JP2008249333A - Arrival direction estimating device and wireless communication apparatus employing the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an arrival direction estimating device capable of estimating an arrival direction of an arriving wave with accuracy higher than that of an antenna pattern table. <P>SOLUTION: A replica signal generating section 105 and a propagation parameter estimating section 106 are configured to repeatedly estimate the propagation parameter of the arrival wave which arrives at an array antenna 101, at the same resolution as an angle step of the antenna pattern table. If the propagation parameter is converged, the replica signal generating section 105 and the propagation parameter estimating section 106 output the converged propagation parameter to an angle high-resolution processing section 108 through a convergence determining section 107. The angle high-resolution processing section 108 estimates the arrival direction of the arriving wave at the resolution higher than that of the angle step of the antenna pattern table by using the converged propagation parameter. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an arrival direction estimation apparatus that estimates an arrival direction of an incoming wave and a wireless communication apparatus including the arrival direction estimation apparatus.

  2. Description of the Related Art Recently, in a wireless communication system, in order to further increase subscriber capacity and further increase transmission speed, time domain and spatial domain represented by AAA (Adaptive Antenna Array) and MIMO (Multiple Input Multiple Output). Application of spatio-temporal signal processing technology using and is being studied.

  In order to evaluate these techniques with high accuracy, a spatio-temporal path model that can simultaneously handle the time direction and the direction of arrival of incoming radio waves is essential. In order to propose this model, it is necessary to perform a detailed analysis of the spatio-temporal multipath propagation path in the real environment.

  In this way, MUSIC (Multiple Signal Classification) (Non-patent Document 1) and ESPRIT (Estimation of Signal Parameters via Rotational Institial Induction) There is an algorithm called Non-Patent Document 2).

However, the accuracy of these algorithms deteriorates in an environment such as a low SNR (Signal to Noise Ratio) when the number of snapshots is small or when a coherent wave arrives. In these algorithms, since the shape of the array is also limited, recently, it is easy to make parameters multidimensional regardless of the shape of the array. Therefore, SAGE (Space Alternating Generalized Em based on maximum likelihood estimation is used. algorithm) (Non-Patent Document 3).
ROSchmidt “Multiple Emitter Location and Signal Parameter Estimation,” IEEE Trans., Vol.AP-34, No.3, pp.276-280 (Mar. 1986). R.Roy and T.Kailath “Estimatio of Signal Parameters via Rotational Invariance Techniques,” IEEE Trans., Vol.ASSP-37, pp.984-995 (July 1989). Bernard H. Fleury, Patrik Jourdan, Andreas Stucki “High-Resolution Channel Parameter Estimation for MIMO Applications Using the SAGE Algorithm”.

  However, since the SAGE algorithm sequentially detects incoming waves by angle scanning, the limit resolution in estimating the arrival angle depends on the array response vector. Then, considering that the SAGE algorithm is used in an actual apparatus, the array response vector varies depending on various error factors. For example, there may be errors due to gain / phase variations in the analog circuit section of each port of the receiver, errors in the position of each element, or manufacturing errors in the element shape such as the array element length.

  Therefore, by estimating / measuring the error factor and removing the influence, an array response vector that compensates for the influence of the error can be obtained. This is called array calibration of a direction-of-arrival system using an array antenna, and there is a method of directly measuring an array response vector in an anechoic chamber or the like in order to compensate for the above error factors. As a result, data is acquired for each certain angle step, stored as an antenna pattern table, and referred to when the corresponding angle is estimated.

  However, when estimating the direction of arrival with high resolution, it is necessary to create an antenna pattern table that has a small angle step and covers the entire search range, but this is difficult to measure when there are many array elements. There is.

  In addition, when an actual machine holds this huge antenna pattern table, there is a problem that the memory capacity increases.

  Accordingly, the present invention has been made to solve such a problem, and an object thereof is to provide an arrival direction estimation device capable of estimating the arrival direction of an incoming wave with higher accuracy than the accuracy of an antenna pattern table. It is.

  Another object of the present invention is to provide a wireless communication device including an arrival direction estimation device capable of estimating the arrival direction of an incoming wave with higher accuracy than the accuracy of an antenna pattern table.

  According to the present invention, the arrival direction estimation apparatus includes an array antenna, a reception unit, and first and second estimation units. An array antenna is composed of a plurality of antenna elements and receives incoming waves. The receiving means generates a reception signal when the array antenna receives an incoming wave. The first estimating means uses the received signal generated by the receiving means and an antenna pattern table that is a response vector of a plurality of antenna elements when the array antenna receives the incoming wave. A first estimation process is performed to estimate the propagation parameter with the first resolution. The second estimation means executes a second estimation process for estimating the arrival direction of the incoming wave with a second resolution higher than the first resolution using the propagation parameter estimated by the first estimation means.

  Preferably, the first estimation unit starts the first estimation process using the initial value of the propagation parameter, and repeatedly executes the first estimation process until the propagation parameter converges to estimate the propagation parameter. The second estimating means performs the second estimating process using the converged propagation parameter to estimate the arrival direction of the incoming wave.

  Preferably, the first estimation unit starts the first estimation process using the initial value of the propagation parameter, and repeatedly executes the first estimation process until the propagation parameter converges to estimate the propagation parameter. The second estimation means executes the second estimation process every time the first estimation process is executed, and uses the propagation parameter estimated by the first estimation means until the propagation parameter converges. Estimate the direction of arrival.

  Preferably, the antenna pattern table is created for the same angular step as the first resolution.

  Preferably, the antenna pattern table is a pattern table obtained by interpolating the antenna pattern table created for the same angular step as the first resolution.

  Preferably, the initial value of the propagation parameter is a parameter arbitrarily set in advance.

  Preferably, the initial value of the propagation parameter includes a propagation parameter detected so that the transmission signal approaches the reception signal.

  Preferably, the initial value of the propagation parameter includes a propagation parameter detected based on a correlation value between the reception signal and the transmission signal.

  According to the invention, the wireless communication device includes an arrival direction estimation device and a transmission unit. An arrival direction estimation apparatus consists of an arrival direction estimation apparatus of any one of Claims 1-8. The transmission means forms a beam in the direction of arrival estimated by the direction-of-arrival estimation device and transmits a signal.

  In the arrival direction estimation apparatus according to the present invention, the first estimation means estimates the propagation parameter with the first resolution using the antenna pattern table, and the second estimation means is estimated by the first estimation means. The direction of arrival is estimated with the second resolution using the propagation parameter.

  Therefore, according to the present invention, the direction of arrival can be estimated with higher resolution than the angle step of the antenna pattern table.

  As a result, it is not necessary to mount an antenna pattern table having an enormous amount of data, and an increase in memory capacity can be suppressed.

  Embodiments of the present invention will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals and description thereof will not be repeated.

[Embodiment 1]
FIG. 1 is a schematic block diagram showing the configuration of an arrival direction estimation apparatus according to Embodiment 1 of the present invention. The arrival direction estimation apparatus 100 according to Embodiment 1 includes an antenna 101, a reception unit 102, a known signal generation unit 103, a propagation parameter initial value generation unit 104, a replica signal creation unit 105, a propagation parameter estimation unit 106, , A convergence determination unit 107, an angle high resolution processing unit 108, and a propagation parameter storage unit 109.

  The array antenna 101 includes K (K is an integer of 2 or more) antenna elements 101-1 to 101 -K. Array antenna 101 receives incoming waves arriving from L (L is a positive integer) paths by K antenna elements 101-1 to 101 -K, and receives the received signals of the received incoming waves. Output to the unit 102.

