JP2009218936A - Reception device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To precisely estimate a weight in adaptive array reception. <P>SOLUTION: An antenna 10 receives a signal including a known part and a data part alternately. A weight estimation unit 20 performs an adaptive algorithm for a known part obtained by an acquisition unit to estimate a weight vector. An array composition unit 50 performs array composition processing on the received known part with the weight vector estimated by the weight estimation unit 20. A weight correction unit 40 corrects the weight vector used for the known part by a weight multiplication unit 60 based upon an array composite signal from the array composition unit 50 to derive a weight vector by which the data part is to be multiplied. The array composition unit 50 performs array composition processing on the received data part with the weight vector corrected by the weight correction unit 40. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to reception technology, and more particularly, to a reception apparatus that performs adaptive array reception processing.

  2. Description of the Related Art In recent years, for the purpose of improving communication characteristics and the like, a technique related to adaptive array (Adaptive Array) reception has been developed in which reception signals received by a plurality of antennas are multiplied by weights and combined. In this adaptive array reception technique, a weight for multiplying a received signal is estimated using an adaptive algorithm such as an RLS (Recursive Least Square) algorithm or an LMS (Least Mean Square) algorithm, and a desired received signal is demodulated.

However, in such an adaptive algorithm, parameters such as the initial value of the weight and the update step are important, which affects the convergence speed and convergence accuracy. Conventionally, a plurality of parameters are properly used according to the propagation path situation (for example, see Patent Document 1). In addition, a plurality of parameters are set simultaneously for each of a plurality of adaptive algorithms and executed in parallel (see, for example, Patent Document 2). Further, the weight calculated in the immediately preceding received frame is used as the initial value of the current weight (see, for example, Patent Document 3).
JP 2002-026788 A JP 2003-264489 A JP 2002-077644 A

  In general, in an adaptive algorithm, if the known signal is lengthened, the estimation accuracy is improved. On the other hand, in wireless communication, if the known signal is lengthened, the transmission efficiency is reduced. Therefore, a technique for improving the estimation accuracy without changing the length of the known signal has been desired.

  The present invention has been made in view of such circumstances, and an object thereof is to provide a technique capable of accurately estimating the weight in adaptive array reception.

  In order to solve the above-described problem, a receiving apparatus according to an aspect of the present invention includes a receiving unit that receives a signal in which a data portion is arranged after a known portion by a plurality of antennas, and a signal received by the receiving unit. A weight estimation unit that executes an adaptive algorithm on the known part and estimates an initial weight vector, and a known part of the signal received by the reception unit by the initial weight vector estimated by the weight estimation unit Calculated by an array combining unit, a weight correcting unit for correcting an initial weight vector based on the output from the first array combining unit, and a weight correcting unit. A second array combining unit that array-synthesizes the data portion of the signal received by the receiving unit using a weight vector. That.

  Another embodiment of the present invention is also a receiving device. This apparatus receives a signal in which a data portion is arranged after a known portion by a plurality of antennas, and executes an RLS algorithm on the known portion of the signals received by the receiving portion, A first weight estimator for estimating a vector, a first array synthesizer for array-synthesizing a known portion of the signal received by the receiver by the weight vector estimated by the first weight estimator, and a first array Based on the output from the synthesizer, the initial value of the correlation matrix P in the RLS algorithm is corrected, and the RLS algorithm is executed again using the corrected initial value of the correlation matrix P to multiply the data portion. A second weight estimation unit for estimating a weight vector to be performed, and a weight vector estimated by the second weight estimation unit. And a second array combining section for array synthesis data portion of the signal received by the receiving unit.

  It should be noted that any combination of the above-described constituent elements and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, etc. are also effective as an aspect of the present invention.

  According to the present invention, it is possible to accurately estimate the weight in adaptive array reception.

  Before describing embodiments of the present invention specifically, an outline will be described first. Embodiments described herein relate generally to adaptive array reception technology. Adaptive array reception is a technique for improving reception performance by receiving signals using a plurality of antennas and combining the arrays with weights obtained using an adaptive algorithm.

