JP2002076746A - Adaptive array device, calibration method, and program recording media - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an adaptive array device, which conducts calibration without attaching a supplement device, and a calibration method as well as program recording media. SOLUTION: A response vector, which transmits specific signals from Ant2 and shows the incoming direction of interference signals and the specific signals from Ant3 and Ant4, and a virtual response vector, which has no correlation with this response vector, are calculated. A weight vector, which turns the beam in the direction shown by the virtual response vector and turns the null in the direction shown by the response vector calculated by the first calculation unit, is calculated. Signals are array-transmitted to Ant3 and Ant4 using the calculated weight vector. As a phase volume and an amplitude volume of transmission signals by Ant4 are changed, the phase volume and the amplitude volume which minimize the reception signal level of Ant2 are determined as the correction value.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an adaptive array device for measuring a characteristic difference between a plurality of radio units in an adaptive array device, a calibration method, and a recording medium on which a calibration program is recorded.

[0002]

2. Description of the Related Art In recent years, an adaptive array system for adaptively forming a directivity pattern for a mobile communication device has been put to practical use in a radio base station for wirelessly connecting a mobile communication device such as a PHS or a portable telephone. The adaptive array system is a system that includes a plurality of radio units including an antenna, a transmission unit, and a reception unit, and forms a sharp directivity pattern (referred to as an array antenna pattern) in the direction of a mobile communication device as a whole antenna. The array antenna pattern is formed by adjusting the amplitude and phase of the reception signal and the transmission signal for each radio unit. The adjustment of the amplitude and the phase is performed by weighting the reception signal and the transmission signal of each radio unit using a weight coefficient (called a weight vector). This weighting factor is determined by the DSP (Digital Signal Proces
sor). For the calculation of the weight coefficient, for example, see “Adaptive signal processing by array antenna”.
(Nobuyoshi Kikuma, published by Science and Technology Publishing Co., Ltd. on September 20, 1998).

[0003] In a radio base station of a mobile communication system, the same array antenna pattern as that at the time of reception is formed at the time of transmission by weighting a transmission signal using the weight vector calculated at the time of reception. However, even if the weight vector calculated at the time of reception is used at the time of transmission, the same array antenna pattern is not always formed at the time of transmission and reception. This is because transmission characteristics (that is, phase fluctuation characteristics and amplitude fluctuation characteristics) are different between the transmission unit and the reception unit in each wireless unit. The transmission characteristics of the transmitting unit and the receiving unit are different due to the fact that they are physically separate circuits, and there is inherent variation in the characteristics of circuit elements. Variations in the characteristics of the circuit elements are particularly caused by L in the receiver.
In a NA (low noise amplifier), an HPA (high power amplifier) in a transmission unit, or the like, it is caused by a temperature change, a temporal change, or the like.

The difference in transmission characteristics between the receiving unit and the transmitting unit directly affects the error in the array antenna pattern between the time of reception and the time of transmission. For this reason, it is necessary to obtain a transmission characteristic difference between the transmission unit and the reception unit, and perform calibration to compensate for the transmission characteristic difference. For example, Japanese Patent Application Laid-Open No. 11-3
No. 12917 “Array antenna device” includes a calibration method and the like.

This array antenna apparatus comprises a calibration desired signal generating means, a calibration interference signal generating means, a power control means for controlling the power of the calibration interference signal, a calibration desired signal and a power control. Combining means for combining the calibrated interference signal and distribution means for distributing the combined signal to each antenna are provided as additional devices, and are configured to compensate for transmission characteristics of the receiving system.

[0006]

However, according to the above prior art, it is necessary to provide the additional device for measuring the difference in transmission characteristics between the transmitting circuit and the receiving circuit individually in the radio unit in the adaptive array device. There is a problem that the circuit scale increases. In other words, there is a problem that a circuit dedicated to calibration, which is not required for normal communication, must be provided, and the circuit scale increases.

[0007] In view of the above problems, an object of the present invention is to provide an adaptive array device, a calibration method, and a program recording medium for performing calibration without providing a circuit dedicated to calibration.

[0008]

In order to solve the above-mentioned problems, an adaptive array device according to the present invention includes at least three radio units each including a transmitting unit, a receiving unit, and an antenna,
An adaptive array device for measuring a relative phase and amplitude characteristic of a third radio unit with respect to a second radio unit,
First calculating means for receiving a transmission signal from the radio unit by the second and third radio units, and calculating, from the reception signals of the second and third radio units, a response vector indicating a propagation path of the transmission signal; A second calculating unit configured to calculate a weight vector indicating a directivity pattern in which a null is directed to an antenna of the first wireless unit based on the calculated response vector; A control unit for forming a directivity pattern in the radio unit and transmitting the signal, and a second radio unit based on a reception signal level in the first radio unit which has received the transmission signals from the second and third radio units. Measuring means for measuring the relative phase variation and amplitude variation of the three radio units.

