JP2002261531A - Device and method for conrolling array antenna - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an array antenna controller that has a simple constitution and can be manufactured inexpensively, as compared with the conventional controller and performs temporal and spatial adaptive processing for an ESPAR antenna. SOLUTION: A time-domain signal processing section 4 divides a signal y(t) received by mean of an array-antenna device 100, composed of the EXPAR antenna into sub-signals in a plurality of time domains, processes the sub-signals in the time domains by respectively multiplying the sub-signals by prescribed weighting factors and adding the weighted sub-signals to each other, and outputs the added signals as processed signal z(t). An adaptable controller 7 computes the weighting factors, based on a learning sequence signal generated from a learning sequence signal generator 6 and the sub-signals, so that the error signals between the sequence signal and processed signals z(t) become minimal, outputs the weighting factors to the time-domain signal processing section 4, and calculates and sets the reactance value of each variable reactance element, so that the value of a prescribed criterion function, indicating the value corresponding to one of the error signals, becomes minimal by calculating the gradient vector of the criterion function.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an array antenna control device and a control method capable of changing the directional characteristics of an array antenna device comprising a plurality of antenna elements, and more particularly, to an adaptive control method for changing the directional characteristics. Electronically controlled director array antenna device (Electronically
Steerable Passive Array Radiator (ESPAR) Antenn
a; hereinafter, referred to as ESPAR antenna. ) And TDM
The present invention relates to a control device and a control method for an array antenna capable of processing an A reception signal or a CDMA reception signal.

[0002]

2. Description of the Related Art An ESPAR antenna is disclosed, for example, in T. Ohira, "Microwave signal processing and
vices for adaptive beamforming, "IEEE Antenna and
Propagation society International Symposium vol. T
wo, pp. 583-586, Salt LakeCity, Utah July 16-21, 2
000 "and Japanese Patent Application No. 11-194487. The ESPAR antenna is provided with an excitation element through which a radio signal is transmitted and received, at least one non-excitation element that is provided at a predetermined distance from the excitation element, and through which no radio signal is transmitted and received, and is connected to the non-excitation element. An array antenna including a variable reactance element is provided, and the directivity characteristics of the array antenna can be changed by changing the reactance value of the variable reactance element.

[0003] Wireless communication includes multipath propagation,
Co-channel interference (CCI) exists as two problems that adversely affect wireless systems. Each of these problems manifests as inter-symbol interference (ISI) and co-channel interference due to frequency reuse in TDMA wireless systems, or multi-user access interference (MAI) in CDMA wireless systems.

In order to solve the above problem, a spatio-temporal adaptive processing (STAP) (refer to J. Paulraj e
t al., "Space-time processing for wireless communi
cations, "IEEE Signal Processing Magazine, Vol. 1
4, No. 6, pp. 49-83, November 1997. " ) Has been proposed, and it is believed that this process exhibits outstanding performance in suppressing both ISI and CCI. More recently, TDMA or direct sequence (sequence) CDMA
For DS-CDMA wireless communication systems, a method of space-time adaptive processing (STAP) has been proposed and analyzed.

[0005]

However, the STA
Due to the complexity and high cost of realizing the antenna array channels of the P system, these may be practically widespread, especially in situations where cost is a very important factor, such as in wireless local area network systems or user terminals. It is difficult to apply. This means that developing a simpler and lower cost STAP system is of practical importance.

SUMMARY OF THE INVENTION An object of the present invention is to solve the above-mentioned problems, to have a simpler structure than in the prior art, to have a low manufacturing cost, and to be able to perform spatio-temporal adaptive processing for an ESPAR antenna. An object of the present invention is to provide a control device and a control method for an array antenna.

[0007]

According to the present invention, there is provided an array antenna control apparatus comprising: a radiating element for receiving a radio signal; and a plurality of parasitic elements provided at a predetermined distance from the radiating element. A plurality of variable reactance elements respectively connected to the plurality of parasitic elements, and by changing the reactance value of each of the variable reactance elements, the plurality of variable reactance elements as a director or a reflector, respectively. Operate, in the array antenna control device to change the directional characteristics of the array antenna, the radio signal received by the array antenna is divided into a plurality of time-domain sub-signals, for each of the divided sub-signals Performs signal processing in the time domain by multiplying by a predetermined weighting factor and then adding it to output as a processed signal Time domain signal processing means, based on a predetermined learning sequence signal and each of the sub-signals, calculates the weighting coefficient such that an error signal between the processing signal and the learning sequence signal is minimized, and Output to the area signal processing means, calculate a gradient vector of a predetermined reference function indicating a value corresponding to the error signal, and calculate a reactance value of each of the variable reactance elements so that the value of the reference function is minimized. And an adaptive control means for setting.

Further, the method for controlling an array antenna according to the present invention includes a radiating element for receiving a radio signal, a plurality of non-exciting elements provided at a predetermined distance from the radiating element, A plurality of variable reactance elements respectively connected to the non-excitation elements, and by changing the reactance value of each of the variable reactance elements, the plurality of variable reactance elements operate as a director or a reflector, respectively, and In a method of controlling an array antenna that changes the directional characteristics of an antenna, a radio signal received by the array antenna is divided into a plurality of time-domain sub-signals, and a predetermined weighting factor is assigned to each of the divided sub-signals. Multiplied by the sum and then added to execute signal processing in the time domain and output as a processed signal. And, based on a predetermined learning sequence signal and each of the sub-signals, calculate the weighting coefficient so that an error signal between the processing signal and the learning sequence signal is minimized, and perform signal processing in the time domain. Calculating a gradient vector of a predetermined reference function indicating a value corresponding to the error signal, and calculating and setting a reactance value of each of the variable reactance elements so that the value of the reference function is minimized. And characterized in that:

[0009]

DETAILED DESCRIPTION OF THE INVENTION Unlike nulls formed by conventional adaptive algorithms or beams formed in advance before signal processing, the present invention provides a variable beam pattern and time domain. And a control method of an array antenna capable of realizing spatio-temporal adaptive filtering using the equalizer of (1). Espar antennas are believed to have the ability to spatially beam-form a desired signal with minimal cost. The present invention proposes a control apparatus and a control method of an array antenna for realizing space-time adaptive filtering (STAF) using an ESPAR antenna for a TDMA or CDMA signal waveform.

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of a control device for an array antenna according to an embodiment of the present invention. As shown in FIG. 1, the control device for an array antenna according to the present embodiment is an array antenna device configured by a conventional ESPAR antenna including one excitation element A0 and six non-excitation elements A1 to A6. 100, a time-domain signal processing unit 4 for processing a radio signal received by the array antenna device 100,
And an adaptive control type controller 7 for controlling them.

In FIG. 1, an array antenna device 100
Is composed of an excitation element A0 and non-excitation elements A1 to A6 provided on the ground conductor 11, and the excitation element A0 includes six non-excitation elements A1 to A provided on the circumference of the radius r.
6 so as to be surrounded. Preferably, the non-exciting elements A1 to A6 are provided at equal intervals on the circumference of the radius r. The length of each of the excitation element A0 and the non-excitation elements A1 to A6 is configured to be, for example, about 波長 of the wavelength λ of the desired wave, and the radius r is configured to be λ / 4. You. The feeding point of the excitation element A0 is connected via a coaxial cable 9 to a low noise amplifier (LNA).
1 and the parasitic elements A1 to A6 are connected to the variable reactance elements 12-1 to 12-6, respectively, and these variable reactance elements 12-1 to 12-6 are connected.
Are set by a reactance value signal from the adaptive control type controller 7.