The receiving unit 102 receives K baseband signals y _k (corresponding to the K antenna elements 101-1 to 101-K) based on the received signals received from the K antenna elements 101-1 to 101-K. t) (k = 1 to K) is generated, and the generated K baseband signals y _k (t) are output to the replica signal creation unit 105.

The known signal generation unit 103 generates a known signal s (t) composed of a transmission signal, and outputs the generated known signal s (t) to the replica signal creation unit 105. Propagation parameter initial value generation section 104 has initial values τ _{10 to} τ _L0 of delay times τ _{1 to} τ _L of incoming waves arriving through the _{first to} Lth paths, and incoming waves arriving through the _{first to} Lth paths. Initial values θ _{10 to} θ _L0 , ψ _{10 to} ψ _L0 , and the incoming waves arriving from the _{first to} Lth paths, respectively, of the arrival angles of azimuth angles θ _{1 to} θ _L and elevation angles ψ _{1 to} ψ _L When the initial values α _{10 to} α _L0 of the complex amplitude attenuation α _{1 to} α _L are held and the estimation of the propagation parameter including the delay time of the arrival wave, the arrival angle, and the complex amplitude attenuation is started, the held delay time τ _{1 to} τ _L initial values τ _{10 to} τ _L0 , arrival angle initial values θ _{10 to} θ _L0 , ψ _{10 to} ψ _L0 , and complex amplitude attenuation α _{1 to} α _L initial values α _{10 to} α _L0 are used as replica signals. The data is output to the creation unit 105.

The replica signal creation unit 105 receives an antenna pattern table V _k (ψ _l , θ _l ) (k = _{1 to} K, K) that is a response vector of the K antenna elements 101-1 to 101-K when receiving the incoming wave. l = 1 to L), K baseband signals y _k (t) are received from the receiver 102, the known signal s (t) is received from the known signal generator 103, and the propagation parameter initial value generator 104 To initial values τ _{10 to} τ _{L0 of} delay times τ ₁ to τ _L , initial values of arrival angles θ _{10 to} θ _L0 , ψ _{10 to} ψ _L0 , and initial values α _{10 to} α _L of complex amplitude attenuation α _{1 to} α _L. Receive _L0 . Then, the replica signal creation unit 105 includes initial values τ _{10 to} τ _{L0 of} delay times τ _{1 to} τ _L , initial values of arrival angles θ _{10 to} θ _L0 , ψ _{10 to} ψ _L0 , and complex amplitude attenuation α _{1 to} α. initial value the initial value alpha ₁₀ to? _L0 of _{L [τ 10, θ 10,} ψ 10, α 10], [τ 20, θ 20, ψ 20, α 20], ···, [τ l0, θ l0 , Ψ ₁₀ , α ₁₀ ],..., [Τ _L0 , θ _L0 , ψ _L0 , α _L0 ].

In addition, the replica signal creation unit 105 generates initial values of the K baseband signals y _k (t), the antenna pattern table V _k (ψ _l , θ _l ), the known signal s (t), and the delay times τ _{1 to} τ _L. Based on the values τ _{10 to} τ _L0 , the initial values of arrival angles θ _{10 to} θ _L0 , ψ _{10 to} ψ _L0 , and the initial values α _{10 to} α _L0 of the complex amplitude attenuation α _{1 to} α _L , The replica signal x _{k, l} (t) (l = 1 to L) of the l-th path in the k-th antenna element 101-k is created.

Then, the replica signal creation unit 105 creates the created replica signal x _{k, l} (t), the known signal s (t), and the propagation parameters [τ _ln , θ _ln , ψ _ln , α _ln ] (n is Is output to the propagation parameter estimation unit 106.

Further, upon receiving the propagation parameter [τ _ln , θ _ln , ψ _ln , α _ln ] of the l-th path estimated from the propagation parameter estimation unit 106, the replica signal creation unit 105 receives the propagation parameter [τ _ln , Θ _ln , ψ _ln , α _ln ] update the propagation parameter of the l-th path.

When replica signal creation section 105 receives stop signal STP for stopping propagation parameter estimation from convergence determination section 107, replica signal creation section 105 stops creating replica signal x _{k, l} (t) and propagates parameter estimation section 106. Output nothing. As a result, the propagation parameter estimation operation in the propagation parameter estimation unit 106 is stopped.

The propagation parameter estimation unit 106 holds the antenna pattern table V _k (ψ _l , θ _l ) (k = 1 to K, l = 1 to L), and the replica signal x _{k, l} (t ) (L = 1 to L), the known signal s (t), and the initial values [τ ₁₀ , θ ₁₀ , φ ₁₀ , α ₁₀ ] of the propagation parameters. Then, the propagation parameter estimation unit 106 is based on the replica signal x _{k, l} (t) (l = 1 to L), the known signal s (t), and the propagation parameter [τ _ln , θ _ln , ψ _ln , α _ln ]. Thus, the delay time τ _l , the arrival angles θ _l , ψ _l and the complex amplitude attenuation α _l are estimated by the method described later, and the propagation parameter estimation count n is counted. Further, the propagation parameter estimation unit 106 uses the estimated delay time τ _l , arrival angles θ _l , ψ _l and complex amplitude attenuation α _l as a replica signal for the propagation parameters [τ _ln , θ _ln , ψ _ln , α _ln ]. The output is output to the creation unit 105, and the propagation parameters [τ _ln , θ _ln , ψ _ln , α _ln ] and the estimated number n are output to the convergence determination unit 107.

Convergence determination unit 107, the propagation parameter estimation unit 106 from the propagation parameters _{[τ ln, θ ln, ψ} ln, α ln] undergo and the estimated number n, by a method described later, propagation parameters _{[τ ln, θ ln, ψ} ln , Α _ln ] has converged. When the convergence determination unit 107 determines that the propagation parameter has converged, the angle high resolution processing unit 108 uses the propagation parameter [τ _ln , θ _ln , ψ _ln , α _ln ] received last from the propagation parameter estimation unit 106. And a stop signal STP for stopping the estimation of the propagation parameter is output to the replica signal creation unit 105.

Based on the propagation parameters [τ _ln , θ _ln , ψ _ln , α _ln ] received from the convergence determination unit 107, the angle high resolution processing unit 108 performs the arrival direction (azimuth angle θ and elevation angle) of the incoming wave by a method described later. and the estimated arrival direction (consisting of azimuth angle θ and elevation angle ψ) is output to propagation parameter storage section 109.

  The propagation parameter storage unit 109 stores the arrival direction (consisting of the azimuth angle θ and the elevation angle ψ) of the incoming wave received from the angle high resolution processing unit 108.

Details of the propagation parameters will be described. FIG. 2 is a conceptual diagram of a propagation parameter table held by the replica signal creation unit 105 shown in FIG. When the estimation of the propagation parameter is started, the replica signal creation unit 105 uses the known signal s (t), the initial value of the propagation parameter [θ ₁₀ , ψ ₁₀ , α ₁₀ ], and the first antenna element 101-1. By substituting baseband y ₁ (t) into the following equation, replica signal x _1,1 (t) of the first path in the first antenna element is created.

  Note that l ′ in equation (1) represents a component due to a path other than the l-th path, and the second term in equation (1) represents the l-th path when estimating the propagation parameter of the l-th path. This means that the components other than the path are removed.

Then, when the replica signal creation unit 105 creates the replica signal x _1,1 (t), the created replica signal x _1,1 (t), the known signal s (t), and the initial value of the propagation parameter [ τ ₁₀ , θ ₁₀ , ψ ₁₀ , α ₁₀ ] are output to the propagation parameter estimation unit 106.