Here, the second calculating means calculates a virtual response vector having no correlation with the response vector calculated by the first calculating section, and outputs the beam in the direction indicated by the virtual response vector by the first calculating section. It may be configured to calculate a weight vector that turns null in the direction indicated by the calculated response vector. Here, the control means includes a second and a third control unit.
During the transmission by the radio unit, the phase and amplitude of the transmission signal of the third radio unit are changed, and the measuring unit is configured to receive the transmission signal from the second and third radio units during the change. The amount of phase fluctuation and the amount of amplitude fluctuation may be measured based on the values of the phase and the amplitude when the received signal level at the minimum is obtained.

Here, the control means may further comprise:
Activating the first and second calculation means by relatively replacing the third radio section, and causing the measurement means to measure the relative phase variation and amplitude variation of each radio section with respect to one radio section; The configuration may be such that the phase correction value and the amplitude correction value for each wireless unit other than the one wireless unit are calculated based on each measured phase variation and each amplitude variation.

[0011] The control means may further determine the validity of the sum of the relative phase fluctuations and the product of the amplitude fluctuations measured for each radio unit within a predetermined range. Alternatively, it may be configured to calculate the phase correction value and the amplitude correction value when it is determined to be valid. Further, the calibration method of the present invention is a calibration method for measuring a correction value of a third antenna with respect to a second antenna using first to third antennas, wherein the first signal is transmitted from the first antenna. 1st step, 2nd, 3rd
Based on the received signal of the first signal at each antenna,
A second step of calculating a response vector indicating a propagation path of the first signal; and a third step of calculating a weight vector indicating directivity of pointing null to the first antenna based on the calculated response vector. A fourth step of array-transmitting the second signal from the second and third antennas using the obtained weight vector and causing the first antenna to receive the second signal, and the amplitude and the amplitude of the second signal transmitted from the third antenna A fifth step of changing the phase and measuring a correction value of the third antenna with respect to the second antenna based on the received signal level of the changing first antenna.

Here, the third step is a sub-step of calculating a virtual response vector having no correlation with the response vector calculated in the second step, and the first calculation of the beam in the direction indicated by the virtual response vector. And calculating a weight vector for turning a null in a direction indicated by the response vector calculated by the unit.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A radio base station and a mobile phone according to an embodiment of the present invention will be described in the following order. 1. First embodiment (wireless base station) 1.1. Overview 1.1.1 Schematic configuration 1.1.2 Schematic operation 1.1.3 Capture description 1.2. Configuration of Radio Base Station 1.2.1 Configuration of Signal Processor 1.2.2 Configuration of User Processor 1.2.2.1 First Weight Calculator 1.2.2.2 Response Vector Calculator 1. 2.2.3 Virtual response vector calculator 1.2.2.4 Second weight calculator 1.2.3 Calibration process 1.2.3.1 Δθ measurement process 1.2.3.2 Amp measurement process 1.2.3.3 Verification and correction value calculation processing Second embodiment (wireless base station) 2.1. Overview 2.2. Configuration of wireless base station 2.2.1 Calibration process 3 Third Embodiment (Mobile Phone) 3.1 Configuration of Mobile Phone 3.2 Configuration of Measuring Device 3.3 Calibration Process 4 Other Modifications <1. First embodiment (wireless base station)><1.1.Overview> First, an overview will be given of a case where an adaptive array device according to an embodiment of the present invention is configured as a wireless base station of a mobile communication system. <1.1.1 Schematic Configuration> FIG. 1 is a diagram showing a schematic configuration of a main part of an adaptive array device according to an embodiment of the present invention.

As shown in FIG. 1, the adaptive array device includes radio units 1 to 4 and a DSP (digital signal processor) 50. Although four DSPs 50 are shown for convenience, one DSP may actually be used. This adaptive array device measures a correction value by itself and performs communication using the measured correction value during normal communication. In other words, it also serves as a calibration device.

The radio unit 1 includes an antenna 10 and a transmitting unit 111
(TX1 in the figure), a receiving unit 112 (RX1), and an antenna switch 113 (SW1). The same applies to the wireless units 2 to 4. ΘRX1 and ARX1 in the figure are antenna 1
0, the phase fluctuation amount and the amplitude fluctuation amount caused by the signal passing through the antenna switch 113 and the receiving unit 112, respectively. θTX1 and ATX1 indicate the amount of phase change and the amount of amplitude change caused by the signal passing through the transmission unit 111, the switch 113, and the antenna 10, respectively. θRX2〜
θRX4, ARX2 to ARX4, θTX2 to θTX4, and ATX2 to ATX4 also indicate the same phase variation and amplitude variation in the respective radio units.

