JP6632466B2 - Receiving apparatus and receiving method, and program and recording medium - Google Patents

Description

  The present invention relates to a receiving method and a receiving apparatus, and more particularly to a technique for receiving a radio wave emitted from a transmitter and specifying a direction of arrival of a direct wave from the transmitter based on a received signal. The present invention also relates to a program for causing a computer to execute the processing in the above-described receiving device or receiving method, and a computer-readable recording medium on which the program is recorded.

  When receiving radio waves such as mobile phones, wireless LANs, terrestrial digital broadcasts, etc., in addition to the arriving waves directly arriving from the transmitter (hereinafter referred to as direct waves), arriving waves arriving at buildings and vehicles by being reflected / scattered ( The reception performance is reduced due to the influence of a delay wave). An environment where a plurality of incoming waves exist is called a multipath environment.

  As one technique for suppressing a decrease in reception performance due to multipath, directivity control using an array antenna is known. The array antenna has a plurality of antenna elements, and the directivity can be given to the array antenna by controlling a weight coefficient used when combining signals received by the antenna elements. In a multipath environment, by controlling the directivity so that the main lobe of the array antenna is directed in the direction in which a direct wave arrives, it is possible to suppress performance degradation due to the influence of a delayed wave. In order to improve the reception performance by controlling the directivity of the array antenna, it is necessary to accurately estimate the direction in which the direct wave arrives.

  If the receiver is fixed and the direction of the transmitter is known in advance, the directivity may be manually adjusted so that the main lobe of the array antenna is oriented in the direction of the transmitter. In the case of receiving radio waves, for example, when receiving by inter-vehicle communication, manual adjustment is not practical because the relative position of the transmitter with respect to the receiving device changes as the vehicle travels. Therefore, it is necessary to automatically estimate the arrival direction of the direct wave from the received signal in which the direct wave and the delayed wave are multiplexed.

  In addition, the radio wave environment of wireless communication is classified into LOS (Line Of Light) where the transmitter is in line of sight and NLOS (None Line Of Light) where the transmitter is out of line of sight from the viewpoint of the receiver. However, in the present invention, an LOS environment is assumed.

  Patent Literature 1 describes an apparatus that measures a multipath arrival direction based on signals received by two antennas. In this apparatus, a frequency characteristic (transfer function in the frequency domain) of a transmission path is estimated from a signal received by each antenna element, and the estimated frequency characteristic of the transmission path is subjected to inverse Fourier transform to calculate a complex delay profile. An incoming wave having a different delay time is separated from the delay profile, and an arrival angle of the separated direct wave is estimated based on a phase difference between antenna elements.

  Patent Document 2 discloses that the delay time is estimated by super-resolution processing such as MUSIC (Multiple Signal Classification) processing and ESPRIT (Estimation of Signal Parameters via Rotational Invariance Technologies) processing. The technique disclosed in Patent Document 2 converts a signal received by a plurality of antennas into a frequency spectrum, estimates the delay time of each arriving wave by super-resolution processing using the frequency spectrum, and determines from the estimation result that each arriving wave The coefficient matrix to be mixed is estimated, and the frequency spectrum is multiplied by the pseudo inverse matrix of the coefficient matrix to separate the components of the direct wave, and the angle of arrival is estimated from the phase difference of the separated direct wave.

Patent No. 4833144 (Fig. 1) JP 2010-286403 A (FIG. 1)

Nobuyoshi Kikuma, "Adaptive Signal Processing by Array Antenna", Science and Technology Publisher, November 1998

  Non-patent document 1 will be mentioned later.

In the technique of Patent Literature 1, there is a problem that a direct wave and a delayed wave cannot be separated when a delayed wave having a short delay time exists. For example, when a building, a vehicle, or the like exists near the receiver, the delay time is very short because there is no large difference in the radio wave propagation path length from the transmitter to the receiver between the direct wave and the delayed wave. . If this delay time is shorter than the delay time resolution of the complex delay profile, the direct wave and the delayed wave overlap on the estimated complex delay profile and cannot be separated. As a result, there is a problem that the angle of arrival estimation accuracy is significantly reduced. For example, the delay time resolution of a complex delay profile that receives and estimates a signal having a bandwidth of 10 MHz is about 100 ns, which is the reciprocal of the bandwidth. On the other hand, when a vehicle or the like exists at a position 3 m away from the transmitter when viewed from the receiver, the delay time for a direct wave is τ = 3 × 2 / c = 20 ns. Here, c is the speed of light (about 3 × 10 ⁸ m / s). The delay time 20 ns is shorter than the delay time resolution 100 ns, and the direct wave and the delayed wave overlap on the estimated delay profile, and as a result, the estimation accuracy of the angle of arrival decreases.

  On the other hand, the method of Patent Literature 2 has a problem that the amount of processing required for separating incoming waves, particularly the amount of calculation for calculating a pseudo inverse matrix, is large.

  The present invention provides a receiving apparatus and a method capable of accurately estimating the angle of arrival of a direct wave in an environment where a delayed wave having a short delay time exists and suppressing the amount of necessary calculations. The purpose is to:

The receiving device of the present invention includes:
A receiving device for receiving a radio wave emitted by a transmitter and estimating an angle of arrival of an incoming wave from the transmitter,
The first to N-th antenna elements are provided corresponding to the first to N-th (N is an integer of 2 or more) antenna elements constituting the array antenna, and the first to N-th antenna elements are obtained by receiving the radio waves with the corresponding antenna elements. A first to an N-th wireless receiving unit that performs frequency conversion and AD conversion on the analog signal to generate first to N-th digital signals;
It is provided corresponding to each of the first to N-th radio reception units, estimates the frequency characteristics of the transmission path from the first to N-th digital signals, and outputs the first to N-th transmission path estimation results. First to N-th transmission path estimating units;
A delay time estimating unit for estimating a delay time of one or more arriving waves included in the radio wave by super-resolution processing from one of the first to Nth transmission path estimation results;
Assuming that the arrival angle is one of M (M is an integer of 2 or more) different arrival angle candidates, for each of the arrival angle candidates, the first through N-th corresponding to the arrival angle candidate are considered. By adding a correction amount for canceling the phase difference between the frequency characteristics of the transmission lines and adding the frequency characteristics of the first to Nth transmission lines, the composite frequency characteristics of the M arrival angle candidates are calculated. A synthesizing part,
A reception power estimator configured to estimate reception power of a direct wave for the M arrival angle candidates from the synthesized frequency characteristics of the M arrival angle candidates and the delay time;
Determining which of the reception powers of the direct waves estimated for the M arrival angle candidates is the largest, and estimating the arrival angle candidate having the largest reception power as the arrival angle of the direct wave And a part.

  According to the present invention, the direct wave and the delayed wave are separated after estimating the delay time of the incoming wave by the super-resolution processing. Therefore, even when there is a delayed wave with a short delay time, the direct wave can be separated from the delayed wave, and the arrival angle of the direct wave can be estimated with high accuracy.

FIG. 2 is a block diagram illustrating a receiving device according to the first embodiment of the present invention. FIG. 2 is a block diagram illustrating a configuration example of a transmission path estimating unit in FIG. 1. FIG. 4 is a block diagram illustrating another configuration example of the transmission path estimator in FIG. 1. FIG. 5 is a conceptual diagram illustrating a relationship between an arrival angle of an incoming wave and a phase difference of a received radio wave with respect to an antenna array having two antenna elements. FIG. 3 is a conceptual diagram illustrating a relationship between an arrival angle of an incoming wave and a phase difference of a received radio wave with respect to an antenna array having N antenna elements. FIG. 2 is a block diagram illustrating a configuration example of a reception power estimating unit in FIG. 1. FIG. 9 is a block diagram illustrating a receiving device according to a second embodiment of the present invention. FIG. 8 is a block diagram illustrating a configuration example of a reception power estimation unit in FIG. 7. FIG. 13 is a block diagram illustrating a receiving device according to a third embodiment of the present invention. FIG. 10 is a block diagram illustrating a configuration example of a reception power estimation unit in FIG. 9. FIG. 14 is a block diagram illustrating a receiving device according to a fourth embodiment of the present invention. FIG. 12 is a block diagram illustrating a configuration example of a reception power estimation unit in FIG. 11. 26 is a flowchart illustrating a procedure of a process in a receiving method according to the fifth embodiment of the present invention. 14 is a flowchart illustrating a procedure of a process in an example of a transmission path estimation step in FIG. 13. 14 is a flowchart illustrating a procedure of a process in another example of the transmission path estimation step in FIG. 13. 14 is a flowchart illustrating a procedure of a process in an example of a received power estimation step in FIG. 13. 26 is a flowchart illustrating a procedure of a process in a receiving method according to the sixth embodiment of the present invention. 18 is a flowchart illustrating a procedure of a process in an example of a received power estimation step in FIG. 17. 28 is a flowchart illustrating a procedure of a process in a receiving method according to the seventh embodiment of the present invention. 20 is a flowchart illustrating a procedure of a process in an example of a received power estimation step in FIG. 19. 28 is a flowchart illustrating a procedure of a process in the receiving method according to the eighth embodiment of the present invention. 22 is a flowchart illustrating a procedure of a process in an example of a received power estimation step in FIG. 21. FIG. 21 is a block diagram illustrating a computer that executes processing according to the first to eighth embodiments.

Embodiment 1 FIG.
FIG. 1 shows a receiving apparatus according to the present embodiment.
The illustrated receiving apparatus receives a radio wave emitted from a transmitter and estimates the direction of the transmitter, that is, the arrival direction of a direct wave.
The illustrated receiving apparatus includes radio receiving sections 11-1 and 11-2, transmission path estimating sections 12-1 and 12-2, delay time estimating section 13, combining section 14, received power estimating section 15, and arrival angle estimating section. The wireless receivers 11-1 and 11-2 are connected to the antenna elements 10-1 and 10-2, respectively.