FIG. 2 is a longitudinal sectional view of the array antenna device 100. The excitation element A0 is electrically insulated from the ground conductor 11, and each of the non-excitation elements A0 to A6 is grounded at a high frequency to the ground conductor 11 via the variable reactance elements 12-1 to 12-6. Variable reactance element 1
The operation of 2-1 to 12-6 will be described. For example, when the length of the radiating element A0 and the non-exciting elements A1 to A6 in the longitudinal direction are substantially the same, for example, the variable reactance element 12-1 has an inductance property. (L property)
The variable reactance element 12-1 serves as an extension coil, and the electrical length of the non-excitation elements A1 to A6 is longer than that of the excitation element A0, and functions as a reflector. On the other hand, for example, when the variable reactance element 12-1 has a capacitance (C property), the variable reactance element 12-1 becomes a shortened capacitor, and the electrical length of the non-exciting element A1 is
It becomes shorter than 0 and works as a director.

Therefore, the array antenna device 100 shown in FIG.
, By changing the reactance values of the variable reactance elements 12-1 to 12-6 connected to the respective non-exciting elements A1 to A6,
Can be changed.

In the control device for an array antenna shown in FIG. 1, an array antenna device 100 receives a radio signal, and the received signal is input to a low noise amplifier (LNA) 1 via a coaxial cable 9 and amplified. Next, the down-converter (D / C) 2 performs low-frequency conversion of the amplified signal into a signal of a predetermined intermediate frequency (IF signal). Furthermore, A
The / D converter 3 A / D converts the low-frequency converted analog signal into a digital signal, and outputs the A / D converted digital signal to the time domain signal processing unit 4. Next, the time-domain signal processing unit 4 divides the radio signal y (t) received by the array antenna device 100 into a plurality of time-domain sub-signals, and generates a signal vector [Y] including the divided sub-signals. The signal is output to the adaptive control type controller 7, and a plurality of divided sub-signals are each multiplied by a predetermined weighting coefficient and then added to perform signal processing in the time domain to obtain a processed signal z (t). Output. Then, the adaptive control type controller 7 subtracts the processing signal z (t) from the learning sequence signal generated by the learning sequence signal generator 6 to calculate an error signal. Based on the learning sequence signal and the signal vector [Y], the adaptive control process is performed by calculating the optimum weight coefficient vector [W] so as to minimize the error signal and outputting the calculated weight coefficient vector [W] to the time domain signal processing unit 4. Here, specifically, the adaptive control type controller 7 performs the weighting based on the learning sequence signal and the respective sub-signals so that the error signal between the processing signal z (t) and the learning sequence signal is minimized. The coefficient is calculated and output to the time domain signal processing unit 4 to calculate a gradient vector of a predetermined reference function indicating a value corresponding to the error signal, and each variable reactance element is set so that the value of the reference function is minimized. The reactance values of 12-1 to 12-6 are calculated and set.

The transmitting station transmitting the radio signal received by the array antenna 100 is a digital data signal of a predetermined symbol rate including the same learning sequence signal as the predetermined learning sequence signal generated by the learning sequence signal generator 6. , A carrier signal of a radio frequency is modulated using a digital modulation method such as QPSK or a direct spread spectrum spread modulation method, and the modulated signal is power-amplified and transmitted to the array antenna device 100 of the receiving station. . In the embodiment according to the present invention, before performing data communication, a radio signal including a learning sequence signal is transmitted from the transmitting station to the receiving station, and the adaptive control type controller 7 executes an adaptive control process at the receiving station. Is done.

Next, the time domain signal processing section 4 of FIG. 1 will be described in more detail with reference to FIGS. FIG.
Is TD, which is the first embodiment of the time domain signal processing unit 4.
It is a block diagram of the time domain signal processing part 4-1 for MA.
The TDMA time domain signal processing unit 4-1 includes a plurality of (J-1) shift registers (SR) 1 cascaded with each other.
3-1 to 13- (J-1), a plurality J of downsamplers 14-1 to 14-J, a plurality J of transversal filter circuits 23-1 to 23-J, and an adder 17
It is comprised including. Shift register (SR) 1
Each of 3-1 to 13- (J-1) outputs the input signal delayed by one symbol period based on the input clock. The transversal filter circuits 23-1 to 23-J convert the signal data Dx1 to DxJ divided into a plurality of time-delayed sub-signals for calculating the weighting coefficients.
Is output to the adaptive control type controller 7, and the weight coefficient data D calculated by the adaptive control type controller 7
The input signals are multiplied by w1 to DwJ and output.

The received signal y (t) output from the A / D converter 3 in FIG. 1 is input to a downsampler 14-1 and a shift register 13-1. Down sampler 14-1
Down-samples the input received signal y (t) at a sampling frequency of 1 / J times the sampling frequency of the A / D converter 3, and processes the processed signal in a transversal filter circuit 23-1 described in detail later. Adder 1 via
7 is output. The signal output from the shift register 13-1 is sent to the down sampler 14-2 and the shift register 1
Input to 3-2. The downsampler 14-2 downsamples the input signal at a sampling frequency of 1 / J times the sampling frequency of the A / D converter 3, and converts the processed signal into a transversal filter circuit 23.
Output to the adder 17 via -2. Similarly, a signal output from the shift register 13-j (j = 2, 3,..., J-1) is a down-sampler 14- (j + 1),
Output to the shift register 13- (j + 1). The down sampler 14- (j + 1) converts the input signal to A /
Down-sampling is performed at a sampling frequency of 1 / J times the sampling frequency of the D converter 3, and the processed signal is output to the adder 17 via the transversal filter circuit 23- (j + 1). Further, the adder 17 adds the input plurality of J signals, and outputs a signal of the addition result as a processed signal z (t).

FIG. 4 is a block diagram showing a configuration of the transversal filter circuit 23-1 in FIG. The transversal filter circuit 23-1 is connected to the down sampler 22.
-1 is delayed by, for example, 1/4 to 1/2 of one symbol, and a plurality (M-1) of delay circuits 25-1 to 25- connected in cascade with each other. (M-1) and a plurality of M multipliers 26-1 to 26-2
6-M and an adder 27. The signal input to the transversal filter circuit 23-1 is output to the adaptive control type controller 7 as sub-signal data, and is added via the multiplier 26-1 having a multiplication coefficient of the weight coefficient w _1,1. Output to the container 27,
(M-1) delay circuits 25-1 cascaded with each other
To 25- (M-1) and a multiplier 26-M having a multiplication coefficient of the weighting coefficients w1 _{and M} is output to the adder 27. Here, the suffix of the weight coefficient w is the first suffix and the serial numbers 1 to J of the transversal filter circuits 23-1 to 23-J, and the second suffix is the serial numbers 1 to J of the transversal filter circuits 23-J. Represents the serial numbers 1 to M of the multipliers in 1 to 23-J. The delay circuit 25
-1 is output from the adaptive control type controller 7.
And output to the adder 27 via a multiplier 26-2 having a multiplication coefficient of the weight coefficient w1, ₂ .
Further, the signal output from the delay circuit 25-2 is output to the adaptive control type controller 7 is output to the adder 17 via a multiplier 26-3 with a multiplication factor of the weighting factor w _{1, 3} You. Hereinafter, similarly, the delay circuit 26
The signal output from −ma (ma = 3,..., M−1) is output to the adaptive control type controller 7,
Multiplier 26 having a multiplication coefficient of weight coefficient w1 _{, ma + 1}
The signal is output to the adder 27 via − (ma + 1). Then, the adder 27 adds the input M signals, and outputs a signal of the addition result to the adder 17.