The propagation parameter estimation unit 106 receives the replica signal x _1,1 (t), the known signal s (t), and the initial values [τ ₁₀ , θ ₁₀ , ψ ₁₀ , α ₁₀ ] of the propagation parameter from the replica signal creation unit 105. Then, the received replica signal x _1,1 (t), the known signal s (t) and the initial value [θ ₁₀ , ψ ₁₀ , α ₁₀ ] of the propagation parameter are substituted into the following equation and the evaluation function Z of the delay time τ (Τ) is calculated.

That is, the propagation parameter estimation unit 106 calculates the evaluation function Z (τ) by taking the cross-correlation between the replica signal x _1,1 (t) and the known signal s (t).

  Then, the propagation parameter estimation unit 106 estimates the delay time τ by searching for the delay time τ when the evaluation function Z (τ) is maximized by the following equation.

As a result, the propagation parameter estimation unit 106 obtains an estimated value τ ₁₁ of the delay time τ.

After that, the propagation parameter estimation unit 106 substitutes the replica signal x _1,1 (t), the known signal s (t), and the initial values [τ ₁₁ , θ ₁₀ , α ₁₀ ] of the propagation parameters into the following expressions, An evaluation function Z (θ) of θ is calculated.

That is, the propagation parameter estimation unit 106 uses the estimated delay time value τ ₁₁ estimated by the equations (2) and (3), and uses the replica signal x _1,1 (t) and the estimated delay time value τ _11. The evaluation function Z (θ) is obtained by taking a cross-correlation between the known signal s (t−τ ₁₁ ) shifted by only the antenna pattern table V _k (ψ ₁₀ , θ) with the elevation angle ψ fixed to the initial value ψ _10. Is calculated.

  Then, the propagation parameter estimation unit 106 estimates the azimuth angle θ by searching for the azimuth angle θ when the evaluation function Z (θ) is maximized by the following equation.

Accordingly, the propagation parameter estimation unit 106 obtains an estimated value θ ₁₁ of the azimuth angle θ.

Subsequently, the propagation parameter estimation unit 106 substitutes the replica signal x _1,1 (t), the known signal s (t), and the initial value [τ ₁₁ , ψ ₁₀ , α ₁₀ ] of the propagation parameter into the following equation, and the elevation angle ψ The evaluation function Z (ψ) is calculated.

That is, the propagation parameter estimation unit 106 uses the estimated delay time value τ ₁₁ estimated by the equations (2) and (3), and uses the replica signal x _1,1 (t) and the estimated delay time value τ _11. The evaluation function Z (ψ) is obtained by taking the cross-correlation between the known signal s (t−τ ₁₁ ) shifted by only the antenna pattern table V _k (ψ, θ ₁₀ ) with the azimuth angle θ fixed to the initial value θ _10. ) Is calculated.

  Then, the propagation parameter estimation unit 106 estimates the elevation angle ψ by searching for the elevation angle ψ when the evaluation function Z (ψ) is maximized by the following equation.

Thereby, the propagation parameter estimation unit 106 obtains an estimated value ψ ₁₁ of the elevation angle ψ.

Further, after that, the propagation parameter estimation unit 106 substitutes the replica signal x _1,1 (t), the known signal s (t), and the propagation parameters [τ ₁₀ , θ ₁₁ , ψ ₁₁ ] into the following equation to attenuate the complex amplitude. It calculates the estimated value alpha ₁₁ for alpha.

That is, the propagation parameter estimation unit 106 uses the antenna pattern table V _k (ψ ₁₁ , θ using the estimated value ψ ₁₁ of the elevation angle ψ ₁₁ and the estimated value θ ₁₁ of the azimuth angle θ estimated by the equations (4) to (7). ₁₁ ) and the component of the known signal s (t) are removed from the replica signal x _1,1 (t) to calculate the estimated value α ₁₁ of the complex amplitude attenuation α.

Then, the propagation parameter estimation unit 106 counts the number of propagation parameter estimations as “1”, outputs the counted number of estimations “1” to the convergence determination unit 107, and also estimates the propagation parameter estimation values [τ ₁₁ , θ ₁₁ , ψ ₁₁ , α ₁₁ ] are output to the replica signal creation unit 105.

The replica signal creation unit 105 initially holds a propagation parameter table PTB1 (see FIG. 2), and receives propagation parameter estimation values [τ ₁₁ , θ ₁₁ , ψ ₁₁ , α ₁₁ ] from the propagation parameter estimation unit 106. Then, the initial values [τ ₁₀ , θ ₁₀ , ψ ₁₀ , α ₁₀ ] of the propagation parameter table PTB1 are updated with the estimated values [τ ₁₁ , θ ₁₁ , ψ ₁₁ , α ₁₁ ] of the received propagation parameters, and the propagation parameters are updated. The table PTB1 is updated to the propagation parameter table PTB2 (see FIG. 2).

Thereafter, the replica signal creation unit 105 transmits the known signal s (t), the initial values [τ ₂₀ , θ ₂₀ , ψ ₂₀ , α ₂₀ ] of the propagation parameters, and the baseband y ₁ ( By substituting t) into equation (1), the replica signal x _1,2 (t) of the second path in the first antenna element is created.

Thereafter, the above-described operation is repeated to obtain the estimated values [τ ₂₁ , θ ₂₁ , ψ ₂₁ , α ₂₁ ] of the propagation parameters of the second path. Then, the propagation parameter estimation unit 106 outputs the propagation parameter estimation values [τ ₂₁ , θ ₂₁ , ψ ₂₁ , α ₂₁ ] to the replica signal creation unit 105, and the replica signal creation unit 105 receives the propagation parameter estimation unit 106 from the propagation parameter estimation unit 106. The initial values [τ ₂₀ , θ ₂₀ , ψ ₂₀ , α ₂₀ ] of the propagation parameter table PTB2 are updated with the received estimated values [τ ₂₁ , θ ₂₁ , ψ ₂₁ , α ₂₁ ] of the propagation parameters, and the propagation parameter table PTB2 is updated. The propagation parameter table PTB3 is updated (see FIG. 2).

Thereafter, the above-described operation is repeatedly performed, and the propagation parameter estimation unit 106 estimates the propagation parameters [τ ₃₀ , θ ₃₀ , ψ ₃₀ , α] of the third to Lth paths in the first antenna element 101-1. ₃₀ ] to [τ _L0 , θ _L0 , ψ _L0 , α _L0 ] are output to the replica signal creation unit 105, and the replica signal creation unit 105 updates the propagation parameter table PTB 3 to the propagation parameter table PTBL (see FIG. 2). .

  Further, after that, the above-described operation is repeatedly executed, and n-th estimated values of the propagation parameters of the first to Lth paths in the first to Kth antenna elements 101-1 to 101-K are obtained. Then, the replica signal creation unit 105 finally updates the propagation parameter table PTBL to the propagation parameter table PTBLn (see FIG. 2).

In this way, the propagation parameters are sequentially estimated and updated starting from the initial values [τ ₁₀ , θ ₁₀ , ψ ₁₀ , α ₁₀ ].