The wireless receiving unit 11-1 and the transmission path estimating unit 12-1 constitute a first system, which correspond to each other and are provided corresponding to the first antenna element 10-1. .
The wireless receiving unit 11-2 and the transmission path estimating unit 12-2 constitute a second system, which correspond to each other and are provided corresponding to the second antenna element 10-2. .
The processing in the first system and the processing in the second system are different from each other in the signals input to the respective systems (that is, they are obtained by reception at the antenna elements 10-1 and 10-2, respectively). Are similar to each other, except for
The delay time estimating unit 13, the combining unit 14, the received power estimating unit 15, and the arrival angle estimating unit 16 are provided in common for the two systems described above.

  Although the receiving apparatus shown in FIG. 1 has a configuration in which the number of antenna elements is two, the present invention is applicable even when the number of antenna elements is three or more. May be described as N.

The wireless receiving units 11-1 and 11-2 in FIG. 1 are provided corresponding to the antenna elements 10-1 and 10-2, respectively, and are obtained by receiving radio waves with the corresponding antenna elements. It converted into a baseband signal obtained signal by frequency conversion, _{(the n} 1 or 2) digital signal Sr n by AD conversion, and outputs the digital signal Sr _n that generated.

The transmission path estimating units 12-1 and 12-2 in FIG. 1 are provided corresponding to the wireless receiving units 11-1 and 11-2, respectively, and the digital signal Sr output from the corresponding wireless receiving unit is provided. _{From n} , the frequency characteristic of the transmission path (transfer function in the frequency domain) X _n (t) is estimated.
The method of estimating the frequency characteristics of the transmission path depends on the transmission method used in the communication system. The present invention can be applied to any transmission scheme. Here, as an example, an OFDM (Orthogonal Frequency Division Multiplex) transmission scheme adopted in many communication systems is adopted, and a DSSS (Direct Sequence) is adopted. Estimation of the frequency characteristic of the transmission path when the (Spectrum Spread) transmission method is adopted will be described.

  First, a description will be given of the estimation of the frequency characteristics of the transmission path, which is performed when the OFDM transmission method is adopted. The OFDM transmission method is a transmission method in which a symbol obtained by multiplexing a plurality of orthogonal subcarriers is used as a transmission unit. In many communication systems employing the OFDM transmission scheme, some subcarriers are used as known pilot subcarriers on the transmission side and the reception side in order to compensate for transmission line distortion on the reception side. In the present embodiment, the frequency characteristics of the transmission path are estimated using the pilot subcarriers.

FIG. 2 shows an example of the transmission channel estimator 12-n (n is 1 or 2) used when the OFDM transmission method is adopted.
The transmission channel estimation unit 12-n illustrated in FIG. 2 includes an FFT unit 20-n, a pilot extraction unit 21-n, a pilot generation unit 22-n, a division unit 23-n, and an interpolation unit 24-n.

FFT unit 20-n is, the FFT digital signal Sr _n output from the radio reception unit 11-n in FIG. 1 (Fast Fourier Transform), to convert the frequency axis from the time axis for each symbol, each subcarrier Is output.
Pilot extraction section 21-n extracts pilot carriers from subcarriers output from FFT section 20-n.
The pilot generation unit 22-n generates a known pilot carrier in the receiving device.
The division unit 23-n divides the pilot carrier extracted by the pilot extraction unit 21-n by the pilot carrier generated by the pilot generation unit 22-n to output the frequency characteristic of the transmission path acting on the pilot carrier. I do.
The interpolation unit 24-n interpolates the frequency characteristics of the transmission line acting on the pilot carrier in the symbol direction and the subcarrier direction, thereby obtaining the frequency characteristics of the transmission lines of all the subcarriers (transmission channel estimation result) _Xn (t). Get.

  Next, estimation of the frequency characteristics of the transmission path, which is performed when the DSSS transmission method is adopted, will be described. The DSSS transmission method is a method of transmitting a signal spread using a pseudo noise sequence for each symbol and despreading the signal on the receiving side.

FIG. 3 shows an example of the transmission path estimator 12-n (n is 1 or 2) used when the DSSS transmission scheme is adopted.
The transmission channel estimation unit 12-n illustrated in FIG. 3 includes a pseudo noise sequence generation unit 25-n, a despreading unit 26-n, and an FFT unit 27-n.

The pseudo-noise sequence generation unit 25-n generates the same pseudo-noise sequence Ns as used at the time of spreading on the transmission side.
Despreader 26-n, and outputs the calculated a sliding correlation between the digital signal Sr _n and the pseudo-noise sequence Ns output from the radio reception unit 11-n in FIG. 1 in symbol units.
The FFT unit 27-n obtains the frequency characteristic of the transmission path (transmission path estimation result) _Xn (t) by converting the output of the despreading unit 26-n into the frequency domain by FFT.

The frequency characteristics (transmission path estimation results) of the transmission path estimated by the transmission path estimating units 12-1 and 12-2 in FIG. 1 are represented by the following equations (1A) and (1B).

In equations (1A) and (1B), the superscript "T" represents transposition.
Further, f _k (k = 1, 2,..., K) is a frequency obtained by dividing the range from the lowest frequency f ₁ to the highest frequency f _K at equal intervals. Equal to the carrier frequency.
Further, t indicates the time at which the transmission channel estimation was performed.

Returning to FIG. 1, the delay time estimating unit 13 outputs the output X ₁ (t) of one of the transmission channel estimating units 12-1 and 12-2, for example, the first transmission channel estimating unit 12-1. ), The delay time of the arriving wave included in the radio wave received by the corresponding antenna is estimated.
Estimation of the delay time is performed by super-resolution processing (processing according to a super-resolution algorithm) such as MUSIC (Multiple Signal Classification) processing and ESPRIT (Estimation of Signal Parameters via Rotational Invariance Technologies) processing. Here, the number of arriving waves is L, and the delay time of each arriving wave is τ ₁ , τ ₂ ,..., Τ _L (τ ₁ <τ ₂ <... <τ _L ).

  A method of estimating the delay time by the super-resolution processing is described in, for example, Chapter 13 of Non-Patent Document 1 described above.

Synthesizer 14 synthesizes the frequency characteristics of the transmission path estimated by the channel estimation unit 12-1,12-2 _X 1 (t) and _X 2 (t), to produce a combined frequency characteristic X (t) .
At the time of synthesis, correction is performed to make the phases of the frequency characteristics X ₁ (t) and X ₂ (t) uniform.

When radio waves are received by the antenna elements 10-1 and 10-2, the received radio waves have a phase difference corresponding to the angle of arrival.
For example, as shown in FIG. 4, the antenna element spacing d _{2 [m],} the wavelength of the carrier wave lambda [m], when the angle of arrival of the received radio wave and theta _a, the path difference d _p between the antenna elements, between the arrival angle θ _a,
d _p = d ₂ × sin θ _a
There is a relationship. In addition, between the propagation path difference d _p and the carrier wavelength λ and the phase difference φ _a of the received radio wave,
d _p = λ × φ _a / 2π
There is a relationship.

Thus, between the arrival angle theta _a phase difference phi _a, a relationship of the following equation (2).

That is, if the arrival angle theta _a is as shown in FIG. 4, the received radio waves of the antenna elements 10-2, with respect to the received radio wave in the antenna elements 10-1, phi _a of the formula (2) (= The phase is delayed by 2π (d ₂ / λ) θ _a ).
In this case, for example, if the phase of X ₂ (t) is advanced by φ _a (= 2π (d ₂ / λ) θ _a ), the phase difference between the arriving waves can be canceled and the phases can be aligned.

In this embodiment, assuming that the arrival angle theta _a is any angle theta ₁ through? _M of M (M is an integer of 2 or more), so as to cancel the phase difference corresponding to angle of arrival assumed Phase correction is performed to calculate a composite frequency characteristic, and using the calculated composite frequency characteristic, the incoming wave having the shortest delay time, that is, the received power of the direct wave is calculated, and the calculated received power of the direct wave is maximized. angle (represented by "theta _x") is estimated to be the arrival angle theta _a is. Hereinafter, these assumed angles θ _{1 to} θ _M are referred to as “estimated arrival angle candidates” or simply “candidates”.
Such a process can be said to be a process of changing the estimated value of the angle of arrival and searching for an estimated value at which the received power of the direct wave becomes maximum (peak value).

The composition when the arrival angle is a candidate θ _m (m is any one of 1 to M) is represented by the following equation.

In Equation (3), j2π (d ₂ / λ) sin θ _m is a correction amount for canceling the phase difference of the received radio wave.
の is for canceling a phase difference due to a difference in delay time between two systems generated in an analog circuit in the antenna elements 10-1 and 10-2 and analog circuits in the radio receiving units 11-1 and 11-2 and a phase difference. This is the correction amount.
As described above, synthesis by the formula (3) above, if the arrival angle theta _a match to the candidate theta _m, the result of the synthesis is the maximum value (peak value).
The combining unit 14 performs combining for each of the _M different arrival angle candidates θ _{1 to} θ _M to generate _M combined frequency characteristics X (t, θ ₁ ) to X (t, θ _M ).

Although the number of antenna elements is two in the present embodiment, the present invention is applicable even when the number of antenna elements is three or more.
When the number of antenna elements is three or more, similarly to the above, the correction amount for canceling the phase difference of the arriving wave depending on the arrangement of the antenna elements and the phase difference generated in the antenna element and the analog circuit in the radio receiving unit are canceled. May be added and combined.