  In general, in an adaptive algorithm, if the known signal is lengthened, the estimation accuracy is improved. On the other hand, in wireless communication, if the known signal is lengthened, the transmission efficiency is reduced. Therefore, a technique for improving the estimation accuracy without changing the length of the known signal has been desired.

  The adaptive algorithm uses parameters such as an update coefficient and an update step in addition to the initial weight value. Generally, when the update coefficient and the update step are set large, the weight convergence speed may be increased. However, if it is set too large, it may diverge or vibrate, resulting in a slow convergence. On the other hand, if it is set small, the convergence accuracy may be improved. Therefore, it is important to appropriately set the update coefficient and the update step.

  Here, when examining the setting of the initial value using simulation, the smaller the update coefficient and the update step, the smaller the error of the phase component of the weight, but the greater the error of the amplitude component. found. Therefore, in the embodiment of the present invention, in order to reduce the error of the amplitude component, an array synthesis error is derived using a known signal, and the weight is corrected using the error. Thereby, the weight estimation accuracy can be improved without changing the length of the known signal. Details will be described later. In the present embodiment, it is assumed that the RLS algorithm is used as the adaptive algorithm.

  FIG. 1 is a diagram illustrating a configuration example of a receiving device 100 according to an embodiment of the present invention. The receiving apparatus 100 includes a first antenna 10a to a fourth antenna 10d represented by an antenna 10, a reception RF unit 12, a weight estimation unit 20, an initial value setting unit 30, a weight correction unit 40, and an array synthesis unit. 50 and a demodulation processing unit 80. The array combining unit 50 includes a first weight multiplying unit 60 a to a fourth weight multiplying unit 60 d typified by a weight multiplying unit 60, and an adding unit 70.

  The antenna 10 receives a signal from a transmission device (not shown) and outputs the signal to the reception RF unit 12. Note that the number of antennas 10 may not be four. FIG. 2 is a diagram illustrating an example of a format 400 of a signal received by the antenna 10 of FIG. The format 400 includes a known signal section 410 and a data signal section 420. In the known signal section 410, a known symbol sequence is arranged between the transmission device and the reception device 100. In the data signal section 420, a transmission signal from the transmission device is arranged. The antenna 10 may repeatedly receive a signal configured in the format 400.

  The reception RF unit 12 executes processing for converting the signal transmitted from the antenna 10 from the RF band to the baseband, and outputs the signal to the weight estimation unit 20 and the array synthesis unit 50. The initial value setting unit 30 notifies the weight estimation unit 20 of the initial value P (0) of the correlation matrix P for executing the RLS algorithm and the initial value W (0) of the weight, and stores them. (Hereinafter, P (0) and W (0) are collectively referred to as “initial value”). When the signal output from the reception RF unit 12 is a known signal as illustrated in FIG. 2, the weight estimation unit 20 uses P (0) and W (0) notified from the initial value setting unit 30. Then, the RLS algorithm is executed to derive a weight vector for the known signal (hereinafter referred to as a known signal weight vector).

  Here, the RLS algorithm will be described. The RLS algorithm estimates the known signal weight vector by repeating the following (1) to (4) at t = 1 to n. Note that n indicates the length of the known signal section 410 in FIG.

(1) A Kalman gain vector K (t) is calculated based on the following equation. X (t) represents a known signal output from the reception RF unit 12. X ^H (t) represents a conjugate transpose matrix of X (t).

（３）下式に基づき、ウエイトベクトルＷ（ｔ）を更新する。

（４）下式に基づき、相関行列Ｐ（ｔ）を更新する。λは、忘却係数を示す。

(2) The error e (t) is calculated based on the following equation. d (t) represents a sequence of known signals stored in advance.

(3) The weight vector W (t) is updated based on the following expression.

(4) The correlation matrix P (t) is updated based on the following equation. λ represents a forgetting factor.

  When the data signal as shown in FIG. 2 is output from the reception RF unit 12 after the known signal weight vector is derived, the weight estimation unit 20 uses the known signal weight vector derived in the known signal portion. It is set as a weight vector for the data signal (hereinafter referred to as a data weight vector). The weight estimation unit 20 notifies the weight correction unit 40 of the data weight vector in consideration of the timing at which the data signal is output from the reception RF unit 12.