The transversal filter circuits 23-2 to 23-J of FIG. 3 include a plurality (M-1) of delay circuits, a plurality M of multipliers, and an adder connected in cascade with each other. Provided, the transversal filter circuit 23-1
The configuration is the same as The time domain signal processing unit 4 combines the signal data Dx1 to DxJ output from each of the transversal filter circuits 23-1 to 23-J into a signal vector [Y] and outputs the resultant to the adaptive control type controller 7. Further, the time domain signal processing unit 4 decomposes the weight coefficient vector [W] input from the adaptive control type controller 7 into weight coefficient data Dw1 to DwJ, and converts the transversal filter circuits 23-1 to 23-J. Is multiplied by the signal input thereto.

FIG. 5 is a block diagram of a second embodiment of the time domain signal processing unit 4 which replaces the first embodiment of FIG.
FIG. 3 is a block diagram of a DMA time domain signal processing unit 4-2. In the present embodiment, instead of the transversal filter circuits 23-1 to 23-J according to the first embodiment, a plurality of J matched filters (matched filters;
Also called a matched filter. 15) through 15-J, and sub-signal processing circuits 16-1 through 16-J connected to the matched filters 15-1 through 15-J. TD of the first embodiment
This is the same as the MA time domain signal processing unit 4-1 and a detailed description thereof will be omitted.

In FIG. 5, J-
One shift register 13-1 to 13- (J-1)
And J downsamplers 14-1 to 14-J,
The configuration is the same as that of the time domain signal processing unit 4-1 for TDMA. The signal output from the downsampler 14-1 is input to the matched filter 15-1, and the matched filter 1
5-1, the down-sampled signal is buried in the white noise based on the data Dcp of the spreading code of the user terminal of the desired wave input from the controller (not shown) of the receiver. The desired wave signal is detected at the maximum SN ratio, and specifically, a pulse signal is output for each period of the spreading code. Next, the signal from the matched filter 15-1 is output to the adder 17 via the sub-signal processing circuit 16-1 described in detail later. Also, down sampler 14-2
Are output to the adder 17 via the matched filter 15-2 and the sub-signal processing circuit 16-2. Similarly, each matched filter 15-j (j =
, J) output the signal output from the downsampler 14-fa to the adder 17 via the sub-signal processing circuit 16-j.

Next, the sub signal processing circuit 16-1 shown in FIG.
A detailed configuration will be described. Sub signal processing circuit 16-
Reference numeral 1 denotes a plurality (Nc-1) of delay circuits 21-1 to 21- connected in cascade with a predetermined delay time Tc.
(Nc-1) and a plurality of Nc downsamplers 22-1
To 22-Nc, a plurality of Nc transversal filter circuits 23-1 to 23-Nc, and an adder 24. The signal output from the matched filter 15-1 is supplied to the delay circuit 21-1 and the downsampler 22.
-1 is output. The downsampler 22-1 downsamples the input signal at a sampling frequency 1 / Nc times the sampling frequency of the downsamplers 14-1 to 14-J, and processes the processed signal through the transversal filter circuit 23-1. And outputs the result to the adder 24.

The transversal filter circuits 23-1 to 23-Nc are used for calculating weighting coefficients. The signal data Dx1 to Dx divided into a plurality of time-delayed sub-signals are calculated.
Nc is output to the adaptive control type controller 7, and the input signals are multiplied by weight coefficient data Dw 1 to DwNc calculated by the adaptive control type controller 7, respectively. The detailed configuration of the transversal filter circuits 23-1 to 23-Nc is the same as that of the TD according to the first embodiment.
This is the same as the transversal filter circuit of the MA time domain signal processing unit 4-1 (see FIG. 4). Here, in order to distinguish each weight coefficient w to be multiplied, the suffix of the weight coefficient w is the first suffix and the serial numbers 1 to J of the sub-signal processing circuits 16-1 to 16-J are replaced by the first suffix. Subscripts of 2 indicate serial numbers 1 to Nc of transversal filter circuits in each of the sub-signal processing circuits, and third subscripts indicate serial numbers 1 to M of multipliers in each of the transversal filter circuits. Shall be represented.

The signal output from the delay circuit 21-1 is input to the delay circuit 21-2 and the downsampler 22-2, and the downsampler 22-2 converts the input signal into
Down-sampling is performed at a sampling frequency 1 / Nc times the sampling frequency of the down-samplers 14-1 to 14-J, and the processed signal is output to the adder 24 via the transversal filter circuit 23-2. Hereinafter, similarly, the delay circuit 21-nc (nc = 2, 3,..., Nc−
1) is output from the delay circuit 21- (nc +
1) and input to the down sampler 22- (nc + 1),
The downsampler 22- (nc + 1) downsamples the input signal at a sampling frequency of 1 / Nc times the sampling frequency of the downsamplers 14-1 to 14-J, and processes the processed signal in the transversal filter circuit 23-. Output to the adder 24 via (nc + 1). Further, the adder 24 adds the input Nc signals and outputs a signal of the addition result to the adder 17.

The sub signal processing circuits 16-2 to 16-J also have the same internal configuration as the sub signal processing circuit 16-1. The adder 17 includes a sub signal processing circuit 16
-1 to 16-J are added, and a plurality of J adaptively controlled signals are added, and the added signal is processed signal z
Output as (t). The time domain signal processing unit 4 includes a plurality of J × Nc transversal filter circuits 23-1 to 23-J in the sub signal processing circuits 16-1 to 16 -J.
The signal data Dx1 to DxNc output from Nc are
The signal is synthesized with the signal vector [Y] and output to the adaptive control type controller 7. Further, the time-domain signal processing unit 4 decomposes the weight coefficient vector [W] input from the adaptive control type controller 7 into weight coefficient data Dw1 to DwNc, and obtains a plurality of J × Nc transversal filter circuits 23. -2 to 23-Nc multiply by the signal inputted thereto.

In the control device for an array antenna configured as described above, the adaptive control type controller 7 controls the signal vector [Y] output from the time domain signal processing section 4.
And a predetermined learning sequence signal, for example, using a predetermined adaptive control algorithm using a minimum mean square error (MMSE) criterion, a plurality of J × Nc × M multiplications such that the error signal is minimized. Calculate the respective weight coefficients for the multipliers 26-1 to 26-M, and calculate the respective multipliers 26-1 to 26-M
Feedback and set.

The adaptive control type controller 7 further includes
Rear for controlling the directivity of the ray antenna device 100
Outputs the reactance value signal. Here, the adaptive control type controller
The roller 7 is, for example, a digital computer such as a computer.
Before starting data communication.
Signal generated by the signal processing unit 4 and the learning sequence signal.
Learning sequence signal sb generated by signal generator 6
_p(M) and is shown in the flowchart of FIG.
By executing the adaptive control process
Point the main beam of the tenor device 100 in the desired wave direction and
Each variable reactance element to direct nulls in the direction of interference
Reactance value X of slaves 12-1 to 12-6₁, ..., X
₆Is calculated and set. Specifically
, The adaptive control type controller 7 controls each variable reactance
Reactance value X of elements 12-1 to 12-6₁,…,
X₆Are sequentially perturbed by a predetermined shift amount ΔX,
A predetermined reference function using the distance value as a variable (in this embodiment,
Is derived from the received signal y (t) in Equation 68 described later.
Calculated sub-signal and the generated learning sequence signal
sb _pCalculate the gradient vector of the function fh) with (m)
You. Next, based on the calculated gradient vector,
Reactance value X such that the quasi-function value is maximized₁,
X₂, ..., X₆And the reactance value X₁, X₂,
…, X₆Variable reactance signal
Output to the switching elements 12-1 to 12-6.
Therefore, the main beam of the array antenna device 100 is desired.
Point nulls in the direction of the wave and in the direction of the interfering wave
Set

Next, the principle of the control device and the control method of the array antenna according to the embodiment of the present invention will be described.