The generalization assuming that N is 2 or more is as follows.
As shown in FIG. 5, the antenna element 10-1 and the antenna element 10-n (n is any one of 2 to N) is the distance between the _d n [m], the antenna element 10-1 and the radio receiver unit 11 an analog circuit of -1, the difference in delay time between the systems produced by an analog circuit of the antenna element 10-n and the wireless reception section 11-n, and the correction amount for canceling a phase difference due to the difference of the phase characteristic [psi _n In this case, the synthesizing unit 14 may perform synthesis according to the following equation (4) instead of the above equation (3). Equation (3) above corresponds to the case where N = 2 in equation (4) below.

The received power estimating unit 15 calculates the synthesized frequency characteristics X (t, θ ₁ ) to X (t, θ _M ) calculated by the synthesizing unit 14 and the delay times τ _{1 to} τ _L estimated by the delay time estimating unit 13. Thus, the received power of the incoming wave having the shortest delay time, that is, the direct wave, is estimated for each of the arrival angle candidates θ _{1 to} θ _M. That is, to estimate the reception power of the direct wave for each angle of arrival candidates theta _m. Since the received power is estimated for each of the _M candidates θ _{1 to} θ _M , an estimated value of the M received powers is obtained.

Each of the synthetic frequency characteristics X (t, θ ₁ ) to X (t, θ _M ) supplied from the synthesizing unit 14 to the reception power estimating unit 15, that is, the synthetic frequency characteristics X (t, θ _m ) for the candidate θ _m Is represented by the following equation (5).

In Expression (5), F (t, θ _m ) is a column vector composed of complex amplitudes, and is represented by Expression (6) below.

In the equation (6), F ₁ (t, θ _m ) (1 is any one of 1 to L) is a complex amplitude (wave source) of the l-th arriving wave.

In the equation (5), A is called a mode matrix, and is represented by the following equation (7).

In the equation (7), a (τ ₁ ) to a (τ _L ) are called mode vectors, and each a (τ ₁ ) (1 is any one of 1 to L) is represented by the following equation (8). ).

In equation (8), τ ₁ is the arrival delay time of the l-th incoming wave.
_N a (t, _{θ m)} in equation (5) is the noise vector, represented by the following formula (9).

The correlation matrix _{Rxx at} this time is represented by the following equation (10).

上記の式（１１）で、σ^２は雑音電力、Ｉは単位行列を表す。 In equation (10A), E [•] is an ensemble average.
In Equation (10B), N (t, θ _m ) is a noise power matrix.
If the noise is incandescent noise having the same intensity at all frequencies, the noise power matrix is represented by the following equation.

In the above equation (11), σ ² represents noise power, and I represents a unit matrix.

In the above equation (10B), S (θ _m ) is called a signal (wave source) correlation matrix, and its component (element) is represented by the following equation (12).

From the equation (10B), the signal (wave source) correlation matrix S (θ _m ) is obtained by the following equation (13).

  In Equation (13), the upper suffix “H” represents a complex conjugate transpose, and the upper suffix “−1” represents an inverse matrix.

If the incoming waves are uncorrelated, the signal correlation matrix S (θ _m ) is a diagonal matrix.
When there is a correlation between the incoming waves, a diagonal matrix may be generated by using a method called a spatial averaging method, a moving average method (Smoothing Preprocessing), or the like. Such a method is described in Non-Patent Document 1, for example.

When the signal correlation matrix S (θ _m ) is a diagonal matrix, its components are represented by the following equation (14).

In the formula (14), the l th row and l columns of component _{s ll} (theta _m) is representative of the power of the l-th arrival wave. The component of l = 1, that is, the component s ₁₁ (θ _m ) in the first row and first column represents the power of the direct wave.

  FIG. 6 shows a configuration example of the reception power estimation unit 15 for calculating the power of the direct wave according to the above idea. The illustrated received power estimation unit 15 includes a mode matrix generation unit 51, a correlation matrix calculation unit 52, a noise power calculation unit 53, and an incoming wave power calculation unit 54.

The mode matrix generation unit 51 receives the delay times τ _{1 to} τ _L estimated by the delay time estimation unit 13 as inputs, and each of the input delay times τ _{1 to} τ _L τ ₁ (1 is any one of 1 to L). from equation (generates mode vector a (tau _l) shown in 8), the mode vector a (tau ₁ for all of the delay time _{τ 1 ~τ L) ~a (τ} L) from), the formula (7) Is generated.

The correlation matrix calculation unit 52 receives the synthesized frequency characteristics X (t, θ ₁ ) to X (t, θ _M ) calculated by the synthesis unit 14 and receives M correlation matrices R _xx (θ ₁ ) to R _xx ( θ _M ) is calculated. That is, the calculation of Expression (10A) is performed for each arrival angle candidate θ _m to calculate the correlation matrix R _xx (θ _m ).

The noise power calculation unit 53 receives the synthesized frequency characteristics X (t, θ ₁ ) to X (t, θ _M ) calculated by the synthesis unit 14, and receives a noise power matrix N ( t, θ _m ) is calculated.
Various methods can be applied to the calculation of the noise power. For example, the variance of the synthetic frequency characteristics X (t, θ ₁ ) to X (t, θ _M ) may be calculated.
The noise power may be calculated from the transmission path estimation results X ₁ (t) and X ₂ (t) for each antenna element instead of the composite frequency characteristics.

The arriving wave power calculation unit 54 receives the mode matrix A (Equation (7)) from the mode matrix generation unit 51 and receives the correlation matrices R _xx (θ ₁ ) to R _xx (θ _M ) (Equation ( 10A)), the noise power matrix N (t, θ _m ) from the noise power calculation unit 53, and the above-described noise power matrix N (N) from the correlation matrices R _xx (θ ₁ ) to R _xx (θ _M ). t, to remove the effects of noise by subtracting the theta _m), with respect to the result of this subtraction, including inverse matrix calculation of a product of the mode matrix a (equation (7)) and its complex conjugate transpose matrix a ^H By performing the processing, signal correlation matrices S (θ ₁ ) to S (θ _M ) are generated. That is, a signal correlation matrix S (θ _m ) is generated for each arrival angle candidate θ _{m according} to the above equation (13).

From the signal correlation matrices S (θ ₁ ) to S (θ _M ), the incoming wave power calculation unit 54 further calculates the first row and first column components s ₁₁ (θ ₁ ) to s ₁₁ (θ _M ). It is extracted and output as the power of the direct wave for the arrival angle candidates θ _{1 to} θ _M. That is, the component s ₁₁ (θ _m ) of the first row and the first column is extracted from each signal correlation matrix S (θ _m ) shown in Expression (14), and the power of the direct wave for the arrival angle candidate θ _m is extracted. Output as
The direct wave powers s ₁₁ (θ ₁ ) to s ₁₁ (θ _M ) calculated by the arriving wave power calculation unit 54 are supplied to the arrival angle estimation unit 16 as outputs of the reception power estimation unit 15.

In the above equations (10B) and (13), it is assumed that the noise is white noise having the same intensity at all frequencies. However, when the noise power has a frequency characteristic (the noise power depends on the frequency, The present invention can be applied to other cases.
In that case, the noise power matrix N (t, θ _m ) is expressed by noise components n ₁ (f ₁ ), n ₂ (f ₂ ),. It becomes a diagonal matrix having _K (f _K ) as an element.

The arrival angle estimation unit 16 determines which of the direct wave powers s ₁₁ (θ ₁ ) to s ₁₁ (θ _M ) supplied from the reception power estimation unit 15 is the maximum (peak value), and There _(either theta ₁ through? _M) arrival angle candidate theta _x is maximum, and estimates the arrival angle theta _a direct wave.

  The device of Patent Document 1 has a problem that when a delayed wave with a short delay time exists, the direct wave and the delayed wave cannot be separated, and the accuracy of estimation of the angle of arrival of the direct wave decreases. On the other hand, according to the present embodiment, the direct wave and the delayed wave are separated after estimating the delay time of the incoming wave by the super-resolution processing. Therefore, even when there is a delayed wave with a short delay time, the direct wave can be separated from the delayed wave, and the arrival angle of the direct wave can be estimated with high accuracy.

  In addition, the influence of noise can be reduced by subtracting the noise power in the frequency domain before calculating the received power of the incoming wave separated in the delay time domain. That is, when noise power is removed after the incoming wave is separated in the delay time domain, it is necessary to find the noise power in the delay time domain, which complicates the processing. On the other hand, in the present embodiment, as described above, after the noise power is subtracted in the frequency domain, the received power of the arriving wave is separated in the delay time domain, so that the processing is simple. In particular, when the noise power has a frequency characteristic (when the noise power varies in frequency), the processing is simplified by subtracting the noise power in the frequency domain.

Embodiment 2 FIG.
In the first embodiment, the arrival angle of the direct wave is calculated. On the other hand, in the second embodiment, the arrival angle of an incoming wave other than the direct wave is estimated. For example, the arrival angle of at least one arriving wave other than the direct wave in addition to the direct wave is estimated. In the following, a case will be described in which the arrival angles of all incoming waves, including direct waves, are estimated.

FIG. 7 shows a receiving apparatus according to the second embodiment.
The receiving device in FIG. 7 is generally the same as the receiving device in FIG. 1, except that a receiving power estimating unit 15b and an angle of arrival estimating unit 16b are provided instead of the receiving power estimating unit 15 and the angle of arrival estimating unit 16. .

FIG. 8 shows a configuration example of the reception power estimation unit 15b.
The received power estimating unit 15b is generally the same as the received power estimating unit 15 of FIG. 6, but includes an incoming wave power calculating unit 54b instead of the incoming wave power calculating unit 54.

The arriving wave power calculation unit 54b performs the same processing as that of the arriving wave power calculation unit 54 of the first embodiment, and generates signal correlation matrices S (θ ₁ ) to S (θ _M ).
Arrival wave power calculation unit 54b further from the generated signal correlation matrix _{S (θ 1) ~S (θ} M), their first row and first column of the component _{s 11 (θ 1) ~s 11} (θ M) extracts, second L row and the L columns of the component _{s LL} (θ _{1) ~s LL} a (θ _M), for the arrival angle candidate theta ₁ through? _M, the first incoming wave ~L th arrival wave Output as electric power. That is, all the diagonal components s ₁₁ (θ _m ) to s _LL (θ _m ) are extracted from each signal correlation matrix S (θ _m ), and the first to L-th components of the arrival angle candidate θ _m are extracted. Output as power of incoming wave.