Then, the propagation parameter estimation unit 106 transmits the n-th estimated values [τ _1n , θ _1n , ψ _1n , α _1n ] to [τ _Ln , θ _Ln , ψ _Ln , α _Ln ] of the propagation parameters to the convergence determination unit 107. The convergence determination unit 107 outputs the n-th estimated value [τ _1n , θ _1n , ψ _1n , α _1n ] to [τ _Ln , θ _Ln , ψ _Ln , α _Ln ] of the propagation parameter and the estimated number n. Then, it is determined whether or not the propagation parameter has converged. In this case, the convergence determination unit 107 determines whether or not the propagation parameter has converged depending on whether or not the estimated number n has reached a predetermined number. The convergence determination unit 107 also estimates the propagation parameters estimated by n-1 estimations [τ _1n−1 , θ _1n−1 , ψ _1n−1 , α _1n−1 ] to [τ _1n−1 ]. , Θ _1n−1 , ψ _1n−1 , α _1n−1 ] and the nth estimated value [τ _1n , θ _1n , ψ _1n , α _1n ] to [τ _Ln , θ _Ln , ψ _Ln , It is determined whether or not the propagation parameter has converged based on whether or not the difference from α _Ln ] is smaller than a threshold value.

When the convergence determining unit 107 determines that the propagation parameter has converged, the n-th estimated value [τ _1n , θ _1n , ψ _1n , α _1n ] to [τ _Ln , θ _Ln , ψ _Ln , α of the propagation parameter is determined. _Ln ] is output to the angle high resolution processing unit 108.

FIG. 3 is a flowchart for explaining an estimation operation in the propagation parameter estimation unit 106 shown in FIG. When a series of operations is started, the reception unit 102 generates baseband signals y ₁ (t) to y _K (t) and outputs them to the replica signal creation unit 105. The known signal generation unit 103 s (t) is generated and output to the replica signal creation unit 105, and the propagation parameter initial value generation unit 104 outputs the held initial value of the propagation parameter to the replica signal creation unit 105. Then, the replica signal creation unit 105 performs the replica signal x _k, using the above-described method based on the baseband signals y ₁ (t) to y _K (t), the known signal s (t), and the initial values of the propagation parameters _{. l} (t) is created (step S1), and the created replica signal x _{k, l} (t), the known signal s (t), and the initial value of the propagation parameter are output to the propagation parameter estimation unit 106.

The propagation parameter estimation unit 106 receives the replica signal x _{k, l} (t), the known signal s (t), and the initial value of the propagation parameter from the replica signal creation unit 105. The propagation parameter estimation unit 106 sets the propagation parameter estimation count n to “1” (step S2), and sets the number of paths l to “1” (step S3).

Then, the propagation parameter estimation unit 106 determines the replica of the l-th path based on the replica signal x _{k, l} (t) received from the replica signal creation unit 105, the known signal s (t), and the initial value of the propagation parameter. The signal x _{k, l} (t) is reproduced (step S4).

Then, the propagation parameter estimation unit 106 _converts the propagation parameter [τ _kl , ψ _kl , θ _kl , α _kl ] of the l-th path into the delay time τ, the arrival angle (azimuth angle θ), and the arrival angle ( The estimated value [τ _kl , ψ _kl , θ _kl , α _kl ] of the estimated l-th path is output to the replica signal creation unit 105. .

The replica signal creation unit 105 receives the propagation parameter estimation value [τ _kl , ψ _kl , θ _kl , α _kl ] of the l-th path from the propagation parameter estimation unit 106 and receives the propagation parameter of the l-th path received. The propagation parameter is updated with the estimated values [τ _kl , ψ _kl , θ _kl , α _kl ] (step S5).

  Thereafter, the replica signal creation unit 105 generates a replica signal using the updated propagation parameter (step S6), sets l to l + 1 (step S7), and determines whether l is greater than L (step S7). Step S8).

  If it is determined in step S8 that l is not greater than L, the series of operations returns to step S4, and step S4 to step S8 described above are performed until it is determined in step S8 that l is greater than L. Is repeatedly executed.

  When it is determined in step S8 that l is greater than L, the propagation parameter estimation unit 106 sets n = n + 1 (step S9), and determines whether n is equal to N (step S10). ).

  When it is determined in step S10 that n is not equal to N, the series of operations returns to step S3, and the above-described steps S3 to S10 are repeated until it is determined in step S10 that n is equal to N. Executed. If it is determined in step S10 that n is equal to N, the series of operations ends.

Next, a method of estimating the direction of arrival in the angle high resolution processing unit 108 will be described. The angle high resolution processing unit 108 holds the antenna pattern table V _k (ψ, θ), and the n-th estimated value [τ _1n , θ _1n , ψ _1n , α _1n ] of the propagation parameter from the convergence determination unit 107. [Τ _Ln , θ _Ln , ψ _Ln , α _Ln ] is received. Then, the angle high resolution processing unit 108 obtains the n-th estimated values [τ _1n , θ _1n , ψ _1n , α _1n ] to [τ _Ln , θ _Ln , ψ _Ln , α _Ln ] of the propagation parameter ( Substituting into 4) and (6), the evaluation function Z (ψ) and the evaluation function Z (θ) are calculated, respectively.

  FIG. 4 is a diagram for explaining a method of estimating the direction of arrival in the angle high resolution processing unit 108 shown in FIG. FIG. 4A shows an evaluation function Z (θ) of the azimuth angle θ. In FIG. 4A, the vertical axis represents the evaluation function value, and the horizontal axis represents the azimuth angle θ.

  FIG. 4B shows a Fourier transform of the sample value of the evaluation function Z (θ) shown in FIG. In FIG. 4B, the vertical axis represents the Fourier transform value of the evaluation function value, and the horizontal axis represents the number of samples. Further, FIG. 4C shows an inverse Fourier transform of the spectrum shown in FIG. 4B from which the high frequency components have been removed. In FIG. 4C, the vertical axis represents the evaluation function value, and the horizontal axis represents the azimuth angle θ. Furthermore, (d) of FIG. 4 shows the estimation result of the arrival direction by the conventional method. In FIG. 4D, the vertical axis represents the evaluation function value, and the horizontal axis represents the azimuth angle θ. The desired value of the azimuth angle θ is −1.2 degrees.

When calculating the evaluation function Z (θ), the angle high resolution processing unit 108 samples the calculated evaluation function Z (θ) with a sample width smaller than the angle step of the antenna pattern table V _k (ψ, θ) ( (See (a) of FIG. 4). The angle high resolution processing unit 108 samples the evaluation function Z (θ) with a sample width of 0.5 degrees, for example.

  Then, the angle high resolution processing unit 108 performs Fourier transform on the sampled evaluation function value. As a result, a spectrum represented by the curve k1 is obtained (see FIG. 4B).

Then, the angle high resolution processing unit 108 removes the high frequency component of the spectrum k1, and performs inverse Fourier transform on the spectrum k1 from which the high frequency component has been removed. As a result, evaluation function values at intervals of 0.5 degrees shown in FIG. 4C are obtained. Then, the angle high resolution processing unit 108 estimates the azimuth angle θ (= −1.5 degrees) at which the evaluation function value shown in FIG. 4C is maximum as the azimuth angle θ _opt of the arrival direction (FIG. 4). (See (c)).

In this way, the angle high resolution processing unit 108 samples the evaluation function Z (θ) with a sample width smaller than the angle step of the antenna pattern table V _k (ψ, θ), and estimates the azimuth angle θ _{opt in the} direction of arrival. To do.

The evaluation function value shown in (d) of FIG. 4 is an evaluation function value when the evaluation function Z (θ) is sampled with the same sample width as the angle step of the antenna pattern table V _k (ψ, θ). As a result, the azimuth angle θ estimated by the conventional method was −2 degrees (see FIG. 4D).

