna shape a cross shape and performing the simulation The figure which shows the result of having applied and simulated to ADR

Explanation of symbols

1-array antenna 1a to 1n-element antenna 2-complex weight assigning section 3-adder section 6-data processing section 11-element antenna 21-amplitude adjuster 22-phase adjuster

Claims (4)

A data matrix of time series data is obtained from a received signal at each receiving unit of a receiving array in which a plurality of receiving units that receive an incoming wave signal are arranged, and a correlation matrix of a subarray matrix of the data matrix is calculated, and the correlation Estimating the direction of arrival of a signal based on a spatial average correlation matrix obtained by weighted spatial averaging of the matrix;
Based on the arrival direction of the desired wave and / or unnecessary wave of the signal, a constraint condition is determined to determine a complex weight vector, each of the sub-array matrices is a column vector, and each column vector and the transposed conjugate of the complex weight vector are A rearrangement matrix obtained by rearranging the inner product of the subarrays in the order of the array arrangement is obtained, a correlation matrix for direction of arrival re-estimation is obtained from the rearrangement matrix, and the spatial average correlation matrix is replaced with the correlation matrix for direction of arrival re-estimation. A method of optimizing the characteristics of an adaptive array, comprising re-applying the step of estimating the direction of arrival of the signal.   The adaptive array characteristic optimization method according to claim 1, wherein the reception array has the reception units arranged one-dimensionally or two-dimensionally.   The adaptive array characteristic optimization method according to claim 1, wherein the reception array is arranged such that the reception units are orthogonally crossed on a plane.   Each of the plurality of receiving units receives an incoming signal from a transmission source, obtains an integrated delta range (ADR) obtained by tracking the carrier phase of the received signal by the reception, and receives the value of the integrated delta range. The adaptive array characteristic optimization method according to claim 1, wherein the received signal is a reception signal in a unit.

Images