  When the weight vector notified from the weight estimation unit 20 is a known signal weight vector, the weight correction unit 40 outputs the weight vector to the array synthesis unit 50 without performing the correction process. When the weight vector notified from the weight estimation unit 20 is a data weight vector, the weight correction unit 40 uses the array synthesis signal related to the known signal output from the array synthesis unit 50 to generate the data weight vector. After performing the correction processing, the data is output to the weight multiplication unit 60.

  That is, the weight correction unit 40 corrects the data weight vector before the data signal is output from the reception RF unit 12, and when the data signal is input to the array synthesis unit 50, the weight correction unit 40 The corrected data weight vector is output. Here, the same data weight vector is output during the data signal section 420 shown in FIG. The data weight vector may be updated by tracking using the LMS algorithm in the data section instead of the same data weight vector.

  In the array synthesis unit 50, the weight multiplication unit 60 multiplies the plurality of signals output from the reception RF unit 16 by the weight vectors output from the weight correction unit 40, respectively. Specifically, the weight multiplication unit 60 multiplies the known signal by a known signal weight vector, and multiplies the data signal by a data signal weight vector. The adder 70 combines the outputs from the respective weight multipliers 60 and outputs an array combined signal to the demodulation processor 80 and the weight corrector 40. The demodulation processing unit 80 performs demodulation processing on the array composite signal output from the addition unit 70.

  Here, a detailed configuration of the weight correction unit 40 will be described. FIG. 3 is a diagram illustrating a configuration example of the weight correction unit 40 of FIG. The weight correction unit 40 includes an averaging unit 42, a correction coefficient derivation unit 44, a memory 46, and a weight correction multiplication unit 48. As described above, when the known signal weight vector is output from the weight estimation unit 20, the weight correction unit 40 outputs it to the weight multiplication unit 60 as it is. In the following, a case where the data weight vector is output from the weight estimation unit 20 will be described.

  The averaging unit 42 performs an amplitude averaging process on the array combined signal corresponding to the known signal among the array combined signals output from the adding unit 70. The correction coefficient deriving unit 44 derives a correction coefficient using the average value of the array composite signal related to the known signal output from the averaging unit 42 and the known signal stored in the memory 46. The memory 46 stores a symbol pattern of a known signal arranged in the known signal section 410 of FIG.

  Specifically, the correction coefficient deriving unit 44 first derives the ideal average amplitude of the known signal stored in the memory 46 (hereinafter referred to as the ideal average amplitude). Next, the correction coefficient deriving unit 44 performs a division process using the ideal average amplitude as the dividend and the average value of the combined signal as the divisor, and sets the result as the correction coefficient.

  Note that the memory 46 may hold in advance the ideal average amplitude instead of the symbol pattern of the known signal. Thereby, the processing load of the correction coefficient deriving unit 44 is reduced. Further, the symbol pattern of the known signal may be defined so that the ideal average amplitude is 1. As a result, the reciprocal of the average value of the amplitude of the array composite signal can be derived as a correction coefficient, so that the memory 46 is not necessary and the processing in the correction coefficient deriving unit 44 is simplified. Note that a statistical value such as a median value may be used instead of the average value.

  The weight correction multiplication unit 48 performs a multiplication process using the data weight vector output from the weight estimation unit 20 as a multiplicand and the correction coefficient output from the correction coefficient derivation unit 44 as a multiplier, and the result is a weight multiplication unit. 60.

  These configurations described above can be realized in terms of hardware by a CPU, memory, or other LSI of an arbitrary computer, and in terms of software by a program loaded in the memory. Describes functional blocks realized through collaboration. Accordingly, those skilled in the art will understand that these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof.

  An operation example in the above configuration will be described. FIG. 4 is a flowchart showing an operation example of the receiving apparatus 100 of FIG. The weight estimation unit 20 executes an RLS algorithm using a known signal portion of the received signal, and estimates a known signal weight vector. (S10). The array synthesizing unit 50 performs array synthesis of the known signals among the received signals using the estimated known signal weight vector (S12).