Calculated in the arrival wave power calculating portion 54b 1 th arrival wave power _{s 11 (θ 1) ~s 11} (θ M) ~L th arrival wave power _{s LL (θ 1) ~s LL} (θ M ) Is supplied to the arrival angle estimation unit 16b as an output of the reception power estimation unit 15b.

The arrival angle estimation unit 16b compares the power of the same arriving wave among the powers of the first to Lth arriving waves supplied from the reception power estimation unit 15b, and determines which is the largest, , The arrival angle candidate having the largest power is estimated as the arrival angle of the arrival wave. That is, which of the power _{s ll} of l-th arrival wave for the arrival angle candidate _{θ 1 ~θ M (θ 1)} ~s ll (θ M) of determining whether the maximum (peak value), the power There _(either theta ₁ through? _M) arrival angle candidate theta _x is maximum, and estimates the arrival angle theta _al in l th arrival wave. This process is performed for all possible values of l, that is, from l = 1 to l = L.
The processing when l is 1 is the same as the processing in the arrival angle estimation unit 16 of the first embodiment, and the arrival angle of the direct wave is estimated by this processing.
When 1 is any of 2 to L, the arrival method of each of the second and subsequent arriving waves is estimated.

  As described above, the case has been described in which the arrival angles of all the incoming waves are estimated. However, the arrival angles of only some of the incoming waves may be estimated. For example, the angle of arrival of the direct wave may be estimated, and the angle of arrival of one of the incoming waves other than the direct wave may be estimated.

  In the second embodiment, the same effect as in the first embodiment can be obtained. In addition, it is possible to highly accurately estimate the angle of arrival of an incoming wave other than the direct wave.

Note that, from the signal correlation matrices S (θ ₁ ) to S (θ _M ) calculated by the reception power estimating unit 15 or 15b according to the first or second embodiment, the delay for each of m = ₁ to _M is determined. A change in received power with respect to time, in other words, received power (received power profile) P (τ, θ ₁ ) to P (τ, θ _M ) corresponding to each of the delay times τ _{1 to} τ _L can also be calculated. . For that purpose, the reception power profile P (τ, θ _m ) may be calculated by performing the calculation of the following equation (16) for each arrival angle candidate θ _m .

In the formula _{(16), s ll} is a component of the diagonal of the signal correlation matrix S (θ _m), l is either 1 to L.
Δ (τ) is a Dirac delta function.

Embodiment 3 FIG.
In the first embodiment, the generation of the mode matrix (A) and the signal correlation matrix (S (θ) are performed using all the delay times estimated by the delay time estimation unit 13, that is, using the arriving waves of all the delay times. )). On the other hand, in the third embodiment, an incoming wave whose delay time is equal to or longer than the threshold is not used for generating a mode matrix, calculating a signal correlation matrix, and the like. This is done to reduce the amount of calculation.

FIG. 9 shows a receiving apparatus according to the third embodiment.
The receiving device in FIG. 9 is generally the same as the receiving device in FIG. 1, except that a receiving power estimating unit 15 c and an angle of arrival estimating unit 16 c are provided instead of the receiving power estimating unit 15 and the angle of arrival estimating unit 16. .
FIG. 10 shows a configuration example of the reception power estimation unit 15c.

  The received power estimating unit 15c in FIG. 10 includes a mode matrix generating unit 51c, a correlation matrix calculating unit 52c, a noise power calculating unit 53c, an incoming wave power calculating unit 54c, a threshold setting unit 56, and a delayed wave removing unit 57. .

The threshold setting unit 56 sets a delay time threshold τ _th . The delay time threshold τ _th is used to remove a relatively long delay time. The delay time threshold τ _th may be determined in advance, or may be determined based on the delay times τ _{1 to} τ _L estimated by the delay time estimating unit 13.
In the case where it is determined in advance, for example, it is desirable to determine the threshold τ _th in consideration of the radio wave environment so that the number of delay times after removal does not exceed the upper limit. When determined based on the delay times τ _{1 to} τ _L estimated by the delay time estimating unit 13, it is desirable that the threshold value τ _th be determined so that the number of delay times after removal does not exceed the upper limit value.

Mode matrix generating unit 51c, with an input of a delay time τ ₁ ~τ _L estimated by the delay time estimation section 13 receives the delay time threshold value tau _th set by the threshold setting unit 56, the delay time threshold value tau _th A mode matrix A ′ is generated from a shorter delay time.

Specifically, among the delay times τ _{1 to} τ _L , τ _{1 to} τ _q (q is any of 1 to L) is shorter than the threshold τ _th , and τ _{(q + 1) to} τ _L is greater than or equal to the threshold τ _th Then, a mode vector a (τ ₁ ) shown in the above equation (8) is generated from each of τ ₁ (1 is any of 1 to q) of the delay times τ _{1 to} τ _q , and the threshold τ _th from all of the delay time shorter than τ ₁ ~τ _q mode vector a for _{(τ 1) ~a (τ q} ), and generates a mode matrix a 'as shown in the following equation (17).

The delayed wave removing unit 57 removes a component of the delayed wave whose delay time is equal to or _larger than the threshold τ _th from the combined frequency characteristics X (t, θ ₁ ) to X (t, θ _M ) calculated by the combining unit 14. , And outputs synthesized frequency characteristics X ′ (t, θ ₁ ) to X ′ (t, θ _M ) for the arriving wave whose delay time is shorter than the threshold τ _th , that is, the first to qth arriving waves. That is, a component of a delay wave whose delay time is equal to or greater than the threshold τ _th is removed from each of the synthesized frequency characteristics X (t, θ _m ) (m is any one of 1 to M), and the delay time is reduced to the threshold τ _th The synthesized frequency characteristic X ′ (t, θ _m ) for the shorter incoming wave is output.

A digital filter such as an FIR (Finite Impulse Response) filter and an IIR (Infinite Impulse Response) filter is used for removing the delayed wave. Since each of the above-described synthesized frequency characteristics X (t, θ ₁ ) to X (t, θ _M ) is a frequency characteristic of a transmission path, a desired component in a delay time domain is obtained by performing a filtering process in a frequency domain. Can be extracted.

The composite frequency characteristic X ′ (t, θ _m ) for the arriving wave whose delay time is shorter than the threshold τ _th , that is, the 1st to qth arriving waves, is represented by the following equation (18). .

In Expression (18), F ′ (t, θ _m ) is a column vector composed of complex amplitudes, and is represented by Expression (19) below.

In the equation (19), F _l (t, θ _m ) (1 is any one of 1 to q) is a complex amplitude (wave source) of the l-th arriving wave.
In Expression (18), A ′ is called a mode matrix, and is represented by Expression (17).

In Expression (18), N ′ _a (t, θ _m ) is a noise vector, and is represented by Expression (9). The noise vector N _a (t, θ _m ) used in equation (5) is calculated from the synthetic frequency characteristic X (t, θ _m ) for all arriving waves, whereas the noise vector N _a (t, θ _m ) is used in equation (18). The noise vector N ′ _a (t, θ _m ) is different in that the noise vector N ′ _a (t, θ _m ) is calculated from the synthetic frequency characteristic X ′ (t, θ _m ) of the arriving wave whose delay time is shorter than the threshold τ _th .

The correlation matrix calculation unit 52c calculates M correlation matrices R ′ _xx (θ ₁ ) from the synthesized frequency characteristics X ′ (t, θ ₁ ) to X ′ (t, θ _M ) output from the delayed wave removal unit 57. ＲR ′ _xx (θ _M ) is calculated. That is, the following equation (20) is calculated for each arrival angle candidate θ _m to calculate a correlation matrix R ′ _xx (θ _m ).

The noise power calculation unit 53c calculates the noise power using the noise power in the frequency domain as an element based on the synthesized frequency characteristics X ′ (t, θ ₁ ) to X ′ (t, θ _M ) output from the delay wave removal unit 57. The matrix N ′ (t, θ _m ) is calculated.

The arriving wave power calculation unit 54c receives the mode matrix A ′ (Equation (17)) from the mode matrix generation unit 51c, and receives the correlation matrices R ′ _xx (θ ₁ ) to R ′ _xx (θ _M ) from the correlation matrix calculation unit 52c. (Equation (20)), the noise power matrix N ′ (t, θ _m ) from the noise power calculation unit 53c, and the above-mentioned correlation matrices R ′ _xx (θ ₁ ) to R ′ _xx (θ _M ) The noise power matrix N ′ (t, θ _m ) is subtracted to remove the influence of noise, and the mode matrix A ′ (Equation (17)) and its complex conjugate transpose A ′ ^H The signal correlation matrices S ′ (θ ₁ ) to S ′ (θ _M ) are generated by performing a process including an inverse matrix operation of the product of That is, a signal correlation matrix S ′ (θ _m ) is generated for each arrival angle candidate θ _m according to the following equation (21).

The components of the signal correlation matrix S ′ (θ _m ) of Expression (21) are represented by Expression (22) below.

In the formula (22), the l th row and l columns of component _{s ll} (theta _m) is representative of the power of the l-th arrival wave. The component of l = 1, that is, the component s ₁₁ (θ _m ) in the first row and first column represents the power of the direct wave.
Arrival wave power calculation unit 54c from the signal correlation matrix _{S '(θ 1) ~S'} (θ M), their first row and first column of the component _{s 11} (θ _{1) ~s 11} a (theta _M) It is extracted and output as the power of the direct wave for the arrival angle candidates θ _{1 to} θ _M. That is, the component s ₁₁ (θ _m ) of the first row and first column is extracted from each signal correlation matrix S ′ (θ _m ), and is output as the power of the direct wave for the arrival angle candidate θ _m .
The direct wave power s ₁₁ (θ ₁ ) to s ₁₁ (θ _M ) calculated by the arriving wave power calculation unit 54c is supplied to the arrival angle estimation unit 16c as an output of the reception power estimation unit 15c.