Therefore, when the estimation method according to the present invention is used for a desired value of azimuth angle θ (= −1.2 degrees), an azimuth angle θ of −1.5 degrees is estimated as the azimuth angle θ _{opt of the} arrival direction, The accuracy is higher than the estimated value (= −2 degrees) by the conventional estimation method.

Similarly, the angle high resolution processing unit 108 samples the evaluation function Z (ψ) with a sample width smaller than the angle step of the antenna pattern table V _k (ψ, θ), and estimates the elevation angle ψ _opt in the direction of arrival. .

When the angle high resolution processing unit 108 estimates the azimuth angle θ _opt and elevation angle ψ _opt in the direction of arrival by the method described above, the estimated azimuth angle θ _opt and elevation angle ψ _opt are stored in the propagation parameter storage unit 109. The propagation parameter storage unit 109 stores the azimuth angle θ _opt and the elevation angle ψ _opt .

FIG. 5 is a flowchart for explaining the direction of arrival estimation operation in the first embodiment. When a series of operations are started, according to the flowchart shown in FIG. 3, the angle step of the antenna pattern table V _k (ψ, θ) is set using the initial value of the propagation parameter preset in the propagation parameter initial value generation unit 104. The propagation parameters are estimated with the same resolution until they converge (step S11).

Then, the arrival direction is estimated with a resolution higher than the angle step of the antenna pattern table V _k (ψ, θ) using the converged propagation parameter (step S12). Thereby, the operation for estimating the arrival direction is completed.

As described above, according to the first embodiment, the propagation parameter estimation unit 106 estimates the propagation parameter using the antenna pattern table V _k (ψ, θ) until the propagation parameter converges, and the angle high resolution processing unit. 108 samples the evaluation function of the azimuth angle θ and the elevation angle ψ in the direction of arrival using the propagation parameter estimated by the propagation parameter estimation unit 106 with a higher resolution than the angle step of the antenna pattern table V _k (ψ, θ). Thus, the azimuth angle θ and the elevation angle ψ are estimated. That is, the propagation parameter estimation unit 106 estimates the propagation parameter with the same resolution as the angle step of the antenna pattern table V _k (ψ, θ) until the propagation parameter converges, and the angle high resolution processing unit 108 converges the propagation parameter. Then, using the converged propagation parameters, the evaluation function of the azimuth angle θ and the elevation angle ψ in the arrival direction is sampled with a resolution higher than the angle step of the antenna pattern table V _k (ψ, θ), and the azimuth angle θ and the elevation angle ψ. Is estimated.

  Therefore, according to the present invention, it is possible to estimate the direction of arrival with higher accuracy than the conventional estimation method by using the antenna pattern table created for the same rough angle step as the conventional one.

  The propagation parameter estimation unit 106 constitutes a “first estimation unit”, and the angle high resolution processing unit 108 constitutes a “second estimation unit”.

  Further, the estimation process by the propagation parameter estimation unit 106 constitutes a “first estimation process”, and the estimation process by the angle high resolution processing unit 108 constitutes a “second estimation process”.

[Embodiment 2]
FIG. 6 is a schematic block diagram illustrating a configuration of the arrival direction estimation apparatus according to the second embodiment. Arrival direction estimation apparatus 200 according to Embodiment 2 is obtained by replacing propagation parameter initial value generation section 104 of arrival direction estimation apparatus 100 shown in FIG. 1 with correlation value calculation section 111 and angular spectrum calculation section 112. This is the same as the arrival direction estimation apparatus 100.

In arrival direction estimating apparatus 200, receiving section 102 outputs baseband signals y ₁ (t) to y _K (t) to correlation value calculating section 111, and known signal generating section 103 includes known signal s ( t) is output to replica signal creation section 105 and correlation value calculation section 111.

Correlation value calculation section 111 calculates the cross-correlation between known signal s (t) received from known signal generation section 103 and each of baseband signals y ₁ (t) to y _K (t) received from reception section 102. A composite waveform is generated by combining the calculated K cross-correlations.

  FIG. 7 is a diagram showing a combined waveform synthesized by the correlation value calculation unit 111 shown in FIG. In FIG. 7, the vertical axis represents the correlation value, and the horizontal axis represents the delay time. The synthesized waveform k2 has peaks P1 to P7.

  When the correlation value calculation unit 111 generates the combined waveform shown in FIG. 7, the correlation value calculation unit 111 performs a peak search in order from the peak having the maximum correlation value of the generated combined waveform k2 and extracts a path. In this case, the correlation value calculation unit 111 has a rule of extracting paths as many as a preset number of paths, a rule of extracting paths from the peak to what [dB], and what [dB] or more paths from the noise floor. The path is extracted according to any one of the rules of extracting.

As a result, the correlation value calculation unit 111 extracts, for example, the peaks P1 to P7 as the first path to the seventh path, respectively. Then, the correlation value calculation unit 111 detects the delay times τ _{1 to} τ 7 of the extracted first to seventh paths (= peaks P _{1 to} P ₇ ), and the detected delay times τ _{1 to} τ ₇ , Baseband signals y ₁ (t) to y _K (t) are output to replica signal creation section 105.

In addition, the correlation value calculation unit 111 generates the baseband signals y ₁ (t) to y _K (t), the known signal s (t), the delay times τ _{1 to} τ _{7 of} each path, and the combined waveform k2. Output to the angle spectrum calculation unit 112.

  FIG. 8 is a diagram showing an angle spectrum calculated by the angle spectrum calculation unit 112 shown in FIG. Each angle spectrum in FIG. 8 is an angle spectrum with respect to the azimuth angle.

  In FIG. 8, the vertical axis represents intensity, and the horizontal axis represents angle. A curve k3 indicates an angle spectrum at the peak P1, and a curve k4 indicates an angle spectrum at the peak P7.

The angle spectrum calculation unit 112 holds an antenna pattern table V _k (ψ, θ), a baseband signal y ₁ (t) to y _K (t), a known signal s (t), and a delay time of each path. τ _{1 to} τ ₇ and the combined waveform k2 are received from the correlation value calculation unit 111.

When the angle spectrum calculation unit 112 receives the delay times τ _{1 to} τ ₇ and the combined waveform k2 of each path from the correlation value calculation unit 111, the angle spectrum calculation unit 112 is based on the received delay times τ _{1 to} τ ₇ and the combined waveform k2. The angle spectra k3 and k4 with respect to the azimuth angle θ are calculated for each delay time τ _{1 to} τ ₇ .

Then, the angle spectrum calculation unit 112 performs a peak search for each of the calculated angle spectrums k3 and k4, and detects the peak arrival angles as azimuth angles θ _{1 to} θ ₇ .

The angle spectrum calculating section 112, similarly, based on the delay time τ ₁ ~τ ₇ and synthetic waveform k2, it calculates the angular spectrum for the elevation angle ψ for each delay time τ ₁ ~τ _7, and the calculated An angle spectrum peak search is performed, and the peak arrival angles are detected as elevation angles ψ ₁ to ψ ₇ .

Further, the angle spectrum calculation unit 112 includes a known signal s (t), baseband signals y ₁ (t) to y _K (t), an antenna pattern table V _k (ψ, θ), delay times τ _{1 to} τ _7, and Complex amplitude attenuations α _{1 to} α _K are calculated by substituting the detected arrival angles (= azimuth angles θ _{1 to} θ ₇ and elevation angles ψ _{1 to} ψ ₇ ) into the following equations.