  The weight correction unit 40 derives an average value of the amplitudes of the array synthesized signals (S14). Next, the weight correction unit 40 obtains a ratio between the ideal average amplitude of the known signal stored in advance and the average value derived in S14 and derives it as a correction coefficient. Further, the weight correction unit 40 corrects the known signal weight vector derived in S12 by multiplying it by a correction coefficient, and outputs it as a data weight vector to the array combining unit 50 (S16). The array combining unit 50 performs array combining processing on the data signal portion of the received signal using the data weight vector (S18).

  Here, when the array combining process of the data signal portion is not completed (N in S20), the process returns to S18. When the weight calculation process for the data signal portion is completed (Y in S20) and there is an unprocessed signal (N in S22), the process returns to S10. On the other hand, when the weight calculation processing for the data signal portion is completed (Y in S20) and the processing for all signals is completed (Y in S22), the processing is terminated.

  As described above, in order to reduce the amplitude component error, it is possible to accurately estimate the weight by deriving an array demodulation error using a known signal and correcting the weight using the error. Further, it is possible to estimate at a higher speed by correcting the weight vector used for the known signal based on the array combined output of the known signal and deriving the data signal weight vector to be multiplied by the data portion. For the data signal to be multiplied by the data part by deriving the ratio of the average value of the array composite output for the known signal and the average value of the ideal amplitude of the known signal, and multiplying the weight vector by the derived ratio The weight vector can be derived efficiently. In addition, the weight vector estimation accuracy can be improved with a simple configuration.

  The present invention has been described based on the embodiments. This embodiment is an exemplification, and various modifications can be made to the combination of each component and each processing process, and such a modification, a combination of the embodiment and the modification, or between the modifications. Those skilled in the art will appreciate that combinations are also within the scope of the present invention.

  Next, a modified example will be described. The modified example relates to an adaptive array reception technique as in the above-described embodiment. In the above-described embodiment, the weight for the data signal portion is corrected using the average value of the array composite signal for the known signal portion, whereas in the modification, the initial value is corrected instead of the weight. And the point that the adaptive algorithm is executed again after correcting the initial value. In addition, about the structure similar to the above-mentioned embodiment, the same code | symbol is attached | subjected and description is simplified. In this modification, it is assumed that the RLS algorithm is used as the adaptive algorithm.

  FIG. 5 is a diagram illustrating a configuration example of a receiving device 200 according to a modification of the embodiment of the present invention. The receiving apparatus 200 includes a first antenna 10a to a fourth antenna 10d typified by an antenna 10, a reception RF unit 12, a weight estimation unit 20, an initial value setting unit 30, an array combining unit 50, and an initial value correction. Unit 90 and a demodulation processing unit 80. The array combining unit 50 includes a first weight multiplying unit 60 a to a fourth weight multiplying unit 60 d typified by a weight multiplying unit 60, and an adding unit 70. Compared with FIG. 1, the receiving device 200 of FIG. 5 includes an initial value correction unit 90 instead of the weight correction unit 40 of FIG. 1.

In a variant, the adaptive algorithm is executed twice. In the following, processing when each adaptive algorithm is executed will be described.
(1) Execution of First Adaptive Algorithm The initial value setting unit 30 notifies the initial value correcting unit 90 of P (0) and W (0). The initial value correction unit 90 outputs P (0) and W (0) output from the initial value setting unit 30 to the weight estimation unit 20 as they are when the first adaptive algorithm is executed.

  The weight estimation unit 20 performs adaptive signal processing on the known signal output from the reception RF unit 12 using P (0) and W (0) output from the initial value setting unit 30, Estimate the known credit weight. Further, the weight estimation unit 20 stores a known signal output from the reception RF unit 12 (hereinafter also referred to as “reception known signal”) for the second adaptation algorithm. The estimated known signal weight vector is output to the array synthesizing unit 50, and is used for array synthesizing the known signal output from the reception RF unit 12.

(2) Execution of Second Adaptive Algorithm Based on the array composite signal related to the known signal output from the adder 70, the initial value corrector 90 outputs P (0) output from the initial value setter 30. A correction coefficient for correcting the above is derived. Further, the initial value correction unit 90 corrects P (0) with the derived correction coefficient and outputs the corrected value to the weight estimation unit 20. The weight estimation unit 20 executes the second adaptive algorithm using the corrected P (0) and the received known signal, and estimates the weight vector. For W (0), a known signal weight vector estimated in the first adaptation algorithm may be set, or the same value as W (0) used in the first adaptation algorithm may be set. May be.