The arrival angle estimation unit 16c determines which of the direct wave powers s ₁₁ (θ ₁ ) to s ₁₁ (θ _M ) supplied from the reception power estimation unit 15c is the maximum (peak value), and There _(either theta ₁ through? _M) arrival angle candidate theta _x is maximum, and estimates the arrival angle theta _a direct wave.

In the third embodiment, the same effect as in the first embodiment can be obtained.
In addition, since the mode matrix is generated and the signal correlation matrix is calculated after removing the arriving wave having a relatively long delay time, the amount of calculation can be suppressed even in an environment where a lot of arriving waves are included. Moreover, the angle of arrival of the direct wave can be estimated with high accuracy.

  That is, equation (21) is similar to equation (13), but uses mode matrix A ′ instead of mode matrix A, and modal matrix A ′ becomes mode matrix A because the delayed wave has been removed. The number of columns is smaller than that. For this reason, the calculation amount is reduced.

Embodiment 4 FIG.
In the third embodiment, the arrival angle of a direct wave is calculated. On the other hand, in the fourth embodiment, the arrival angles of the incoming waves other than the direct wave among the incoming waves whose delay time is shorter than the threshold τ _th are estimated. For example, the arrival angle of at least one arriving wave other than the direct wave in addition to the direct wave is estimated. In the following, a case will be described in which the arrival angles of all arriving waves, including direct waves, whose delay time is shorter than the threshold τ _th are estimated.

FIG. 11 shows a receiving apparatus according to the fourth embodiment.
The receiving device in FIG. 11 is generally the same as the receiving device in FIG. 9, but includes a received power estimating unit 15 d and an angle of arrival estimating unit 16 d instead of the received power estimating unit 15 c and the angle of arrival estimating unit 16 c. .

FIG. 12 shows a configuration example of the reception power estimation unit 15d.
The received power estimating unit 15d is generally the same as the received power estimating unit 15c of FIG. 10, but includes an incoming wave power calculating unit 54d instead of the incoming wave power calculating unit 54c.

The arriving wave power calculation unit 54d performs the same processing as the arriving wave power calculation unit 54c of the third embodiment to generate signal correlation matrices S ′ (θ ₁ ) to S ′ (θ _M ).
From the generated signal correlation matrices S ′ (θ ₁ ) to S ′ (θ _M ), the arriving wave power calculation unit 54d further calculates the first row and first column components s ₁₁ (θ ₁ ) to s ₁₁ (θ _M) ~ q-th row and the q column component _{s qq} (θ _{1) ~s qq} a (θ _M), for the arrival angle candidate θ ₁ ~θ _{M, 1} th arrival wave ~q th arrival wave power Output as That is, from each signal correlation matrix S ′ (θ _m ), all the diagonal components s ₁₁ (θ _m ) to s _qq (θ _m ) are converted into the first to q-th arriving waves for the arrival angle candidate θ _m. Output as power.

Calculated in the arrival wave power calculation unit 54d 1 th arrival wave power _{s 11 (θ 1) ~s 11} (θ M) ~q th arrival wave power _{s qq (θ 1) ~s qq} (θ M ) Is supplied to the arrival angle estimation unit 16d as an output of the reception power estimation unit 15d.

The arrival angle estimation unit 16d compares the power of the same arriving wave among the powers of the first to qth arriving waves supplied from the reception power estimation unit 15d, and determines which is the largest, , The arrival angle candidate having the largest power is estimated as the arrival angle of the arrival wave. That is, which of the power _{s ll} of l-th arrival wave for the arrival angle candidate _{θ 1 ~θ M (θ 1)} ~s ll (θ M) of determining whether the maximum (peak value), the power There _(either theta ₁ through? _M) arrival angle candidate theta _x is maximum, and estimates the arrival angle theta _al in l th arrival wave. This process is performed for all possible values of l, that is, for all cases from l = 1 to l = q.
The processing when l is 1 is the same as the processing in the arrival angle estimation unit 16c of the third embodiment, and the arrival angle of the direct wave is estimated by this processing.
When 1 is any of 2 to q, the arrival method of each of the second and subsequent arriving waves is estimated.

  As described above, the case has been described in which the arrival angles of all the arriving waves whose delay time is shorter than the threshold value are estimated. However, the arrival angle may be estimated for only a part of the arriving waves whose delay time is shorter than the threshold value. . For example, the angle of arrival of the direct wave may be estimated, and the angle of arrival of one of the incoming waves other than the direct wave whose delay time is shorter than the threshold may be estimated.

  In the fourth embodiment, the same effect as in the third embodiment can be obtained. Moreover, similarly to the second embodiment, it is possible to highly accurately estimate the angle of arrival of an incoming wave other than the direct wave.

Although the receiving apparatus of the present invention has been described above, the receiving method implemented by the above-described receiving apparatus also forms part of the present invention.
Hereinafter, reception methods respectively corresponding to the above-described first to fourth embodiments will be described as fifth to eighth embodiments.

Embodiment 5 FIG.
The fifth embodiment is a receiving method corresponding to the first embodiment.
FIG. 13 shows a processing procedure in the receiving method according to the fifth embodiment.
The receiving method shown in FIG. 13 includes a wireless receiving step ST11, a transmission channel estimation step ST12, a delay time estimation step ST13, a combining step ST14, a reception power estimation step ST15, and an arrival angle estimation step ST16.

  The processing in the wireless reception step ST11 is the same as the processing performed by the wireless reception units 11-1 and 11-2 in FIG. The processing in the transmission path estimation step ST12 is the same as the processing performed by the transmission path estimation units 12-1 and 12-2 in FIG. The processing in the delay time estimation step ST13 is the same as the processing performed by the delay time estimation unit 13 in FIG. The processing in the combining step ST14 is the same as the processing performed by the combining unit 14 in FIG. The processing in received power estimation step ST15 is the same as the processing performed by received power estimation section 15 in FIG. The processing in the arrival angle estimation step ST16 is the same as the processing performed by the arrival angle estimation unit 16 in FIG.

In the wireless reception step ST11, the first and second analog signals obtained by receiving radio waves with the antenna elements 10-1 and 10-2 are respectively frequency-converted into baseband signals, and AD-converted. Thereby, the first and second digital signals Sr ₁ and Sr ₂ are generated.

In the transmission channel estimation step ST12, the frequency characteristics of the transmission channel are estimated from the first and second digital signals Sr ₁ and Sr ₂ generated in the radio reception step ST11, and the first and second transmission channel estimation results are obtained. X ₁ (t) and X ₂ (t) are output.

  The method of estimating the frequency characteristics of the transmission path depends on the transmission method used in the communication system. The present invention can be applied to any transmission scheme. Here, as an example, an OFDM (Orthogonal Frequency Division Multiplex) transmission scheme adopted in many communication systems is adopted, and a DSSS (Direct Sequence) is adopted. A case where a (Spectrum Spread) transmission method is adopted will be described.

FIG. 14 shows a procedure of an example of the processing of the transmission channel estimation step ST12 performed when the OFDM transmission scheme is adopted.
The transmission channel estimation step ST12 shown in FIG. 14 includes an FFT step ST20, a pilot extraction step ST21, a pilot generation step ST22, a division step ST23, and an interpolation step ST24.

In the FFT step ST20, the digital signals Sr ₁ and Sr ₂ generated in the wireless reception step ST11 of FIG. 13 are converted from the time axis to the frequency axis for each symbol by FFT (Fast Fourier Transform), so that each sub signal is converted. Output carrier.
In pilot extraction step ST21, pilot carriers are extracted from the subcarriers output in FFT step ST20.
In pilot generation step ST22, a known pilot carrier is generated on the receiving side.

In the division step ST23, the frequency characteristic of the transmission line acting on the pilot carrier is output by dividing the pilot carrier extracted in the pilot extraction step ST21 by the pilot carrier generated in the pilot generation step ST22.
In the interpolation step ST24, the frequency characteristics of the transmission line acting on the pilot carrier are interpolated in the symbol direction and the subcarrier direction, so that the frequency characteristics of the transmission lines of all subcarriers (transmission channel estimation results) X ₁ (t), X ₂ (t) is obtained.

FIG. 15 shows an exemplary procedure of the process of the transmission channel estimation step ST12 performed when the DSSS transmission scheme is adopted.
The transmission channel estimation step ST12 shown in FIG. 15 includes a pseudo noise sequence generation step ST25, a despreading step ST26, and an FFT step ST27.

In pseudo noise sequence generation step ST25, the same pseudo noise sequence Ns as that used at the time of spreading on the transmission side is generated.
In the despreading step ST26, the sliding correlation between the digital signals Sr ₁ and Sr ₂ generated in the radio receiving step ST11 of FIG. 13 and the pseudo noise sequence Ns is calculated in symbol units.
In the FFT step ST27, the result of the calculation in the despreading step ST26 is converted into a frequency domain by the FFT to obtain frequency characteristics (transmission channel estimation results) X ₁ (t) and X ₂ (t) of the transmission path. .

The first and second transmission path estimation results X ₁ (t) and X ₂ (t) calculated in the transmission path estimation step ST12 are represented by the above equations (1A) and (1B).