First, N (N
> 1) Arriving at an antenna array composed of elements
Consider a signal model. Sent from transmitting station
Radio signals are input signals defined in a plane including the ground conductor 11.
Launch angle (also called Angle of Arrival (AOA)
U. ) Incident at θ and received by array antenna device 100
Is done. In the present embodiment, focusing on the excitation element A0,
The direction of the parasitic element A1 is defined as θ = 0. The transmitted message
Signal s of the p-th user terminal of the signal
_p(T) is expressed as follows.

[0031]

(Equation 1)

Here, sb _p (m) indicates the m-th information symbol related to the signal of the p-th user terminal, and ρ
_p (t) represents an information symbol waveform. In a TDMA system, the information symbol waveform ρ _p (t) is often the same for each user terminal signal and can be considered as a spread spectrum cosine modulated waveform. T indicates a symbol duration or a symbol period. In a CDMA system, the following equation holds, which is referred to as a pulse shaping function of the p-th user terminal.

[0033]

(Equation 2) (0 ≦ t ≦ T)

Here, {c _p (j)}, j = 0,..., N
c-1 is the spreading code assigned to the p-th user terminal, T is the symbol duration equal to the product of the chip interval Tc and the number of chips per symbol Nc,
(T) is a normalized chip waveform signal defined in the time section [0, Tc]. Further, when the oversampling period is Δ, Tc / Δ = 2, and when the transmission bit rate is fb, the symbol bit rate is 2 × 12.
It is represented by 7 × fb. The spreading code sequence may be periodic or aperiodic depending on the standard employed. In this specification, a periodic case will be considered. The N-dimensional array received signal vector [x (t)] received by the noise-free array antenna device including N antenna elements is expressed as follows. Hereinafter, in the present specification, a vector or a matrix is represented by [•].

[0035]

(Equation 3)

Here, the N-dimensional vector [g _p
(T)] is referred to as a p-th user terminal space-time symbol waveform signal or a symbol-level space-time channel impulse response as in the following equation.

[0037]

(Equation 4)

Θ_l ^p, Τ_l ^p, Ξ_l ^pIs the p-th
Arrival angle corresponding to the l-th path of the signal of the eye user terminal
(AOA), representing time delay and propagation loss. Furthermore, N
The dimension vector [a (θ)] is an array tear corresponding to θ.
Represents a ring vector, sb _p(M) and L_pEach
And the m-th information symbol related to the signal of the p-th user terminal.
Shows the total number of volts and multipath waves. For the component of Equation 3
Therefore, the following items are assumed. <Assumption 1> The signal to be received is fractionally spaced.
wide if sampled at a symbol period of paced)
Is a periodic steady state in the sense of
When it is ringed, it is a broadly defined steady state. Broad cycle
The steady-state signal vector [x (t)] is defined by the following equation.
Be defined.

[0039]

E ｛[x (t ₁ )] [x (t ₂ )] ^H } = E
{[X (t ₁ + T)] [x (t ₂ + T)] ^H }

Here, [•] ^H represents the conjugate transpose and E ｛· ｛
Indicates a statistical expectation. <Assumption 2> Information symbol sb _p (m), p = 1, 2,
, P are independent and identically distributed, and satisfy the following expression.

[0041]

[Equation 6] E {sb _p (m) sb _p ^* (n)} = δ _{p, q} δ _{m, n}

Here, [•] ^* indicates a complex conjugate, and δ _{p, q} indicates a Kronecker delta function. <Assumption 3> A plurality of channels {g _p (t)}, p = 1,
2,..., P are linear and time-invariant during the period of interest for performing the given data communication and belong to a finite duration in the time interval [0, D _p T].

Next, a model of a signal received by the array antenna device 100 is formulated. A received signal y (t) without noise output from the array antenna apparatus 100 including the excitation element A0 and the non-excitation elements A1 to A6 shown in FIG. 1 is specified by the following equation (prior art document 3).
"Takahiro Ohira et al.," Equivalent Weight Vector of ESPAR Antenna and Array Factor Expression ", IEICE Technical Report, AP2000-44, SAT2000-41, M
W2000-44, July 2000 ").

[0044]

Y (t) = [i] ^T [x (t)]

The steering vector [a (θ)]
Is represented by the following equation.

[0046]

[A (θ)] = [1, exp (j (2πr /
λ) cos (θ)), ..., exp (j (2πr / λ) c
os (θ−5 × 2π / 6))] ^T

Here, the diameter of the array is r = λ / 4, λ represents the wavelength of the radio frequency of the desired wave, and the equivalent weight vector [i] considered in the prior art document 3 is as follows: Derived.

[0048]

[I] = C [I + YX] ⁻¹ [y ₀ ]

Here, I is a unit matrix.

[0050]

[Y ₀ ] = [y ₀₀ , y ₁₀ , y ₁₀ , y
₁₀ , y ₁₀ , y ₁₀ , y ₁₀ ] ^T

[Equation 11]

(Equation 12)

X is a reactance matrix for adjusting the antenna pattern, R ₀ = 50Ω is the input impedance of the radio receiver, and X ₁ ,..., X ₆ are reactance value signals from the adaptive control type controller 7. Is a parameter output as Y is an admittance matrix representing the mutual coupling between the elements of the antenna, and [y ₀ ] is an associated admittance vector, the components of which include:

[0052] _{(a) y 00} represents the self-input admittance of the radiating element A0. _{(B) y 10} is the excitation element A0 and the parasitic elements A1 through A6
Represents the binding admittance with _{(C) y 11} represents a self-input admittance of the parasitic elements A1 through A6. (D) y ₂₁ is the parasitic elements A1 and A2 adjacent to each other,
The binding admittance of A2 and A3, A3 and A4, A4 and A5, A5 and A6, or A6 and A1. (E) y ₃₁ has two parasitic elements interposed therebetween
Two parasitic elements A1 and A3, A2 and A4, A3 and A5,
(F) y ₄₁ denotes two non-exciting elements A1 and A4, A2 and A5, A3 and A6, which face each other across the exciting element A0. Represents the binding admittance of

Because of the reciprocity and the cyclic symmetry of the array antenna device 100, only the six components are independent as described above. C is a coefficient relating to the gain of the antenna. In the case of the array antenna apparatus 100 shown in FIG. 1, a value of C = 131.2 is approximately obtained from the actual measurement result. Table 1 shows the admittance vector [y ₀ ] and different input values (entries) to the admittance matrix Y.

[0054]

[Table 1] ―――――――――――――――――――――――――――――――――――― y ₀₀ = 0.00860035−0.0315844j _{y 10 = -0.00372642 + 0.0072319j y 11} = 0.00962295-0.01656835j y 21 = -0.000377459 + 0.0117867j y 31 = 0.00002720885-0.0063736j y 41 = 0.001779525 + 0.002208335j ---- ――――――――――――――――――――――――――――――――

By substituting Equation 3 into Equation 7 and considering additive noise, the received signal y (t) output from a single port of the array antenna apparatus 100 can be expressed by the following equation.

[0056]

(Equation 13)

Here, the following function included in Expression 13 is also expressed by p
It is called the spatio-temporal symbol waveform signal of the user terminal.

[0058]

[Equation 14]

Here, f (θ) represented by the following equation is a pattern of the array antenna device 100.