  The weight estimation unit 20 outputs the estimated weight vector to the array synthesis unit 50 as a data weight vector for the data signal section. The array combining unit 50 performs an array combining process on the data signal output from the reception RF unit 12 by using the data weight vector output from the weight multiplying unit 60.

  Here, the configuration of the initial value correction unit 90 will be described in detail. FIG. 6 is a diagram illustrating a configuration example of the initial value correction unit 90 of FIG. The initial value correction unit 90 includes an averaging unit 92, a correction coefficient derivation unit 94, a memory 96, and an initial value correction multiplication unit 98. When executing the first adaptive algorithm, the initial value correction unit 90 outputs P (0) and W (0) output from the initial value setting unit 30 to the weight estimation unit 20 as they are.

  At the time of executing the second adaptive algorithm, the averaging unit 92 derives the average value of the amplitude of the array composite signal related to the known signal portion output from the adding unit 70 and outputs the average value to the correction coefficient deriving unit 94. . Next, the correction coefficient deriving unit 94 executes a division process using the ideal average amplitude of the known signal stored in the memory 46 as a dividend and the average value of the array composite signal as a divisor, and the result is used as a correction coefficient to obtain an initial value. The result is output to the correction multiplier 98. The initial value correction multiplication unit 98 multiplies P (0) output from the initial value setting unit 30 by the correction coefficient output from the correction coefficient derivation unit 94 and outputs the result to the weight estimation unit 20.

  An operation example in the above configuration will be described. FIG. 7 is a flowchart showing an operation example of the receiving apparatus 200 of FIG. This flowchart may be started when the reception RF unit 12 outputs a known signal.

  The weight estimation unit 20 estimates the known signal weight vector using the known signal portion output from the reception RF unit 12 (S30). Next, the array combining unit 50 performs array combining of the known signals output from the reception RF unit 12 using the estimated known signal weight vector, and outputs an array combined signal (S32). The initial value correction unit 90 derives the average value of the array composite signal (S34).

  The initial value correction unit 90 derives a correction coefficient based on the ratio between the average value of the array composite signal and the ideal average amplitude of the known signal stored in advance. Further, the initial value P (0) output from the initial value setting unit 30 is corrected with the derived correction coefficient, and output to the weight estimation unit 20 (S36). Here, the weight estimation unit 20 executes the adaptive algorithm again using the initial value output from the initial value correction unit 90 and the known signal output from the reception RF unit 12, and performs the data weight. A vector is estimated (S38).

  Next, the array composition unit 50 performs array composition on the data signal output from the reception RF unit 12 using the data weight vector output from the weight estimation unit 20 (S40). When the weight calculation processing for all data signal portions is not completed (N in S42), the process returns to S40. When the weight calculation process for the data signal portion is completed (Y in S42) and there is an unprocessed signal (N in S44), the process returns to S30. On the other hand, when the weight calculation processing for the data signal portion is completed (Y in S42) and the processing for all signals is completed (Y in S44), the processing is terminated.

  According to the above aspect, the data signal weight vector to be multiplied by the data portion is estimated by correcting the initial value used for estimation of the known signal weight vector based on the array combined output of the known signal. Thus, the weight can be estimated with higher accuracy. In addition, the data signal to be multiplied by the data portion is derived by deriving the ratio between the average value of the array composite output for the received known signal and the average value of the ideal amplitude of the known signal and multiplying the derived ratio by the initial value. The initial value for estimating the weight vector can be efficiently derived. In addition, this makes it possible to improve the weight estimation accuracy with a simple configuration.

  Next, another modified example (hereinafter referred to as a second modified example) will be described. The second modified example relates to an adaptive array reception technique as in the above-described embodiment and modified examples. Specifically, the second modification is a process that combines the weight correction process in the embodiment and the initial value correction process in the modification. In the second modification, it is assumed that the RLS algorithm is used as the adaptive algorithm.