Returning to FIG. 13, in the delay time estimation step ST13, one of the first and second transmission path estimation results X ₁ (t) and X ₂ (t), for example, the first transmission path estimation result X ₁ ( Based on t), the delay time of one or more arriving waves included in the radio wave received by the corresponding antenna is estimated.
The estimation at the time of delay is performed by super-resolution processing such as MUSIC processing and ESPRIT processing. Here, the number of arriving waves is L, the delay time of each arriving wave is τ ₁ , τ ₂ ,..., Τ _L , and the estimated delay time is τ (hat) ₁ , τ (hat) ₂ ,. (Hat) _L. It is assumed that τ ₁ <τ ₂ <... <τ _L.

In the combining step ST14, the frequency characteristics X ₁ (t) and X ₂ (t) of the transmission path estimated in the transmission path estimation step ST12 are combined for each of the _M different arrival angle candidates θ _{1 to} θ _M. , M synthesized frequency characteristics X (t, θ ₁ ) to X (t, θ _M ). That is, each arrival angle candidate θ _m (m is any one of 1 to M) is synthesized according to the above equation (3) to generate a synthesized frequency characteristic X (t, θ _m ).

  Although the present embodiment assumes that the number of antenna elements is two, the present invention is applicable even when the number of antenna elements greater than two is three or more.

In the received power estimation step ST15, synthetic step ST14 combined frequency characteristic X calculated by _{(t, θ 1) ~X (} t, θ M) and the delay time estimated in estimating step ST13 delay time τ ₁ ~τ _{From L} , the received power of the incoming wave having the shortest delay time, that is, the received power of the direct wave is estimated for each of the arrival angle candidates θ _{1 to} θ _M. That is, to estimate the reception power of the direct wave for each angle of arrival candidates theta _m. Since the received power is estimated for each of the _M candidates θ _{1 to} θ _M , an estimated value of the M received powers is obtained.

FIG. 16 shows a procedure of a process in an example of the received power estimation step ST15.
The received power estimation step ST15 shown in FIG. 16 includes a mode matrix generation step ST51, a correlation matrix calculation step ST52, a noise power calculation step ST53, and an incoming wave power calculation step ST54.

In Mode matrix generation step ST51, shown from each of the delay time estimated by the delay time estimation step ST13 τ ₁ ~τ _L τ _l (either l is from 1 to L), the above equation (8) Mode It generates a vector a (tau _l), from all of the delay time τ ₁ ~τ _L mode vector a for _{(τ 1) ~a (τ L} ), generates a mode matrix a shown in the above formula (7) I do.

In correlation matrix calculating step ST52, calculated in the synthesis step ST14 synthetic frequency response _{X (t, θ 1) ~X} (t, θ M) on the basis of, M-number of the correlation matrix _{R xx (θ 1) ~R xx} (θ _M ) is calculated. That is, the calculation of Expression (10A) is performed for each arrival angle candidate θ _m to calculate the correlation matrix R _xx (θ _m ).

In the noise power calculation step ST53, a noise power matrix having noise power in the frequency domain as an element based on the synthesized frequency characteristics X (t, θ ₁ ) to X (t, θ _M ) calculated in the synthesis step ST14. N (t, θ _m ) is calculated.
Various methods can be applied to the calculation of the noise power. For example, the variance of the synthetic frequency characteristics X (t, θ ₁ ) to X (t, θ _M ) may be calculated.
The noise power may be calculated from the transmission path estimation results X ₁ (t) and X ₂ (t) for each antenna element instead of the composite frequency characteristics.

In the arriving wave power calculation step ST54, the noise power is calculated in the noise power calculation step ST53 from the correlation matrices R _xx (θ ₁ ) to R _xx (θ _M ) (Equation (10A)) calculated in the correlation matrix calculation step ST52. noise power matrix N (t, theta _m) to remove the effects of noise by subtracting the, to the result of this subtraction, the mode matrix a (equation (7)) and the product of the complex conjugate transposed matrix a ^H By performing a process including an inverse matrix operation, signal correlation matrices S (θ ₁ ) to S (θ _M ) are generated. That is, a signal correlation matrix S (θ _m ) is generated for each arrival angle candidate θ _{m according} to the above equation (13).

In arriving wave power calculating step ST54 Furthermore, from the signal correlation matrix _{S (θ 1) ~S (θ} M), their first row and first column of the component _{s 11} (θ _{1) ~s 11} a (theta _M) It is extracted and output as the power of the direct wave for the arrival angle candidates θ _{1 to} θ _M. That is, the component s ₁₁ (θ _m ) of the first row and first column is extracted from each signal correlation matrix S (θ _m ), and is output as the power of the direct wave for the arrival angle candidate θ _m .
The direct wave powers s ₁₁ (θ ₁ ) to s ₁₁ (θ _M ) calculated in the incoming wave power calculation step ST54 are the estimation results in the reception power estimation step ST15.

  Note that in equations (10B) and (13), it was assumed that the noise was white noise having the same intensity at all frequencies, but the noise power has frequency characteristics as described in the first embodiment. The present invention can also be applied to the case (when the intensity of the noise power differs depending on the frequency).

In the arrival angle estimation step ST16, it is determined which of the direct wave powers s ₁₁ (θ ₁ ) to s ₁₁ (θ _M ) estimated in the reception power estimation step ST15 is the maximum (peak value), arrival angle candidate power is maximum theta _{x (either} θ ₁ ~θ _M), and estimates the arrival angle theta _a direct wave.

  In the fifth embodiment, the same effect as in the first embodiment can be obtained. That is, even when there is a delayed wave with a short delay time, the direct wave can be separated from the delayed wave, and the arrival angle of the direct wave can be estimated with high accuracy. Further, since the noise power is subtracted in the frequency domain, the processing is simple.

Embodiment 6 FIG.
The sixth embodiment is a receiving method corresponding to the second embodiment.
In the fifth embodiment, the arrival angle of a direct wave is calculated. On the other hand, in Embodiment 6, the angle of arrival of an incoming wave other than the direct wave is estimated. For example, the arrival angle of at least one arriving wave other than the direct wave in addition to the direct wave is estimated. In the following, a case will be described in which the arrival angles of all incoming waves, including direct waves, are estimated.

FIG. 17 shows a processing procedure in the receiving method according to the sixth embodiment.
The reception method of FIG. 17 is generally the same as the reception method of FIG. 13, but includes a reception power estimation step ST15b and an arrival angle estimation step ST16b instead of the reception power estimation step ST15 and the arrival angle estimation step ST16.

FIG. 18 shows a processing procedure in an example of the received power estimation step ST15b.
The received power estimating step ST15b is generally the same as the received power estimating step ST15 of FIG. 13, but includes an incoming wave power calculating step ST54b instead of the incoming wave power calculating step ST54.

In arrival wave power calculation step ST54b, the same processing as in arrival wave power calculation step ST54 of the fifth embodiment is performed to generate signal correlation matrices S (θ ₁ ) to S (θ _M ).
Arrival wave power calculating step ST54b further in the generated signal correlation matrix S (theta ₁₎ from ~S (θ _M), their first row and first column of the component _{s 11 (θ 1) ~s 11} (θ M ) to the L row and the L columns of the component _{s LL (θ 1) ~s} extracts _{LL (θ M),} for the arrival angle candidate θ ₁ ~θ _{M, 1} th arrival wave ~L th arrival wave Output as power. That is, all the diagonal components s ₁₁ (θ _m ) to s _LL (θ _m ) are extracted from each signal correlation matrix S (θ _m ), and the first to L-th components of the arrival angle candidate θ _m are extracted. Output as power of incoming wave.

The powers s ₁₁ (θ ₁ ) to s ₁₁ (θ _M ) of the _first arriving wave calculated in the arriving wave power calculation step ST54b to the powers s _LL (θ ₁ ) to s _LL (θ _M ) of the Lth arriving wave ) Is the estimation result in the received power estimation step ST15b.

In the angle-of-arrival estimation step ST16b, the power of the same arriving wave among the powers of the first to Lth arriving waves estimated in the received power estimation step ST15b is compared to determine which is the largest, For each arriving wave, the arrival angle candidate having the highest power is estimated as the arrival angle of the arriving wave. That is, which of the power _{s ll} of l-th arrival wave for the arrival angle candidate _{θ 1 ~θ M (θ 1)} ~s ll (θ M) of determining whether the maximum (peak value), the power There _(either theta ₁ through? _M) arrival angle candidate theta _x is maximum, and estimates the arrival angle theta _al in l th arrival wave. This process is performed for all possible values of l, that is, from l = 1 to l = L.
The processing when 1 is 1 is the same as the processing in the arrival angle estimation step ST16 of the fifth embodiment, and the arrival angle of the direct wave is estimated by this processing.
When 1 is any of 2 to L, the arrival method of each of the second and subsequent arriving waves is estimated.

  As described above, the case has been described in which the arrival angles of all the incoming waves are estimated. However, the arrival angles of only some of the incoming waves may be estimated. For example, the angle of arrival of the direct wave may be estimated, and the angle of arrival of one of the incoming waves other than the direct wave may be estimated.

  The same effects as in the fifth embodiment can be obtained in the sixth embodiment. In addition, it is possible to highly accurately estimate the angle of arrival of an incoming wave other than the direct wave.

Embodiment 7 FIG.
The seventh embodiment is a receiving method corresponding to the third embodiment.
In the fifth embodiment, the generation of the mode matrix (A) and the signal correlation matrix (S (θ) are performed using all the delay times estimated in the delay time estimation step ST13, that is, using the incoming waves of all the delay times. )). On the other hand, in the seventh embodiment, an incoming wave whose delay time is equal to or longer than the threshold is not used for generating a mode matrix, calculating a signal correlation matrix, and the like. This is done to reduce the amount of calculation.

FIG. 19 shows a procedure of processing in the receiving method according to the seventh embodiment.
The receiving method of FIG. 19 is generally the same as the receiving method of FIG. 13, but includes a received power estimation step ST15c and an angle of arrival estimation step ST16c instead of the received power estimation step ST15 and the angle of arrival estimation step ST16.