[0060]

F (θ) = f (θ, X ₁ ,..., X ₆ ) =
[I] ^T [a (θ)]

The two spatiotemporal impulse responses g _p (t) and [g _p (t)] clearly have the same duration. Additive noise satisfies the following assumptions. <Assumption 4> Additive noise is zero-mean white noise that satisfies the following two equations, and is uncorrelated with the signal of the user terminal.

[0062]

E ｛n ² (t)} = 0

E | n (t) | ² = σ ²

Here, σ ² represents the power of the noise.

From the equations 9 to 14, the array antenna apparatus
The output signal of 100 is also reactance X ₁, X₂, ..., X₆
It can be seen that this is a nonlinear function of

Next, a description will be given of spatio-temporal adaptive filtering performed by the time-domain signal processing unit 4 for removing an undesired signal. Array antenna device 1
00 variable reactance elements 12-1 to 12-6,
The temporal processing of the array antenna control device having a given set of reactance values is first executed using the TDMA time domain signal processing unit 4-1 shown in FIG. In the case of TDMA, processing is performed based on a symbol waveform signal. Shift registers 13-1 to 13-1
3- The sampling period in (J-1) is represented by Δ,
J = T / Δ (J is an integer of 1 or more) is used as an oversampling coefficient. The received signal y (t) is assumed to have a time t = iΔ + mT (where m is an arbitrary integer; i = 0,
1,..., J−1), Equation 13 is as follows.

[0066]

(Equation 18) (I = 0, 1,..., J-1)

When the periodic steady state of the signal of the terminal described in assumption A1 is used (refer to L. Tong et.
al., "Blind identification and equalization based
onsecond-order statistics: a time domain approac
h, "IEEE Transaction. Information Theory, Vol. 40,
pp. 340-349, March 1994 ". 3), the extended multi-channel model method of the transversal filter circuits 23-1 to 23-J, which are fractionally spaced equalizers shown in FIG. 3, can be easily established as follows. .

[0068]

[Equation 19]

Here, the J-dimensional signal vector [yb
(T)], a spatiotemporal impulse response vector [h
_p (d)] and the noise vector [nb (m)] are represented by the following equations.

[0070]

[Yb (m)] = [y (mT), y (mT−
Δ),..., Y (mT− (J−1) Δ)] ^T

[Number 21] _{[h p (d)] =} [ga p (dT), ga p
_{(DT-Δ), ...,} ga p (dT- (J-1) Δ)] T

[Nb (m)] = [n (mT), n (mT−
Δ),..., N (mT− (J−1) Δ)] ^T

For each m, the dimension of the received signal [yb (m)] is J, where J is called the “number of oversampling channels”. Regarding the limitation of the number of extension channels due to oversampling, refer to the prior art document 5 “AJ
van der Veen, "Resolutionlimits of blind multi-use
r multi-channel identification scheme-the band-l
imited case ", in Proceeding of ICASSP '96, Atlant
a, GA, May 1996. " For successive samples within the period of the M symbols, the next J × M-dimensional signal vector [Y _T (m)] and the symbol vector [S _T composed of M + D _p information symbols related to the signal of the p-th user terminal] _p (m)] and a J × M-dimensional noise vector [N
(M)].

[0072]

[Y _T (m)] = [yb (m), yb (m−
1),..., Yb (m−M + 1)] ^T

[S _p (m)] = [sb _p (m), sb
_p (m−1),..., sb _p (m−M−D _p +1)] ^T

[N (m)] = [nb (m), nb (m−
1),..., Nb (m−M + 1)] ^T

The next sylvester (Sy) related to the user terminal p
lvester) is the convolution matrix of the channel (D _p
+1) × J length (dimension) impulse response [[h
^{_{p (0)] T, [}} h p (1)] T, ..., [h
_p (D _p )] ^T ] When defined by the term ^T , M
It becomes a J × (M + D _p ) order matrix.

[0074]

(Equation 26)

Here, “0” represents a J-dimensional zero vector. Equation 19 can be extended to the following equation.

[0076]

[Equation 27]

Therefore, the time domain signal processing unit 4 for TDMA
The equalization for the sub-signal at -1 can be performed by:

[0078]

(28) z _T (m) = [W] ^T [Y _T (m)]

Here,

[W] = [w _1,0 ,..., W _{J, 0} ,.
_{1, M−1} ,..., W _{J, M−1} ] ^T is a weight coefficient vector for the transversal filter circuit 23-1 which is the equalizer illustrated in FIG.
Based on the minimum mean square error (MMSE) criterion, the optimal weighting factor for the transversal filter circuit 23-1 is solved from equation (30) and given by the known Wiener-Hop solution, ie, equation (31). .

[0080]

[Equation 30]

[W _MMSE ] ^* = [ _RT ] ⁻¹ [r (v)]

Here, sb ₁ (m) is a learning sequence signal of a signal of a desired user terminal, and v ≧ 0 is causal filtering (causal filteri) in consideration of delay time v.
ng), which is the delay of the learning sequence signal required to realize ng).
As is apparent from Equation 30, the adaptive control type controller 7
Calculates a weight coefficient vector [W] such that an error between a signal sb ₁ (m−v) obtained by delaying a learning sequence signal by a predetermined delay time v and a processed signal z _T (m) is minimized. Adaptive control. [ _RT ] and [r (v)]
Are the temporal correlation matrix of the signal vector and the correlation vector between the learning sequence signal and the signal vector, respectively, calculated as follows.

[0082]

[R _T ] = E ｛[Y _T (m)] [Y
_T (m)] ^H ｝

[R (v)] = E ｛sb ₁ ^* (m−v) [Y
_T (m)]｝

The adaptive control type controller 7 outputs the weight coefficient vector [W] obtained by the equations (31) to (33) to the time-domain signal processing unit 4, and outputs the weight coefficient vector [W].
Is multiplied by the signal vector [Y _T ] in a plurality of J × M multipliers 26-1 to 26-M, and the resulting signal is added by the adders 27 and 17 and output. On the basis of the error signal between the output signal z _T (k) and the learning sequence signal, the adaptive control controller 7 repeats the above-described processing to converge the error of Expression 30, thereby obtaining the output signal z _T (k). Is minimized. Also,
Variable reactance element 12 of array antenna device 100
The minimum residual error power when -1 to 12-6 have a given set of reactance values is determined as:

[0084]

(Equation 34)

In practice, the minimum residual error power of Equation 34 is a function of the reactance values X ₁ ,..., X ₆ .

The time domain processing of the array antenna controller when the variable reactance elements 12-1 to 12-6 of the array antenna device 100 have a given set of reactance values is now illustrated in FIG. CDMA
The case where the time domain signal processing unit 4-2 is used will be described. In the case of CDMA, the time domain processing involves a pulse shaping function and their associated matched filter 1.
This is performed on output signals from 5-1 to 15-J.
The sampling period in the shift registers 13-1 to 13- (J-1) is represented by Δ = Tc / J (J is an oversampling coefficient which is a natural number), and the received signal y (t) is represented by time t = laTc-iΔ, ( la is a natural number; i = 0,
1,..., J−1), the discrete form of Expression 13 is as follows.

[0087]

(Equation 35)

The discretized received signal y (laTc-i
Δ), i = 0,..., J−1 are stacked to obtain a signal vector represented by the following equation.

[0089]

[Equation 36]

Here, [yv (laTc)], [gv _p
(LaTc)] and [nv (laTc)] mean a signal vector, a spatio-temporal symbol waveform signal, and a J-dimensional vector indicating noise, respectively, as follows.