Then, the angle spectrum calculation unit 112 outputs the arrival angles (= azimuth angles θ _{1 to} θ ₇ and elevation angles ψ _{1 to} ψ ₇ ) and complex amplitude attenuation α _{1 to} α _K to the replica signal creation unit 105.

In the arrival direction estimation apparatus 200, the replica signal creation unit 105 receives the known signal s (t) from the known signal generation unit 103, and receives the delay times τ _{1 to} τ ₇ and the baseband signal y ₁ ( t) to y _K (t) and the angle of arrival (= azimuth angles θ _{1 to} θ ₇ and elevation angles ψ _{1 to} ψ ₇ ) and complex amplitude attenuation α _{1 to} α _K are received from the angle spectrum calculation unit 112.

Then, the replica signal creation unit 105 sets delay times τ _{1 to} τ ₇ , arrival angles (= azimuth angles θ _{1 to} θ ₇ and elevation angles ψ _{1 to} ψ ₇ ), and complex amplitude attenuation α _{1 to} α _K as initial propagation parameters. The replica signal x _{k, l} (t) is created in the same manner as in the first embodiment based on the known signal s (t) and the baseband signals y ₁ (t) to y _K (t). While outputting to the propagation parameter estimation part 106, a propagation parameter is updated with the propagation parameter received from the propagation parameter estimation part 106.

  Thereafter, the propagation parameter estimation unit 106 estimates until the propagation parameter converges in the same manner as in the first embodiment, and the angle high resolution processing unit 108 uses the converged propagation parameter in the same manner as in the first embodiment. Estimate the direction of arrival.

As described above, in Embodiment 2, the correlation value calculation unit 111 calculates the correlation value between the baseband signals y ₁ (t) to y _K (t) and the known signal s (t), and delays the delay time. τ _{1 to} τ ₇ are detected, and the detected delay times τ _{1 to} τ ₇ are output to the replica signal creation unit 105. In addition, the angle spectrum calculation unit 112 includes the combined waveform k2 obtained from the correlation value between the baseband signals y ₁ (t) to y _K (t) and the known signal s (t) and the delay times τ _{1 to} τ ₇ . Are detected, the arrival angles (= azimuth angles θ _{1 to} θ ₇ and elevation angles ψ _{1 to} ψ ₇ ) are detected, and the delay times τ _{1 to} τ ₇ and the arrival angles (= azimuth angles θ _{1 to} θ ₇ and elevation angle ψ ₁ are detected. ˜ψ ₇ ), the complex amplitude attenuation α _{1 to} α _K is detected. Then, the angle spectrum calculation unit 112 outputs the arrival angles (= azimuth angles θ _{1 to} θ ₇ and elevation angles ψ _{1 to} ψ ₇ ) and complex amplitude attenuation α _{1 to} α _K to the replica signal creation unit 105.

Then, the replica signal creation unit 105 receives the delay times τ _{1 to} τ ₇ received from the correlation value calculation unit 111 and the arrival angles (= azimuth angles θ _{1 to} θ ₇ and elevation angles ψ ₁ to ˜ received from the angle spectrum calculation unit 112. A replica signal x _{k, l} (t) is created using ψ ₇ ) and complex amplitude attenuation α _{1 to} α _K as initial values of propagation parameters.

Therefore, in the second embodiment, the propagation parameter is set so that the known signal s (t) approaches the reception signal (= baseband signals y ₁ (t) to y _K (t)) actually received by the array antenna 101. Is detected, and the arrival direction of the incoming wave is estimated using the detected initial value of the propagation parameter.

FIG. 9 is a flowchart for explaining the direction of arrival estimation operation in the second embodiment. When a series of operations is started, the correlation value calculation unit 111 and the angle spectrum calculation unit 112, based on the known signal s (t) and the baseband signals y ₁ (t) to y _K (t), The initial value of the propagation parameter is detected so that y ₁ (t) to y _K (t) approach the known signal s (t) (step S21), and the detected initial value of the propagation parameter is sent to the replica signal creation unit 105. Output.

Then, the replica signal creation unit 105 and the propagation parameter estimation unit 106 use the initial values of the propagation parameters received from the correlation value calculation unit 111 and the angle spectrum calculation unit 112 to perform the angle step of the antenna pattern table V _{k, l} (t). The propagation parameters are estimated with the same resolution as until convergence (step S22).

Thereafter, the angle high resolution processing unit 108 estimates the arrival direction with a resolution higher than the angle step of the antenna pattern table V _k (ψ _l , θ _l ) using the converged propagation parameter (step S23). As a result, a series of operations is completed.

Thus, in the second embodiment, the initial propagation parameter is set so that the known signal s (t) approaches the actually received signal (= baseband signals y ₁ (t) to y _K (t)). A value is detected, and the arrival direction of the incoming wave is estimated using the detected initial value of the propagation parameter.

  Therefore, according to the present invention, the direction of arrival can be estimated faster than in the first embodiment.

  Others are the same as in the first embodiment.

[Embodiment 3]
FIG. 10 is a schematic block diagram illustrating a configuration of the arrival direction estimation apparatus according to the third embodiment. The arrival direction estimation apparatus 300 according to Embodiment 3 replaces the propagation parameter estimation unit 106 of the arrival direction estimation apparatus 100 shown in FIG. 1 with the propagation parameter estimation unit 106A, and moves the angle high resolution processing unit 108 into the propagation parameter estimation unit 106A. The other parts are the same as those in the arrival direction estimation apparatus 100.

  The propagation parameter estimation unit 106 includes a propagation parameter estimation unit 106 and an angular high resolution processing unit 108.

In the third embodiment, the propagation parameter estimation unit 106 estimates the propagation parameter once by the method described above, and outputs the estimation result to the angle high resolution processing unit 108. Further, the angle high resolution processing unit 108 uses the propagation parameter received from the propagation parameter estimation unit 106 to determine the direction of arrival with a resolution higher than the angle step of the antenna pattern table V _k (ψ _l , θ _l ) by the method described above. presume. When the angular high resolution processing unit 108 receives the stop signal from the convergence determination unit 107, the angular high resolution processing unit 108 outputs the finally estimated arrival direction (= consisting of the azimuth angle θ and the elevation angle ψ) to the convergence determination unit 107.

  In the third embodiment, when the convergence determination unit 107 receives the arrival direction (consisting of the azimuth angle θ and the elevation angle ψ) from the angular high resolution processing unit 108, the convergence determination unit 107 receives the arrival direction (= azimuth angle θ and Is stored in the propagation parameter storage unit 109.

FIG. 11 is a flowchart for explaining the direction of arrival estimation operation in the third embodiment. When a series of operations is started, the reception unit 102 generates baseband signals y ₁ (t) to y _K (t) and outputs them to the replica signal creation unit 105. The known signal generation unit 103 s (t) is generated and output to the replica signal creation unit 105, and the propagation parameter initial value generation unit 104 outputs the held initial value of the propagation parameter to the replica signal creation unit 105.

  Then, the replica signal creation unit 105 and the propagation parameter estimation unit 106 set n = 1 (step S31), and subsequently set l = 1 (step S32).

Thereafter, the replica signal creation unit 105 and the propagation parameter estimation unit 106 estimate the propagation parameters [τ _ln , ψ _ln , θ _ln , α _ln ] by the above-described method using the initial values of the propagation parameters (step S33).

Subsequently, the angle high resolution processing unit 108 uses the estimated propagation parameters [τ _ln , ψ _ln , θ _ln , α _ln ] to be higher than the angle step of the antenna pattern table V _k (ψ _l , θ _l ). The direction of arrival is estimated by the method described above with the resolution (step S34).