FIG. 20 shows a procedure of processing in an example of the received power estimation step ST15c.
The received power estimation step ST15c in FIG. 20 includes a mode matrix generation step ST51c, a correlation matrix calculation step ST52c, a noise power calculation step ST53c, an arrival wave power calculation step ST54c, a threshold setting step ST56, and a delayed wave removal step ST57.

In the threshold setting step ST56, a delay time threshold τ _th is set. The delay time threshold τ _th is used to remove a relatively long delay time. The delay time threshold τ _th may be determined in advance, or may be determined based on the delay times τ _{1 to} τ _L estimated in the delay time estimation step ST13.
In the case where it is determined in advance, it is desirable that the number of delay times after removal does not exceed the upper limit value, for example, taking into account the radio wave environment. Also in the case where the delay time is determined based on the estimated delay times τ _{1 to} τ _L , it is desirable to determine the number of delay times after removal so as not to exceed the upper limit value.

In Mode matrix generation step ST51c, the estimated delay time τ ₁ ~τ _L by the delay time estimation step ST13, based on the delay time threshold value tau _th set in the threshold value setting step ST56, than the delay time threshold value tau _th Generates a mode matrix A ′ from a short delay time.

Specifically, among the delay times τ _{1 to} τ _L , τ _{1 to} τ _q (q is any of 1 to L) is shorter than the threshold τ _th , and τ _{(q + 1) to} τ _L is greater than or equal to the threshold τ _th Then, a mode vector a (τ ₁ ) shown in the above equation (8) is generated from each of τ ₁ (1 is any of 1 to q) of the delay times τ _{1 to} τ _q , and the threshold τ _th from all of the delay time shorter than τ ₁ ~τ _q mode vector a for _{(τ 1) ~a (τ q} ), and generates a mode matrix a 'shown in the above formula (17).

In the delayed wave removing step ST57, a component of the delayed wave whose delay time is equal to or greater than the threshold τ _th is removed from the combined frequency characteristics X (t, θ ₁ ) to X (t, θ _M ) calculated in the combining step ST14. Then, the synthetic frequency characteristics X ′ (t, θ ₁ ) to X ′ (t, θ _M ) for the arriving wave whose delay time is shorter than the threshold τ _th , that is, the first to qth arriving waves are generated. . That is, a component of a delay wave whose delay time is equal to or greater than the threshold τ _th is removed from each of the synthesized frequency characteristics X (t, θ _m ) (m is any one of 1 to M), and the delay time is reduced to the threshold τ _th A synthetic frequency characteristic X ′ (t, θ _m ) is generated for an incoming wave shorter than that.

Digital filters such as an FIR filter and an IIR filter are used to remove the delayed wave. Since each of the above-described synthesized frequency characteristics X (t, θ ₁ ) to X (t, θ _M ) is a frequency characteristic of a transmission path, a desired component in a delay time domain is obtained by performing a filtering process in a frequency domain. Can be extracted.

The synthesized frequency characteristic X ′ (t, θ _m ) of the arriving wave whose delay time is shorter than the threshold τ _th , that is, the first to qth arriving waves, is represented by the above equation (18). Become.

In the correlation matrix calculation step ST52c, M correlation matrices R ′ _xx (θ ₁ ) are obtained from the synthesized frequency characteristics X ′ (t, θ ₁ ) to X ′ (t, θ _M ) generated in the delayed wave removal step ST57. ) To R ′ _xx (θ _M ) are calculated. That is, the calculation of the above equation (20) is performed for each arrival angle candidate θ _m to calculate the correlation matrix R ′ _xx (θ _m ).

In the noise power calculating step ST53c, the noise having the noise power in the frequency domain as an element is obtained from the synthesized frequency characteristics X ′ (t, θ ₁ ) to X ′ (t, θ _M ) generated in the delayed wave removing step ST57. The power matrix N ′ (t, θ _m ) is calculated.

In the arrival wave power calculation step ST54c, the noise power matrix N calculated in the noise power calculation step ST53c is calculated from the correlation matrices R ′ _xx (θ ₁ ) to R ′ _xx (θ _M ) calculated in the correlation matrix calculation step ST52c. '(t, theta _m) to remove the effects of noise by subtracting the, to the result of this subtraction, the mode matrix a' opposite the product (formula (17)) and its complex conjugate transposed matrix a ^'H By performing processing including matrix operation, signal correlation matrices S ′ (θ ₁ ) to S ′ (θ _M ) are generated. That is, a signal correlation matrix S ′ (θ _m ) is generated for each arrival angle candidate θ _{m according} to the above equation (21).

In arriving wave power calculating step ST54c, further signal correlation matrix S '(θ ₁₎ ~S' from (theta _M), their first row and first column of the component _{s 11 (θ 1) ~s 11} (θ _M ) is extracted and output as direct wave power for the arrival angle candidates θ _{1 to} θ _M. That is, the component s ₁₁ (θ _m ) of the first row and the first column is extracted from each signal correlation matrix S ′ (θ _m ), and is output as the power of the direct wave for the arrival angle candidate θ _m .
The direct wave power s ₁₁ (θ ₁ ) to s ₁₁ (θ _M ) calculated in the incoming wave power calculation step ST54c is the estimation result in the reception power estimation step ST15c.

In the arrival angle estimation step ST16c, it is determined which of the direct wave powers s ₁₁ (θ ₁ ) to s ₁₁ (θ _M ) estimated in the reception power estimation step ST15c is the maximum (peak value), arrival angle candidate power is maximum theta _{x (either} θ ₁ ~θ _M), and estimates the arrival angle theta _a direct wave.

  In the seventh embodiment, the same effect as in the third embodiment can be obtained. That is, in addition to the same effects as in the fifth embodiment, the amount of calculation can be suppressed even in an environment where a large number of incoming waves are included, and the angle of arrival of the direct wave can be estimated with high accuracy. .

Embodiment 8 FIG.
The eighth embodiment is a receiving method corresponding to the fourth embodiment.
In the seventh embodiment, the arrival angle of a direct wave is calculated. On the other hand, in the eighth embodiment, the arrival angles of the incoming waves other than the direct wave among the incoming waves whose delay time is shorter than the threshold τ _th are estimated. For example, the arrival angle of at least one arriving wave other than the direct wave in addition to the direct wave is estimated. In the following, a case will be described in which the arrival angles of all arriving waves, including direct waves, whose delay time is shorter than the threshold τ _th are estimated.

FIG. 21 shows a receiving method according to the eighth embodiment.
The reception method of FIG. 21 is generally the same as the reception method of FIG. 19, but includes a reception power estimation step ST15d and an arrival angle estimation step ST16d instead of the reception power estimation step ST15c and the arrival angle estimation step ST16c.

FIG. 22 shows a processing procedure in an example of the received power estimation step ST15d.
The received power estimation step ST15d is generally the same as the received power estimation step ST15c in FIG. 20, but includes an incoming wave power calculation step ST54d instead of the incoming wave power calculation step ST54c.

In arrival wave power calculation step ST54d, the same processing as in arrival wave power calculation step ST54c of the seventh embodiment is performed to generate signal correlation matrices S ′ (θ ₁ ) to S ′ (θ _M ).
Arrival wave power calculating step ST54d further in, the generated signal correlation matrix _{S '(θ 1) ~S'} (θ M), their first row and first column of the component _{s 11 (θ 1) ~s 11} ( theta _M) ~ q-th row and the q column component _{s qq} a _{(θ 1) ~s qq (θ} M), for the arrival angle candidate theta ₁ through? _M, the first incoming wave ~q th arrival wave Output as electric power. That is, from each signal correlation matrix S ′ (θ _m ), all the diagonal components s ₁₁ (θ _m ) to s _qq (θ _m ) are converted into the first to q-th arriving waves for the arrival angle candidate θ _m. Output as power.
The powers s ₁₁ (θ ₁ ) to s ₁₁ (θ _M ) of the first arriving wave and the powers s _qq (θ ₁ ) to s _qq (θ _{M of the} arriving wave calculated in the arriving wave power calculating step ST54d. ) Is the estimation result in the received power estimation step ST15d.

In the angle-of-arrival estimation step ST16d, the power of the same arriving wave is compared among the powers of the first to qth arriving waves estimated in the received power estimation step ST15d, and it is determined which is the largest. For each arriving wave, the arrival angle candidate having the highest power is estimated as the arrival angle of the arriving wave. That is, the angle of arrival candidates theta ₁ of the l-th arrival of the through? _M power _{s ll} (theta _{1) ~s} any _ll of the (theta _M) is to determine the maximum (peak value), the power maximum a is the angle of arrival candidates theta _{x (either} θ ₁ ~θ _M), and estimates the arrival angle theta _al in l th arrival wave. This process is performed for all possible values of l, that is, for all cases from l = 1 to l = q.
The processing when l is 1 is the same as the processing in the arrival angle estimation step ST16c of the fifth embodiment, and the arrival angle of the direct wave is estimated by this processing.
When 1 is any of 2 to q, the arrival method of each of the second and subsequent arriving waves is estimated.

  As described above, the case has been described in which the arrival angles of all the arriving waves whose delay time is shorter than the threshold value are estimated. However, the arrival angle may be estimated for only a part of the arriving waves whose delay time is shorter than the threshold value. . For example, the angle of arrival of the direct wave may be estimated, and the angle of arrival of one of the incoming waves other than the direct wave whose delay time is shorter than the threshold may be estimated.

  The same effects as in the fourth embodiment can be obtained in the eighth embodiment. That is, in addition to the same effects as in the seventh embodiment, similarly to the sixth embodiment, it is possible to highly accurately estimate the angle of arrival of an incoming wave other than the direct wave.

The eighth embodiment has been described as a modification to the seventh embodiment. Similar modifications can be made to the fifth embodiment.
Further, modifications similar to those described in the first to fourth embodiments can be added to the fourth to eighth embodiments.