.DELTA..theta.12 and Amp12 indicate the relative phase variation and the relative amplitude variation of the radio unit 2 with respect to the radio unit 1, respectively. Δθ23, Δθ34, Δθ41, Amp2
3, Amp34 and Amp41 also show the same relative phase variation and amplitude variation. That is, these are represented by the following equations (1) to (8). (1) Δθ12 = ((θTX1-θRX1)-(θTX2-θRX2)) (2) Δθ23 = ((θTX2-θRX2)-(θTX3-θRX3)) (3) Δθ34 = ((θTX3-θRX3)-(θTX4) -θRX4)) (4) Δθ41 = ((θTX4-θRX4)-(θTX1-θRX1)) (5) Amp12 = ((ATX1 / ARX1) / (ATX2 / ARX2)) (6) Amp23 = ((ATX2 / ARX2 ) / (ATX3 / ARX3)) (7) Amp34 = ((ATX3 / ARX3) / (ATX4 / ARX4)) (8) Amp41 = ((ATX4 / ARX4) / (ATX1 / ARX1)) This adaptive array device To transmit and receive known signals to and from each other in the radio unit 1 to the radio unit 4 and obtain a phase amount and an amplitude amount that match an array antenna pattern between transmission and reception while changing the phase amount and the amplitude amount. , The relative transmission characteristics (phase variation, amplitude variation) shown in the above equations (1) to (8) are detected, and a correction value for compensating the detected phase variation and amplitude variation is calculated. calculate. This correction value is expressed by the following equations (9) to (17). (9) θ_hosei_1 = 0 (10) θ_hosei_2 = Δθ12 (11) θ_hosei_3 = Δθ12 + Δθ23 (12) θ_hosei_4 = Δθ12 + Δθ23 + Δθ34 (13) A_hosei_1 = 1 (14) A_hosei_2 = Amp12 (15) A_hosei_3 = Amp12 * Amp23 (16) A_hosei_4 = Amp12 * Amp23 + Amp34 θ_hosei_x and A_hosei_x are correction values for the transmission signal at the time of transmission of the radio unit x (x is 1 to 4).

The above correction values are relative correction values based on the radio unit 1. The reason why the correction value may be such a relative value is that if the ratio of the phase fluctuation amount and the ratio of the amplitude fluctuation amount of the radio unit at the time of reception are equal at the time of transmission, the weight vector calculated at the time of reception is transmitted at the time of transmission. This is because when used, the same array antenna pattern is obtained at the time of transmission and at the time of reception.

Further, in formulas (9) to (16), the radio unit 1 is used as a reference, but any radio unit may be used as a reference. For example,
Based on the wireless unit 3, the phase correction values are (9 ′) to (1
2 ′), the amplitude correction value is represented by (12 ′) to (16 ′). (9 ') θ_hosei_1 = Δθ34 + Δθ41 (10') θ_hosei_2 = Δθ34 + Δθ41 + Δθ12 (11 ') θ_hosei_3 = 0 (12') θ_hosei_4 = Δθ34 (13 ') A_hosei_1 = Amp34 * Amp41 (14') A_hosei_2 = Amp34 * Amp41 * Amp12 (15 ') A_hosei_3 = 1 (16') A_hosei_4 = Amp34 <1.1.2 Outline operation>
A method for roughly measuring the amplitude fluctuation amount will be described.

FIGS. 2 (a) and 2 (b) are explanatory diagrams showing a schematic operation of the adaptive array apparatus when measuring Δθ34 and Amp34 shown in the equations (3) and (7). In the figure, Ant2 to Ant4 in the figure mean logical wireless units that are associated one-to-one with three wireless units among the physical wireless units 1 to 4. Here, it is assumed that Ant2 to Ant4 correspond to wireless units 2 to 4, respectively.

In FIG. 2A, Ant2 alone transmits a specific signal (in the figure). This specific signal may be a known symbol data sequence. On the other hand, Ant3 and Ant4 are 2
As an adaptive array device of an antenna, an array antenna pattern is formed so as to direct a directional beam to a specific signal transmitted from Ant2, and the signal is received. However,
In this array reception, the DSP 50 calculates not the weight vector that directs the directional beam in the direction of arrival of the received specific signal, but the weight vector that directs the directional null in the direction of arrival of the specific signal ().

Further, as shown in FIG.
Reference numerals 3 and 4 denote specific antennas as array antennas using the weight vectors calculated at the time of receiving the array, as a two-antenna adaptive array device (). The array antenna pattern in this array transmission is, as shown by the solid line in FIG. 4B, if the phase variation and amplitude variation of the transmitting unit in Ant 3 and Ant 4 are equal to those of the receiving unit.
It should have directivity in which nulls are correctly directed in the arrival direction of the specific signal at the time of reception (that is, in the direction of Ant2).