[0091]

[Yv (laTc)] = [y (laTc),
…, Y (laTc− (J−1) Δ)] ^T

[Number 38] _{[gv p (laTc)] =} [ga p (laT
_{c), ..., ga p (} laTc- (J-1) Δ)] T

[Nv (laTc)] = [n (laTc),
.., N (laTc- (J-1) Δ)] ^T

The following equation is assumed for the normalized chip waveform.

[0093]

４０ (kTc−laTc) = δ _kl

At this time, the discrete pulse shaping function of the p-th user terminal is represented by the following equation.

[0095]

[Equation 41] (0 ≦ la ≦ Nc-1)

For simplicity of notation,

[Number 42] _{cb p (la) = c p} (Nc-la), 0
If ≤la≤Nc-1, the matched filter 15-1
To the 15-J, the output signal vector after despreading pulse waveform shaping function p ₀ th user terminal can be shown by the following equation.

[0097]

[Equation 43]

Here,

[Equation 44]

[Equation 45]

Similar to the formulation of the vector in Equation 19,
Tc is represented by kT−jTc (here, 0 ≦ j ≦ Nc−1), and a symbol-level signal vector in the sub-signal processing circuits 16-1 to 16-Nc is defined as follows.

[0100]

[Xc (kT)] = [[Xb (kT)] ^T ,
…, [Xb (kT- (Nc-1) Tc)] ^T ] ^T

From Equation 43, Equation 46 can be expressed by the following equation.

[0102]

[Equation 47]

Here,

[Equation 48]

[Equation 49]

From assumption 3, a J × Nc-dimensional waveform signal

[Equation 50] Is known to have a limited duration. Therefore, Equation 47 can be expressed as the following equation.

[0105]

(Equation 51)

Here,

(Equation 52) It is of p ₀ th user terminal channel symbol level in length. Equation 51 shows that the first term on the right-hand side contains all components of the signal from the desired user terminal, while the second term contains the superimposed cross-correlation component of the signal from the unwanted user terminal. Is shown. The cross-correlation component must be suppressed. Based on Equation 51, the symbol-level adaptive processing can be performed as:

[0107]

(Equation 53)

Here, the J × Nc × M-dimensional signal vector [Y _C (k)] and the weight coefficient vector [W] are represented by the following equations.

[0109]

Equation 54] _{[Y C (k)] =} [[Xc (kT)] T,
..., [Xc (kT- (M-1) T)] ^T ] ^T

[0110]

[W] = [[w ₀ ] ^T ,..., [W _M−1 ] ^T ]

[W _ma ] = [w _{1,1, ma} , w]
_{2,1, ma} , ..., wJ _{, 1, ma} , _{w1,2, ma} ,
w _{2,2, ma} , ..., w _{J, 2, ma} , ..., w
_{1, Nc, ma, w2, Nc, ma} , ... _{, wJ, Nc,
ma} ] ^T (ma = 1, 2,..., M)

Expression 56 is a signal vector [Y _C (kT)]
, And a weighting coefficient vector [W] of Expression 55 is generated from all the weighting coefficients. Number of taps of the transversal filter circuit 23-1 to 23-Nc M is the length of _{p 0} th user terminal channel symbol level

[Equation 57] , The number of user terminals on the same channel, and performance requirements. As in the case of TDMA, MM
Based on the SE criterion, that is, Equation 58, an optimal weight coefficient vector is obtained as shown in Equation 59.

[0112]

[Equation 58]

[Equation 59]

Here, the temporal correlation matrix [R _c ] of the signal vector and the correlation vector [γ _p0 (v)] between the learning sequence signal and the signal vector are represented by the following equations, respectively.

[0114]

Equation 60] _{[R C] = E {[} Y C (k)] [Y C (k)] H}

[Equation 61]

Here,

(Equation 62) Denotes a p ₀ th user terminal of the learning sequence signal (learning symbol sequence). As is apparent from Equation 58, the adaptive control type controller 7 outputs the signal sb _p0 (k−
The adaptive control is performed by calculating the weight coefficient vector [W] so that the error between v) and the processed signal z _C (k) is minimized and the array antenna controller outputs the best performance. The adaptive control type controller 7 outputs the weight coefficient vector [W] obtained by Expressions 59 to 61 to the time domain signal processing unit 4, and the weight coefficient vector [W] is J ×
Nc transversal filter circuits 23-1 to 23-2
In 3-Nc, the signal is multiplied by the signal vector [Y _C ], and the resulting signal is added in the adder 24 and output from the sub-signal processing circuits 16-1 to 16-Nc. Based on the error signal between the output processing signal z (k) and the learning sequence signal, the adaptive control type controller 7 repeats the above-described processing to converge, thereby obtaining the residual error of the output processing signal z (k). Minimize power. Number 34
Similarly, the minimum residual error power when the variable reactance elements 12-1 to 12-6 of the array antenna apparatus 100 have a given set of reactance values is expressed by the following equation.

[0116]

[Equation 63]

As described above, the adaptive control type controller 7 can adaptively process the desired signal in the time domain in the time domain signal processing unit 4 and can process the desired signal in the spatial domain in the array antenna apparatus 100 at the same time. (Spatio-temporal adaptive filtering).
From the above description, when the variable reactance elements 12-1 to 12-6 of the array antenna apparatus 100 have a predetermined set of reactance values, the CDM is also used for TDMA.
It can be seen that in both cases A, the process formulation can be included in the same method. Hereinafter, the processing signal z (m) is represented by TDMA
, CDMA, and CDMA, and is represented by the following equation.

[0118]

[Expression 64] z (m) = [W] ^T [Y (m)]

As described above, the signal vector [Y (m)]
Are also variable reactance elements 12-1 to 12-6.
Adaptive filtering using the optimal spatiotemporal processing for the signal of the p-th user terminal together with the weight coefficient vector [W] and the reactance values X ₁ , X ₂ ,..., X ₆ Simultaneously referencing and minimizing Equation 65, that is, expressed as Equation 66.

[0120]

Σ _total ² = E | sb _p (m−v) −z (m) | ²

[Equation 66]

Under a given set of data,
The solution of the six optimal weighting factor vectors and reactance values is known to perform a global search on the associated field. However, it is not possible to actually use such a time-consuming global search. Therefore, it is necessary to consider some alternative method.

The most basic method among the optimization methods is as follows.
It is an alternative search method based on coordinates and has been successfully applied to many uses. In the specification of the present application, using the alternative search method based on the coordinates,
Solve the optimization problem of.

Next, a description will be given of a block update algorithm for actually performing the space-time combined adaptive filtering. In the above description, the procedure for calculating the weight coefficient vector [W] has been described, assuming that the reactance values of the variable reactance elements 12-1 to 12-6 have predetermined values. In the following part of the specification of the present application, the adaptive control processing of the reactance value of the array antenna device 100 executed by the adaptive control type controller 4 will be described based on the flowchart of FIG. In terms of alternative search based on coordinates,
Is formulated from two-stage procedures with different viewpoints. First, as an explanation of a general procedure, it is assumed that reactance values X ₁ , X ₂ ,..., X ₆ are fixed, and an optimal weight coefficient vector is solved. This is shown in Equation 31 or Equation 59. Therefore, Equation 66 is as follows.

[0124]

[Equation 67]

Here,

[W _opt ] ^* = [R] ⁻¹ [r (v)]

[R] = E {[Y (m)] [Y (m)] ^H }

[R (v)] = E {sb _p ^* (m−v) [Y (m)]}

The delay time v of the learning sequence signal is given by
The adaptive control type controller 7 determines in advance such that an error between the learning sequence signal delayed by the time v and the processing signal z (t) is minimized on the basis of 0 and Expression 58. The following equation is the power of the symbol signal of the p-th user terminal.