Embodiment 9 FIG.
In the first, second, third and fourth embodiments, each part (the part shown as a functional block) of the receiving apparatus shown in FIGS. 1, 7, 9 and 11 can be realized by a processing circuit. . The processing circuit may be dedicated hardware or a CPU that executes a program stored in a memory.
For example, the functions of the respective parts in FIGS. 1, 7, 9 and 11 may be realized by separate processing circuits, or the functions of a plurality of parts may be realized by a single processing circuit. .

  When the processing circuit is a CPU, the function of each part of the receiving device is realized by software, firmware, or a combination of software and firmware. Software or firmware is described as a program and stored in a memory. The processing circuit realizes the function of each unit by reading and executing the program stored in the memory. That is, the receiving device stores a program that, when realized by the processing circuit, causes the functions of the respective parts illustrated in FIGS. 1, 7, 9, and 11 to be executed as a result. With memory. In addition, it can be said that these programs cause a computer to execute a processing method or a procedure in a receiving method performed by the receiving device.

In addition, among the functions of each part of the receiving device, a part may be realized by dedicated hardware, and a part may be realized by software or firmware.
As described above, the processing circuit can realize each of the above functions by hardware, software, firmware, or a combination thereof.

FIG. 23 shows an example of a configuration in which all the functions of the receiving device are realized by a computer including a single CPU (indicated by reference numeral 100) in which the processing circuit is a CPU. It is shown together with -2.
The computer 100 shown in FIG. 23 includes a CPU 101, a memory 102, input units 103-1 and 103-2, and an output unit 104, which are connected by a bus 105.
The antenna elements 10-1 and 10-2 are connected to the input units 103-1 and 103-2.
The signals received by the antenna elements 10-1 and 10-2 are supplied to the CPU 101 via the input units 103-1 and 103-2.

The CPU 101 operates according to the program stored in the memory 102, and responds to signals input via the input units 103-1 and 103-2 to each of the units of the receiving apparatus according to the first, second, third, or fourth embodiment. After performing the processing, the output unit 104 outputs an output signal obtained as a result of the processing.
The contents and procedure of the processing by the CPU 101 are the same as those described in the first, second, third and fourth embodiments.

  The case has been described in which the computer executes each process in the receiving apparatus according to the first, second, third, and fourth embodiments. Alternatively, the program may be executed by a computer.

  The receiving method described in the receiving apparatus, the processing of each part of the receiving apparatus, or the program that causes a computer to execute each processing in the receiving method and the computer-readable recording medium on which the program is recorded are also described for the receiving apparatus. The same effect as described above can be obtained.

  10-1, 10-2 antenna elements, 11-1, 11-2 radio receiving section, 12-1, 2 transmission path estimating section, 13 delay time estimating section, 14 combining section, 15, 15b, 15c, 15d received power Estimation unit, 16, 16b, 16c, 16d Arrival angle estimation unit, 20-n FFT unit, 21-n pilot extraction unit, 22-n pilot generation unit, 23-n division unit, 24-n interpolation unit, 25-n Pseudo noise sequence generation unit, 26-n despreading unit, 27-n FFT unit, 51, 51c mode matrix generation unit, 52, 52c correlation matrix calculation unit, 53, 53c noise power calculation unit, 54, 54b, 54c, 54d Arrival wave power calculation unit, 56 threshold setting unit, 57 delay wave removal unit, 101 CPU, 102 memory, 103-1, 103-2 input unit, 104 Output, 105 bus.

Claims (10)

A receiving device for receiving a radio wave emitted by a transmitter and estimating an angle of arrival of an incoming wave from the transmitter,
The first to N-th antenna elements are provided corresponding to the first to N-th (N is an integer of 2 or more) antenna elements constituting the array antenna, and the first to N-th antenna elements are obtained by receiving the radio waves with the corresponding antenna elements. A first to an N-th wireless receiving unit that performs frequency conversion and AD conversion on the analog signal to generate first to N-th digital signals;
It is provided corresponding to each of the first to N-th radio reception units, estimates the frequency characteristics of the transmission path from the first to N-th digital signals, and outputs the first to N-th transmission path estimation results. First to N-th transmission path estimating units;
A delay time estimating unit for estimating a delay time of one or more arriving waves included in the radio wave by super-resolution processing from one of the first to Nth transmission path estimation results;
Assuming that the arrival angle is one of M (M is an integer of 2 or more) different arrival angle candidates, for each of the arrival angle candidates, the first through N-th corresponding to the arrival angle candidate are considered. Calculating a composite frequency characteristic for the M arrival angle candidates by adding a correction amount for canceling the phase difference between the frequency characteristics of the transmission lines and adding the frequency characteristics of the first to Nth transmission lines. A synthesizing part,
A reception power estimator configured to estimate reception power of a direct wave for the M arrival angle candidates from the synthesized frequency characteristics of the M arrival angle candidates and the delay time;
Determining which of the reception powers of the direct waves estimated for the M arrival angle candidates is the largest, and estimating the arrival angle candidate having the largest reception power as the arrival angle of the direct wave A receiving device comprising: a unit;   The receiving apparatus according to claim 1, wherein the super-resolution processing is one of a MUSIC processing and an ESPRIT processing. The reception power estimator,
A mode matrix generation unit that generates a mode matrix from the delay time,
A correlation matrix calculation unit that calculates a correlation matrix for the M arrival angle candidates from the synthesized frequency characteristics for the M arrival angle candidates;
A noise power calculation unit that calculates a noise power matrix having noise power in the frequency domain as an element from the composite frequency characteristics of the M arrival angle candidates;
The influence of noise is removed by subtracting the noise power matrix from the correlation matrix for the M arrival angle candidates, and a signal correlation matrix is generated using the result of the subtraction and the mode matrix. And an incoming wave power calculator that separates the waves in a delay time domain and extracts the received power of the incoming wave having the shortest delay time as the received power of the direct wave. Receiver. The arriving wave power calculation unit, among the arriving waves separated by generating the signal correlation matrix, the reception power of at least one arriving wave other than the direct wave, the M arriving angle candidates Presumed,
The arrival angle estimation unit determines which of the received powers estimated for the M arrival angle candidates is the largest for the at least one arriving wave other than the direct wave, and the received power is maximized. The reception device according to claim 3, wherein the arrival angle candidate is estimated as the arrival angle of the at least one arriving wave. The reception power estimator,
A threshold setting unit for setting a delay time threshold,
Of the delay time, a mode matrix generation unit that generates a mode matrix from a delay time shorter than the delay time threshold,
From the synthetic frequency characteristics of the M arrival angle candidates, the component of the arriving wave whose delay time is equal to or longer than the delay time threshold is removed, and the delay time of the M arrival angle candidates is set to the delay time threshold A delayed wave removing unit that outputs a synthesized frequency characteristic of an incoming wave shorter than
For the M arrival angle candidates, the combined frequency characteristics for short incoming waves than the delay time is the delay time threshold value, the correlation matrix calculator for calculating a correlation matrix for the M arrival angle candidates,
For the M arrival angle candidates, a noise power calculation unit that calculates a noise power matrix having noise power in the frequency domain as an element from a synthesized frequency characteristic of the arriving wave whose delay time is shorter than the delay time threshold. ,
By removing the effect of noise by subtracting the noise power matrix from the correlation matrix for the M arrival angle candidates, by generating a signal correlation matrix using the result of the subtraction and the mode matrix, 3. An incoming wave power calculation unit that separates an incoming wave in a delay time domain and extracts received power of the incoming wave having the shortest delay time as received power of a direct wave. Receiving device. The arriving wave power calculation unit, among the arriving waves separated by generating the signal correlation matrix, the reception power of at least one arriving wave other than the direct wave, the M arriving angle candidates Presumed,
The arrival angle estimating unit, of the at least one arriving wave other than the direct wave among the arriving waves whose delay time is shorter than the delay time threshold, the reception power of the M arrival angle candidates estimated for the M arrival angle candidates. The receiving apparatus according to claim 5, wherein it is determined which one is the maximum, and an arrival angle candidate having the maximum received power is estimated as the arrival angle of the at least one arriving wave.   The receiving device according to claim 3, wherein a component of a first row and a first column of the signal correlation matrix is extracted as received power of the direct wave. A receiving method for receiving a radio wave emitted by a transmitter and estimating an angle of arrival of an incoming wave from the transmitter,
By performing frequency conversion and AD conversion on the first to Nth analog signals obtained by receiving the radio waves with the first to Nth (N is an integer of 2 or more) antenna elements constituting the array antenna A wireless receiving step of generating first to Nth digital signals;
A transmission path estimation step of estimating a frequency characteristic of a transmission path from the first to Nth digital signals and outputting first to Nth transmission path estimation results;
A delay time estimating step of estimating a delay time of one or more arriving waves included in the radio wave by super-resolution processing from one of the first to Nth transmission path estimation results;
Assuming that the arrival angle is one of M (M is an integer of 2 or more) different arrival angle candidates, for each of the arrival angle candidates, the first through N-th corresponding to the arrival angle candidate are considered. By adding a correction amount for canceling the phase difference between the frequency characteristics of the transmission lines and adding the frequency characteristics of the first to Nth transmission lines, the composite frequency characteristics of the M arrival angle candidates are calculated. A synthesizing step,
A reception power estimation step of estimating reception power of a direct wave for the M arrival angle candidates from the synthesized frequency characteristics of the M arrival angle candidates and the delay time;
Determining which of the reception powers of the direct waves estimated for the M arrival angle candidates is the largest, and estimating the arrival angle candidate having the largest reception power as the arrival angle of the direct wave A receiving method comprising:   A program for causing a computer to execute processing of each step of the receiving method according to claim 8.   A computer-readable recording medium recording the program according to claim 9.

Images