  Although details will be described later, in the second modification, first, the second adaptation algorithm is executed after correcting the initial value as in the above-described modification. Further, as in the above-described embodiment, the array synthesized signal of the known signal portion is derived again with the weight derived in the second adaptation algorithm, and the data weight vector for the data signal portion is corrected using the average value. To do. In addition, about the structure similar to the above-mentioned embodiment, the same code | symbol is attached | subjected and description is simplified.

  FIG. 8 is a diagram illustrating a configuration example of the receiving device 300 according to the second modification of the embodiment of the present invention. The receiving apparatus 300 includes a first antenna 10a to a fourth antenna 10d typified by an antenna 10, a reception RF unit 12, a weight estimation unit 20, an initial value setting unit 30, a weight correction unit 40, and an array synthesis unit. 50, an initial value correction unit 90, and a demodulation processing unit 80. The array combining unit 50 includes a first weight multiplying unit 60 a to a fourth weight multiplying unit 60 d typified by a weight multiplying unit 60, and an adding unit 70. The receiving apparatus 200 in FIG. 8 is configured to further include an initial value correction unit 90, as compared with FIG. Compared with FIG. 5, the weight correction unit 40 is further included.

(1) The initial value setting unit 30 outputs P (0) and W (0) as initial values to the initial value correction unit 90. The initial value correction unit 90 notifies the weight correction unit 40 as it is without correcting the initial value notified from the initial value setting unit 30 when executing the first adaptive algorithm. Hereinafter, this initial value is referred to as a known signal initial value.

(2) The weight estimation unit 20 acquires the known signal initial value from the initial value correction unit 90 and acquires the known signal from the reception RF unit 12. The acquired known signal is stored as a received known signal. Further, the weight estimation unit 20 estimates the known signal weight vector using the received known signal and the known signal initial value, and notifies the weight correction unit 40 of the estimated signal weight vector. The weight correction unit 40 outputs the known signal weight vector notified from the weight estimation unit 20 to the weight multiplication unit 60 as it is. The weight multiplication unit 60 multiplies the known signal output from the reception RF unit 12 and the known signal weight vector output from the weight correction unit 40, and outputs the result to the addition unit 70. The adder 70 combines the signals for each antenna and outputs an array combined signal related to the known signals.

(3) The initial value correcting unit 90 corrects the known signal initial value using the array composite signal related to the known signal output from the adding unit 70. The weight estimation unit 20 estimates the known signal weight vector again using the corrected known signal initial value and the stored received known signal, and notifies the weight correction unit 40 of the estimation. The weight correction unit 40 outputs the known signal weight vector notified from the weight estimation unit 20 to the weight multiplication unit 60 as it is.

(4) The weight multiplication unit 60 multiplies the known signal output from the reception RF unit 12 and the known signal weight vector output from the weight correction unit 40, and outputs the result to the addition unit 70. The adder 70 synthesizes the signals for each antenna and outputs an array composite signal related to the known signal to the weight correction unit 40. Here, the weight correction unit 40 performs correction processing on the known signal weight vector using the array composite signal related to the known signal output from the addition unit 70, and the weight multiplication unit 60 serves as the data weight vector. Output to.

(5) The weight multiplication unit 60 multiplies the data signal output from the reception RF unit 12 and the data weight vector output from the weight correction unit 40 and outputs the result to the addition unit 70. Adder 70 synthesizes the signals for each antenna and outputs them to demodulation processor 80.

  An operation example in the above configuration will be described. In the operation in the second modified example, the operation corresponding to the embodiment is inserted during the operation in the modified example described above. Therefore, the description of the operation corresponding to the above-described modification is omitted. FIG. 9 is a flowchart showing an operation example of the receiving apparatus 300 of FIG. This flowchart shows a processing procedure to be inserted between S38 and S40 in the flowchart of FIG.

  After the processing of S38, the weight multiplication unit 60 performs weight multiplication on the known signal output from the reception RF unit 12 using the known signal weight vector output from the weight correction unit 40, and the addition unit 70 Then, the result of the weight multiplication is synthesized and an array synthesized signal related to the known signal is output (S50). The weight correction unit 40 derives the average value of the array composite signal related to the known signal, and derives the ratio of the known signal with the ideal average amplitude as a correction coefficient (S52). Further, the weight correction unit 40 multiplies the correction coefficient by the known signal weight vector, and outputs it as a data weight vector to the weight multiplication unit 60 (S54), and proceeds to the processing after S40.