[0127]

Σ _p ² = E | sb _p (m) | ²

Next, the procedure of block update for solving the optimization problem will be described more specifically. The optimal reactance value is searched for according to the limited length of the received data and Equation 67, and thereby the temporal correlation matrix [R] of the signal vector and the correlation vector [r] between the learning sequence signal and the signal vector.
(V)]. That is, the following two operations are executed.

[0129]

[Equation 72]

[Equation 73]

Here, m_k= N_tl, l = 0,1, ...,
M_tAnd M_tIndicates the number of data blocks required for convergence
You. Number of symbols N_tIs the security of the given reactance value.
Below the Least Mean Square (LMS) algorithm
By use, the weight coefficient vector [W] is N_tPieces of thin
It is chosen so that it can converge to a steady state within the Vol period.
Equation 7, Equation 9, Equation 13, Equation 26, Equation 54 and related items
The signal vector [Y (m)] is calculated by the equation
Value X₁, X₂, ..., X₆Is an explicit function
I understand that there is no. In other words, the quadratic form of
Term [r (v)] ^H[R]^-1[R (v)] is Reactane
Value X₁, X₂, ..., X₆Implies the implicit function of
You.

For the implicit function, the optimal reactance (X
h _{1, Xh} 2, ..., the appropriate update algorithm to discover the Xh _{6) opt} is the steepest descent algorithm for updating the reactance, reactance value _X 1,
The term [r (v)] ^H [R] relating to X ₂ ,..., X ₆
A gradient vector of ^-1 [r (v)] must be evaluated. The term [r (v)] ^H [R] ⁻¹ [r (v)] is
Since it is evaluated according to a given data block of limited length, the block update algorithm uses the term [r
(V)] ^H [R] ⁻¹ [r (v)]. The following equation is assumed as a reference function.

[0132]

[Equation 74]

Here, the second term on the right side is a reactance value X
₁, ..., X₆Is a variable. σh_p ²Is the p-th user
The estimated power of the terminal's symbol signal. Steepest descent
The following algorithm (Prior Art Document 6 “R. A. Monzingo et
al., "Introduction to Adaptive Arrays", John Wiley
 & Sons, Inc., 1980)).
From 74, execute adaptive control processing for reactance value
Then, the following update equation is obtained.

[0134]

[Equation 75]

Here,

Xv = [X ₁ , X ₂ ,..., X ₆ ] ^T

[Number 77] _{^{Xv (k) = [X 1}} (k), X 2 (k),
…, X ₆ ^(k) ] ^T

[0136]

[Equation 78]

[Expression 79]

Here, α is a step size for updating, and takes a value of, for example, 1000 to 2000.

The procedure for updating the reactance value by the adaptive control process shown in FIG. 6 is executed as follows. In step S1, a number ε for controlling the number of repetitions of updating the reactance value is set, and the number k of reactance updates is set to 0 as an initial state. Next, step S2
, The initial value Xv ^{(0 )} of the reactance value vector
= (X ₁ ⁽⁰⁾ , X ₂ ⁽⁰⁾ ,..., X ₆ ⁽⁰⁾ ). Then, in step S3, a reactance value signal corresponding to the reactance value vector Xv ^(k) is generated. The output is set to the variable reactance elements 12-1 to 12-6. For example, the initial value of the reactance value vector can be set to a zero vector, and the update algorithm can be started from an omnidirectional beam pattern (see FIG. 9). Then, in step S4, the received signal vector [Y (m)] and the learning sequence signal vector sb
Based on _p (m), a correlation matrix [R] and a correlation vector [r (v)] are calculated using Equations 72 and 73, and an optimal weight vector [W _opt ] is calculated using Equation 68. And outputs it to the time domain signal processing unit 4. Next, in step S5, the reference function fh is calculated using Expressions 78 and 79.
, And the reactance value vector Xv ^(k) is calculated from the reactance value vector Xv ^(k) by Expression 75.
Calculate ^{(k + 1)} . Next, in step S6, it is determined whether or not the following inequality holds.

[0139]

| Fh (Xv ^{(k )} ) − fh (Xv ^{(k + 1)} ) | ≦ ε

Here, ε is an iterative threshold value. When the inequality of Equation 80 is satisfied in step S6, the process proceeds to step S8. When not satisfied (NO), the process proceeds to step S7, and k is incremented by 1. And step S
Return to 3. In step S8, a reactance value signal corresponding to the reactance value vector Xv ^{(k + 1)} is generated, output to the variable reactance elements 12-1 to 12-6, set, and the adaptive control process ends.

If the spatio-temporal adaptive filtering based on the ESPAR antenna described above is used, the beam of the array antenna device 100 is steered in the arrival direction of the desired signal to suppress spatial interference, and the time domain signal processing unit 4 Accordingly, temporal interference such as ISI included in a received signal can be suppressed.

[0142]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present inventors have performed a computer simulation for the array antenna control device of FIG.
The effectiveness of spatio-temporal adaptive filtering by using this array antenna controller was confirmed. In this simulation, there are 15 DS-CDMA user terminal signals of the same channel in the private network system, and user 1 is a desired user. The code lengths of the signals of all the user terminals are set to 127. The signal of each user terminal has six multipath waves,
The AOA of the signal path of each user terminal whose angles are set to have an interval of 8 degrees from each other has a Gaussian distribution, and their time delays follow an exponential distribution with a delay spread at 1.1 symbol periods. Assume. The propagation loss of the multipath wave is included in the SNR of the array element of the direct wave of the signal of the user terminal. In this case, user 1
Is assumed to be −10 dB, and the SNR of signals of all other user terminals is −26.
It changes randomly from 55 dB to -4.76 dB. All user terminals are assumed to be uniformly distributed within the field of view of the array antenna device 100. Table 2 describes the detailed parameters of the signal of the terminal of the user 1.

[0143]

[Table 2] ―――――――――――――――――――――――――――――――――― Route θ (degrees) τ (symbols) ξ (Propagation loss) ―――――――――――――――――――――――――――――――――― １ 12.30 0 −0.9669 + 0. 2550j2 21.50 0.04 0.7437-0.3081j3 20.20 0.05-0.5206-0.5100j4 8.70 0.12-0.3081-0.4569j5 23.400 .33 −0.1806 + 0.3931j 6 13.20 0.47 −0.1912 + 0.1275j ――――――――――――――――――――――――――――― ―――――――

Here, the oversampling coefficient is J
= 1, and the number of taps of the transversal filter circuit is set to M = 1. As described above, for example, the variable reactance elements 12-1 to 12-6 include:

Xv = [− 53, −136,61,51,5]
9, -146] When having a given set of reactance value, such as ^T, N _t is the number of samples of a symbol level, the number 68 of the number 64 the weighting coefficient vector a conventional LMS algorithm based on this It can converge to its steady state.

[0145] Figure 7 is in the adaptive control process of FIG. 6 is a graph showing an example of the convergence curve of the residual error power to determine the number of samples N _t symbols level to converge to the steady state. As can be seen from FIG. 7, the weight coefficient vector converges to its steady state within about 200 symbol periods. This is the number of symbol periods N _t = 20
It means that 0 can be adopted. In this iteration of the reactance of the simulation, N _t = 200 is adopted as the number of symbol periods. In order to show the behavior at the time of convergence of the update algorithm, the number of data blocks M _t = 7 ×
20 and the iteration threshold ε = 1 × 10 for Equation 81
Set ^-10 .