  Thereafter, replica signal creation section 105 and propagation parameter estimation section 106 determine whether or not l = L (step S35). When it is determined in step S35 that l = L is not satisfied, the replica signal creation unit 105 and the propagation parameter estimation unit 106 set l = 1 + 1 (step S36). Then, the series of operations returns to step S33, and step S33 to step S36 described above are repeatedly executed until it is determined in step S35 that l = L.

  Thereafter, when it is determined in step S35 that l = L, the replica signal creation unit 105 and the propagation parameter estimation unit 106 further determine whether n = N (step S37), and n = N If not, n = n + 1 is set (step S38). Then, the series of operations returns to step S33, and the above-described steps S33 to S38 are repeatedly executed until it is determined in step S37 that n = N.

  Thereafter, when it is determined in step S37 that n = N, the series of operations ends.

As described above, in the third embodiment, every time the propagation parameter estimation unit 106 estimates the propagation parameter once with the same resolution as the angle step of the antenna pattern table V _k (ψ _l , θ _l ), the angle high resolution processing is performed. The unit 108 estimates the direction of arrival with a resolution higher than the angle step of the antenna pattern table V _k (ψ _l , θ _l ) using the estimated propagation parameter (see steps S33 and S34).

  In the third embodiment, the propagation parameter estimation unit 106 and the angle high resolution processing unit 108 are mounted as a single unit, so that a compact arrival direction estimation device can be realized.

  Note that the arrival direction estimation apparatus according to Embodiment 3 replaces the propagation parameter estimation unit 106 of the arrival direction estimation apparatus 200 shown in FIG. 6 with the propagation parameter estimation unit 106A, and replaces the angle high resolution processing unit 108 with the propagation parameter estimation unit 106A. An arrival direction estimation device that has moved to may be used. Thereby, the arrival direction estimation apparatus which can estimate an arrival direction compactly and at high speed is realizable.

  The rest is the same as in the first and second embodiments.

[Embodiment 4]
FIG. 12 is a schematic block diagram showing the configuration of the arrival direction estimation apparatus according to the fourth embodiment. The arrival direction estimation apparatus 400 according to the fourth embodiment includes a replica signal generation unit 105, a propagation parameter estimation unit 106, and an angular high resolution processing unit 108 of the arrival direction estimation apparatus 100 shown in FIG. The estimation unit 106B and the angle high resolution processing unit 108A are replaced with the other components, which are the same as those of the arrival direction estimation device 100.

Each of the replica signal creation unit 105A, the propagation parameter estimation unit 106B, and the angular high resolution processing unit 108A replaces the antenna pattern table V _k (ψ _l , θ _l ) described above with the antenna pattern table V ^* _k (ψ _l , θ _l ).

Then, the replica signal creation unit 105A creates the replica signal x _{k, l} (t) by the method described above using the antenna pattern table V ^* _k (ψ _l , θ _l ). Further, the propagation parameter estimation unit 106B calculates the evaluation functions Z (ψ) and Z (θ) by the method described later using the antenna pattern table V ^* _k (ψ _l , θ _l ), and the elevation angle ψ and azimuth angle θ. , And the complex amplitude attenuation α is estimated by the equation (8) using the antenna pattern table V ^* _k (ψ _l , θ _l ).

Further, the angle high resolution processing unit 108A calculates the evaluation functions Z (ψ) and Z (θ) by the method described later using the antenna pattern table V ^* _k (ψ _l , θ _l ), and the arrival direction (= elevation angle). ψ and azimuth angle θ) are estimated with a resolution higher than the angle step of the antenna pattern table V ^* _k (ψ _l , θ _l ).

FIG. 13 is a conceptual diagram of the antenna pattern table V _k (ψ _l , θ _l ) created in the original angle step. In FIG. 13, the vertical axis represents the antenna pattern value, and the horizontal axis represents the angle. A curve k5 indicates an I component antenna pattern value, and a curve k6 indicates a Q component antenna pattern value.

FIG. 14 is a conceptual diagram of an antenna pattern table V ^* _k (ψ _l , θ _l ) created with an angle step smaller than the original angle step. In FIG. 14, the vertical axis represents the antenna pattern value, and the horizontal axis represents the angle. A curve k7 indicates an I component antenna pattern value, and a curve k8 indicates a Q component antenna pattern value.

  The antenna pattern value of the I component indicated by the curve k5 is sampled at an interval smaller than the original interval, and the sampled I component sampling value is Fourier transformed to remove the high frequency component. Then, an I component antenna pattern value (= curve k7) is created by performing an inverse Fourier transform on the I component sampling value after the Fourier transform from which the high frequency component has been removed.

  Similarly, the antenna component value of the Q component indicated by the curve k6 is sampled at an interval smaller than the original interval, and the sampled Q component is Fourier-transformed to remove high frequency components. Then, the Q component sampling value (= curve k8) is created by performing inverse Fourier transform on the Fourier-transformed Q-component sampling value from which the high-frequency component has been removed.

Then, an antenna pattern table V ^* _k (ψ _l , θ _l ) composed of an I component antenna pattern value (= curve k7) and a Q component antenna pattern value (= curve k8) is used as a replica signal creation unit 105A and propagation parameter estimation. Set in each of the unit 106B and the angle high resolution processing unit 108A.

The propagation pattern estimation unit 106B creates an evaluation function by using any one of the following three methods using the antenna pattern table V ^* _k (ψ _l , θ _l ) described above, and estimates the propagation parameter.

MTH1) Evaluation functions Z (ψ) according to equations (4) and (6) in the first embodiment, respectively.
, Z (θ) is calculated to estimate the propagation parameter MTH2) The evaluation function Z _I (ψ), Z _{I is} calculated for each of the I component and the Q component using the equation (10).
(Θ): Calculate propagation parameters by calculating Z _Q (ψ) and Z _Q (θ) MTH3) Calculate evaluation parameters Z (ψ) and Z (θ) using equation (11) to determine propagation parameters.
Estimate

In addition, the angle high resolution processing unit 108A creates an evaluation function using any one of the following three methods using the antenna pattern table V ^* _k (ψ _l , θ _l ) described above, and determines the arrival direction. presume.

MTH4) The evaluation function Z (ψ) according to the equations (4) and (6) in the first embodiment, respectively.
, Z (θ) is calculated to estimate the direction of arrival. MTH5) The evaluation function Z _I (ψ), Z _{I is} calculated for each of the I component and the Q component using equation (10).
(Θ); Z _Q (ψ), Z _Q (θ) is calculated to estimate the direction of arrival MTH6) The evaluation functions Z (ψ) and Z (θ) are calculated using equation (11) to determine the direction of arrival. Estimation The arrival direction estimation apparatus according to the fourth embodiment includes each of the angle spectrum calculation unit 112, the replica signal creation unit 105, the propagation parameter estimation unit 106, and the angle high resolution processing unit 108 of the arrival direction estimation apparatus 200 shown in FIG. Holds the antenna pattern table V ^* _k (ψ _l , θ _l ), and the angle spectrum calculation unit 112 uses the antenna pattern table V ^* _k (ψ _l , θ _l ) to determine the arrival angle (azimuth angle θ and elevation angle ψ). made) and calculates the complex amplitude attenuation alpha, create replica signals _{x k,} l (t), the replica signal creation unit 105 using the antenna pattern table ^{_{V * k (ψ l, θ}} l), the propagation path Meter estimation unit 106 estimates the propagation parameters using the antenna pattern table ^{_{V * k (ψ l, θ}} l) a, angular high resolution processing section 108 with the antenna pattern table ^{_{V * k (ψ l, θ}} l) the An arrival direction estimation device that estimates an arrival direction may be used.