  According to the above aspect, prior to multiplication of the data signal and the data weight vector, the known signal output from the reception RF unit 12 is multiplied by the data weight vector. Furthermore, a weight with higher accuracy can be derived by correcting the data weight vector to be multiplied by the data signal according to the result obtained by the multiplication.

  In the above-described embodiment, the RLS algorithm is described as being used. However, the LMS algorithm may be used without being limited thereto. Even with the LMS algorithm, accurate weights can be derived and reception performance can be improved by correcting the weights using the error of the array composite signal related to the known signals.

It is a figure which shows the structural example of the receiver concerning embodiment of this invention. It is a figure which shows the example of the format of the signal which the antenna of FIG. 1 receives. It is a figure which shows the structural example of the weight correction | amendment part of FIG. 3 is a flowchart illustrating an operation example of the receiving apparatus in FIG. 1. It is a figure which shows the structural example of the receiver concerning the modification of embodiment of this invention. FIG. 6 is a diagram illustrating a configuration example of an initial value correction unit in FIG. 5. 6 is a flowchart illustrating an operation example of the receiving apparatus in FIG. 5. It is a figure which shows the structural example of the receiver concerning the 2nd modification of embodiment of this invention. FIG. 9 is a flowchart illustrating an operation example of the receiving apparatus of FIG. 8. FIG.

Explanation of symbols

  10 antennas, 12 reception RF units, 20 weight estimation units, 30 initial value setting units, 40 weight correction units, 42 averaging units, 44 correction coefficient derivation units, 46 memories, 48 weight correction multiplication units, 50 array synthesis units, 60 Weight multiplication unit, 70 addition unit, 80 demodulation processing unit, 90 initial value correction unit, 92 averaging unit, 94 correction coefficient derivation unit, 96 memory, 98 initial value correction multiplication unit, 100 receiver, 400 format, 410 known signal Section, 420 Data signal section.

Claims (4)

A receiving unit for receiving a signal in which a data part is arranged at a subsequent stage of a known part with a plurality of antennas;
A first weight estimation unit that executes a RLS (Recursive Least Square) algorithm on a known portion of the signal received by the reception unit and estimates a weight vector;
A first array synthesizing unit that array synthesizes a known portion of the signal received by the receiving unit with the weight vector estimated by the first weight estimating unit;
Based on the output from the first array synthesis unit, the initial value of the correlation matrix P in the RLS algorithm is corrected, and the RLS algorithm is executed again using the corrected initial value of the correlation matrix P. A second weight estimation unit for estimating a weight vector to be multiplied by the data portion;
A second array combining unit that combines the data portion of the signal received by the receiving unit with the weight vector estimated by the second weight estimating unit;
A receiving apparatus comprising:   The second weight estimation unit calculates the ratio of the amplitude of the output of the first array synthesis unit to the amplitude of the known part stored in advance, and multiplies the initial value of the correlation matrix P by the reciprocal of the calculated ratio. The receiving apparatus according to claim 1, wherein an initial value of the correlation matrix P is corrected. Prior to array synthesis of the data portion by the second array synthesis unit, a third array for synthesizing a known portion of the signal received by the reception unit with the weight vector estimated by the second weight estimation unit A synthesis unit;
A weight correction unit that corrects the weight vector estimated by the second weight estimation unit based on the output from the third array synthesis unit;
Further comprising
The said 2nd array synthetic | combination part carries out the array synthesis | combination of the data part of the signal received by the said receiving part using the weight vector correct | amended by the said weight correction | amendment part, It is characterized by the above-mentioned. The receiving device described.   The weight correction unit calculates a ratio of the amplitude of the output of the third array synthesis unit to the amplitude of a known part stored in advance, and is calculated for the weight vector estimated by the second weight estimation unit. 4. The receiving apparatus according to claim 3, wherein the weight vector is corrected by multiplying by the inverse of the ratio.

Images