FIG. 8 is a graph showing an example of a convergence curve of a reference function value for updating a data block in the adaptive control processing of FIG. 6, and FIG. 9 is a graph showing a case where the adaptive control processing of FIG. , The initial value Xv of the reactance value vector
⁶ is a graph of a beam pattern of the array antenna device 100 corresponding to ⁽⁰⁾ = ^{(0, 0, 0, 0, 0, 0)} .
Starting from the initial value Xv ^{(0) of the} reactance value vector, the adaptive control process shown in FIG. 6 is executed and updated, and the number of updates k is 2, 4, 9, 13, and 19, respectively. The beam patterns at this time are shown in FIGS. Table 2
14 that all multipath waves of the desired user terminal signal are encompassed and enhanced by the steady state pattern, and that the undesired user terminal signal multipath wave is reduced by the lower lobes of the steady state pattern. It can be seen that it is reduced. From these two drawings, it can be seen that the spatio-temporal adaptive filtering can be effectively realized by performing the beam pattern formation of the array antenna apparatus 100 and the temporal equalization of the received signal in the time domain signal processing unit. it is obvious.

[0147]

As described above in detail, according to the present invention, a radio signal received by an ESPAR antenna is divided into a plurality of time-domain sub-signals, and a predetermined By performing a signal processing in the time domain by multiplying and adding the weighted coefficients and outputting the processed signals, a processing signal and the learning sequence signal are generated based on a predetermined learning sequence signal and each of the sub-signals. The weighting coefficient is calculated so that the error signal is minimized, and the calculated weighting coefficient is output to the time domain signal processing means, and a gradient vector of a predetermined reference function indicating a value corresponding to the error signal is calculated, and the value of the reference function is calculated. The reactance value of each of the above-mentioned variable reactance elements is calculated and set so that is minimized. Therefore, compared to the prior art, it has a simple configuration and the manufacturing cost is low, and it is possible to perform the spatio-temporal adaptive processing for the ESPAR antenna. In addition, co-channel interference signals can be effectively and spatially suppressed by adaptive beam pattern formation, and inter-symbol interference signals can be effectively suppressed by adaptive equalization based on temporal waveforms. Can be.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a control device for an array antenna according to an embodiment of the present invention.

FIG. 2 is a cross-sectional view of the array antenna device 100 of FIG.

FIG. 3 is a block diagram showing a first embodiment of the time domain signal processing unit 4 of FIG. 1 and showing a configuration of a TDMA time domain signal processing unit 4-1.

4 shows a transversal filter circuit 23- in FIG.
1 is a block diagram showing a configuration of FIG.

FIG. 5 is a block diagram showing a second embodiment of the time domain signal processing unit 4 of FIG. 1 and showing a configuration of a CDMA time domain signal processing unit 4-2.

FIG. 6 is a flowchart illustrating an adaptive control process executed by the adaptive control type controller 7 of FIG. 1;

In the adaptive control process in FIG. 7 6 is a graph showing an example of the convergence curve of the residual error power to determine the number of samples N _t symbols level to converge to the steady state.

8 is a graph showing an example of a convergence curve of a reference function value for updating a data block in the adaptive control processing of FIG. 6;

9 shows an initial value Xv ⁽⁰⁾ = (0,0,
Array antenna device 100 corresponding to (0,0,0,0)
6 is a graph of a beam pattern of FIG.

10 is a graph of the beam pattern of the array antenna device 100 when the adaptive control process of FIG. 6 is executed and the number of updates of the reactance value is two.

11 is a graph of the beam pattern of the array antenna device 100 when the adaptive control process of FIG. 6 is executed and the number of times the reactance value is updated is four.

12 is a graph of the beam pattern of the array antenna device 100 when the adaptive control process of FIG. 6 is executed and the number of times the reactance value is updated is nine.

13 is a graph of the beam pattern of the array antenna device 100 when the adaptive control process of FIG. 6 is executed and the number of times the reactance value is updated is 13.

14 is a graph of the beam pattern of the array antenna device 100 when the adaptive control process of FIG. 6 is executed and the number of times the reactance value is updated is 19 times.

[Explanation of symbols]

A0: Exciting element, A1 to A6: Non-exciting element, 1: Low noise amplifier, 2: Down converter, 3, A / D converter, 4, Time domain signal processing section, 4-1: Time domain signal processing for TDMA Section, 4-2: CDMA time domain signal processing section, 6: learning sequence signal generator, 7, adaptive control type controller, 9: coaxial cable, 11: ground conductor, 12-1 to 12-6: variable reactance element 13-1 to 13- (J-1): shift register; 14-1 to 14-J; 22-1 to 22-Nc: downsampler; 15-1 to 15-J: matched filter; 16-J: sub signal processing circuit; 17, 24, 27: adder; 23-1 to 23-J, 23-1 to 23-Nc: transversal filter circuit; 21-1 to 21- (Nc-1) ), 25-1 to 25-
(M-1) ... delay circuits, 26-1 to 26-M ... multipliers, 100 ... array antenna device.

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Takashi Ohira 2-2-2 Kodai, Seika-cho, Soraku-gun, Kyoto F-term in ATR Eco-Adaptive Communication Research Laboratories (reference) 5J021 AA05 AA08 CA06 DB02 DB03 EA04 FA14 FA15 FA16 FA17 FA20 FA26 FA29 FA32 GA02 HA05 HA10 5K046 AA05 EE06 EE47 EE56 EF13 5K059 CC03 CC04 DD31

Claims (2)

[Claims] A radiating element for receiving a radio signal;
A plurality of passive elements provided at a predetermined distance from the radiating element, and a plurality of variable reactance elements connected to the plurality of passive elements, respectively, wherein a reactance value of each of the variable reactance elements is changed. By operating the plurality of variable reactance elements as a director or a reflector, and controlling the array antenna to change the directional characteristics of the array antenna, a radio signal received by the array antenna is transmitted for a plurality of times. Time-domain signal processing means for performing time-domain signal processing by dividing into a plurality of divided sub-signals, multiplying each of the plurality of divided sub-signals by a predetermined weighting coefficient, and then adding the resultant signals to output as a processed signal Based on a predetermined learning sequence signal and each of the sub-signals, The weight coefficient is calculated and output to the time-domain signal processing means so that an error signal from the sequence signal is minimized, and a gradient vector of a predetermined reference function indicating a value corresponding to the error signal is calculated. An adaptive control means for calculating and setting the reactance value of each of the variable reactance elements so that the value of the reference function is minimized. 2. A radiating element for receiving a radio signal;
A plurality of passive elements provided at a predetermined distance from the radiating element, and a plurality of variable reactance elements connected to the plurality of passive elements, respectively, wherein a reactance value of each of the variable reactance elements is changed. By operating the plurality of variable reactance elements as a director or a reflector, and controlling the array antenna to change the directional characteristic of the array antenna, the radio signal received by the array antenna is transmitted for a plurality of times. Dividing into a plurality of sub-signals in a region, multiplying each of the plurality of divided sub-signals by a predetermined weighting coefficient, and then adding the multiplied sub-signals to perform time-domain signal processing and outputting the processed signals as a processed signal; Based on the learning sequence signal and each of the sub-signals, the processing signal and the learning sequence Calculate the weighting coefficient so that the error signal with the signal is minimized, output the weighting coefficient for signal processing in the time domain, and calculate a gradient vector of a predetermined reference function indicating a value corresponding to the error signal, Calculating and setting the reactance value of each of the variable reactance elements so that the value of the reference function is minimized.