  In this case, the propagation parameter estimation unit 106 estimates the propagation parameter using any one of the three methods MTH1, MTH2, and MTH3 described above, and the angle high resolution processing unit 108 uses the three methods MTH4, MTH5 described above. , MTH6 is used to estimate the direction of arrival.

  Moreover, the arrival direction estimation apparatus according to the fourth embodiment may be one in which the third embodiment is applied to the arrival direction estimation apparatus 400 shown in FIG. That is, the direction-of-arrival estimation apparatus according to Embodiment 4 may be one in which the propagation parameter estimation unit 106B and the angle high resolution processing unit 108 are integrated.

As described above, according to the fourth embodiment, the direction of arrival is estimated using the antenna pattern table V ^* _k (ψ _l , θ _l ) created at an interval smaller than the original interval. The accuracy can be increased.

  Others are the same as those in the first to third embodiments.

[Application example]
FIG. 15 is a schematic block diagram illustrating a configuration of a wireless communication apparatus using the arrival direction estimation apparatus 100 illustrated in FIG. Radio communication apparatus 500 includes arrival direction estimation apparatus 100, transmission beam setting section 110, modulation section 120, and array transmission section 130.

The transmission beam setting unit 110 receives the arrival direction (azimuth angle θ _opt and elevation angle ψ _opt ) from the propagation parameter storage unit 109 of the arrival direction estimation apparatus 100, and receives the arrival direction (azimuth angle θ _opt and elevation angle ψ _opt ). A setting signal for setting the beam is generated and output to array transmitting section 130.

  Modulation section 120 receives transmission data from the outside, modulates the received transmission data by a predetermined method, and outputs the modulated transmission data to array transmission section 130.

Array transmission section 130 superimposes the modulation signal received from modulation section 120 on the setting signal for setting the beam in the arrival direction (azimuth angle θ _opt and elevation angle ψ _opt ) to antenna elements 101-1 to 101-K. Output.

Thereby, array antenna 101 sets a beam in the arrival direction (azimuth angle θ _opt and elevation angle ψ _opt ) and transmits transmission data.

  Radio communication apparatus 500 may include any of arrival direction estimation apparatuses 200, 300, and 400 instead of arrival direction estimation apparatus 100.

  In the present invention, transmission beam setting section 110, modulation section 120, and array transmission section 130 constitute "transmission means" that forms a beam in the arrival direction estimated by arrival direction estimation apparatus 100 and transmits a signal. To do.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is shown not by the above description of the embodiment but by the scope of claims for patent, and is intended to include all modifications within the meaning and scope equivalent to the scope of claims for patent.

  The present invention is applied to an arrival direction estimation device that can estimate the arrival direction of an incoming wave with higher accuracy than the accuracy of an antenna pattern table. The present invention is also applied to a radio communication apparatus including an arrival direction estimation device that can estimate the arrival direction of an incoming wave with higher accuracy than the accuracy of an antenna pattern table.

It is a schematic block diagram which shows the structure of the arrival direction estimation apparatus by Embodiment 1 of this invention. It is a conceptual diagram of the propagation parameter table which the replica signal preparation part shown in FIG. 1 hold | maintains. 4 is a flowchart for explaining an estimation operation in a propagation parameter estimation unit shown in FIG. 1. It is a figure for demonstrating the estimation method of the arrival direction in the angle high resolution process part shown in FIG. 5 is a flowchart for illustrating an arrival direction estimation operation in the first embodiment. 6 is a schematic block diagram illustrating a configuration of an arrival direction estimation apparatus according to Embodiment 2. FIG. It is a figure which shows the synthetic | combination waveform synthesize | combined by the correlation value calculation part shown in FIG. It is a figure which shows the angle spectrum calculated by the angle spectrum calculation part shown in FIG. 10 is a flowchart for explaining an arrival direction estimation operation in the second embodiment. FIG. 10 is a schematic block diagram illustrating a configuration of an arrival direction estimation device according to Embodiment 3. 10 is a flowchart for illustrating an arrival direction estimation operation in the third embodiment. FIG. 10 is a schematic block diagram illustrating a configuration of an arrival direction estimation device according to Embodiment 4. It is a conceptual diagram of the antenna pattern table created by the original angle step. It is a conceptual diagram of the antenna pattern table produced by the angle step smaller than the original angle step. It is a schematic block diagram which shows the structure of the radio | wireless communication apparatus using the arrival direction estimation apparatus shown in FIG.

Explanation of symbols

  100, 200, 300, 400 Arrival direction estimation device, 101 array antenna, 101-1 to 101-K antenna element, 102 receiving unit, 103 known signal generating unit, 104 propagation parameter initial value generating unit, 105, 105A replica signal creation 106, 106A, 106B propagation parameter estimation unit, 107 convergence determination unit, 108, 108A angle high resolution processing unit, 109 propagation parameter storage unit, 111 correlation value calculation unit, 112 angle spectrum calculation unit, 110 transmission beam setting unit, 120 modulation unit, 130 array transmission unit, 500 wireless communication device.

Claims (9)

An array antenna composed of a plurality of antenna elements and receiving incoming waves;
Receiving means for generating a received signal when the array antenna receives the incoming wave;
Propagation parameters when the incoming wave propagates using the received signal generated by the receiving means and an antenna pattern table that is a response vector of the plurality of antenna elements when the array antenna receives the incoming wave First estimation means for executing a first estimation process for estimating
Second estimation means for executing a second estimation process for estimating the arrival direction of the incoming wave with a second resolution higher than the first resolution using the propagation parameter estimated by the first estimation means. A direction-of-arrival estimation apparatus comprising: The first estimation means starts the first estimation process using an initial value of the propagation parameter, and repeatedly executes the first estimation process until the propagation parameter converges to estimate the propagation parameter. And
The arrival direction estimation apparatus according to claim 1, wherein the second estimation unit estimates the arrival direction of the arrival wave by executing the second estimation process using the converged propagation parameter. The first estimation means starts the first estimation process using an initial value of the propagation parameter, and repeatedly executes the first estimation process until the propagation parameter converges to estimate the propagation parameter. And
The second estimating means executes the second estimating process every time the first estimating process is executed, and sets the propagation parameter estimated by the first estimating means until the propagation parameter converges. The arrival direction estimation apparatus according to claim 1, wherein the arrival direction of the arrival wave is estimated using the arrival direction estimation apparatus.   The arrival direction estimation apparatus according to any one of claims 1 to 3, wherein the antenna pattern table is created with respect to the same angular step as the first resolution.   The direction-of-arrival estimation according to any one of claims 1 to 3, wherein the antenna pattern table includes a pattern table obtained by interpolating an antenna pattern table created for the same angular step as the first resolution. apparatus.   The arrival direction estimation apparatus according to any one of claims 1 to 5, wherein the initial value of the propagation parameter includes a parameter arbitrarily set in advance.   The arrival direction estimation apparatus according to any one of claims 1 to 5, wherein the initial value of the propagation parameter includes a propagation parameter detected so that the transmission signal approaches the reception signal.   The arrival direction estimation apparatus according to claim 7, wherein the initial value of the propagation parameter includes a propagation parameter detected based on a correlation value between the reception signal and the transmission signal. An arrival direction estimation device according to any one of claims 1 to 8,
A wireless communication apparatus comprising: a transmission unit configured to transmit a signal by forming a beam in the arrival direction estimated by the arrival direction estimation apparatus.

Images