Actually, the phase fluctuation amount and the amplitude fluctuation amount are not always equal between the transmitting unit and the receiving unit, so that the array antenna pattern is shifted as shown by a broken line and a dashed line in FIG. Therefore, the DSP 50 adds the phase compensation amount Δθ to the transmission signal in Ant 4 while gradually changing the phase compensation amount (for example, from -180 degrees to +180 degrees at a time). On the other hand, Ant2 measures the received signal level according to this change (). The phase compensation amount Δθ when the received signal level is minimized is Δθ34 = (θTX3-θRX
3) It is equal to-(θTX4-θRX4)). Therefore, the phase compensation amount Δθ at this time is determined as Δθ34 ().

Further, the DSP 50 gradually changes only the amplitude compensation amount Amp_coef of the transmission signal of the Ant 4 (for example, 0.01 to 0.5 to 2 times). Ant2 measures the received signal level according to this change (). The amplitude compensation amount Amp_coef when the received signal level becomes minimum is equal to the above Amp34 = ((ATX3 / ARX3) / (ATX4 / ARX4)). Therefore, Amp_coef at this time is determined as Amp34 ().

As described above, the adaptive array apparatus can control the relative phase variation Δθ34 and the relative amplitude variation Am
Measure p34. Similarly, Δθ41 and Amp41, θ
12 and Amp12, θ23 and Amp23 are measured. further,
The DSP 50 verifies whether the measured relative phase change amount and amplitude change amount are appropriate, by using the expressions (17) and (18). (17) | Δθ12 + Δθ23 + Δθ34 + Δθ41 | <θthre Here, θthre is, for example, a threshold value of about 1 (degree). The left side of equation (17) is an equation obtained by adding the right sides of equations (1) to (4), and ideally should be 0 (degree).
Actually, a measurement error or a measurement error due to the influence of an extraneous wave or the like may occur. Therefore, it is desirable to perform verification using θthre. (18) A_thre_min <Amp12 * Amp23 * Amp34 * Amp41 <A_thre
_max Here, A_thre_min and A_thre_max are, for example, threshold values of about 0.95 and 1.05, respectively. The product in the middle of equation (18) is an equation obtained by multiplying the right side of equations (5) to (8), and should ideally be 1. Is desirable.

When the expressions (17) and (18) are satisfied, the adaptive array apparatus is expressed by the expressions (9) to (16) (or the expressions (9 ') to (16')) based on these. A correction value is calculated, and the transmission signal is corrected by the DSP 50 during transmission. <1.1.3 Description of Capture> Here, the relative phase change amount and the relative amplitude change amount will be described in a capture manner.

As shown in FIGS. 2 (a) and 2 (b), when the adaptive array apparatus receives an array at Ant3 and Ant4 and performs array transmission based on the weight vector calculated at the time of array reception, the phase at the time of transmission at the time of reception is determined. The amount of change is
In Ant3 (ΔTX3-ΔRX3), in Ant4 (ΔTX4
-ΔRX4). Similarly, the amplitude fluctuation amount at the time of transmission from the time of reception is (ATX3 / ARX3) in Ant3,
In (4), only (ATX4 / ARX4) occurs.

The fact that the phase Δθ is changed little by little with respect to the transmission signal of Ant 4 and the reception level at Ant 2 is minimized means that the phase fluctuation amount at Ant 3 and Ant 4 has been compensated. That is, (ΔTX3-ΔRX3) = (ΔTX4
−ΔRX4) + Δθ34, and therefore Δθ34 = ((ΔTX3-
ΔRX3)-(ΔTX4-ΔRX4)).

Similarly, the fact that the reception level has become minimum means that the phase fluctuation amount in Ant3 and Ant4 has been compensated. That is, (ATX3 / ARX3) = (ATX4 / ARX4) *
Amp34, therefore Amp34 = ((ATX3 / ARX3) * (ATX4
/ ARX4)). <1.2 Configuration of Radio Base Station> FIG. 3 is a block diagram showing the overall configuration of the radio base station in the first embodiment.
In the figure, a radio base station includes a baseband unit 70, a modem unit 60, a signal processing unit 50, and a front end unit 1.
1, 21, 31, 41, antennas 10 to 40, control unit 8
0. This wireless base station is an adaptive array device that wirelessly connects a mobile communication device by forming an array antenna pattern by weighting transmission and reception signals for each antenna using a plurality of antennas, and is defined by the PHS standard. Bidirectional time division multiplexing (TDMA / TD)
D: Time Division Multiple Access / Time Division Du
A plex) system is installed as a wireless base station for connecting a mobile communication device (PHS telephone).

The baseband unit 70 transmits a plurality of signals (baseband signals indicating voice or data) between a plurality of lines connected via a telephone switching network and the modem unit 60 to T.
TDMA / TDD processing for multiplexing and demultiplexing so as to conform to the DMA / TDD frame is performed. Here, TDMA / T
The DD frame has a period of 5 ms and is composed of four transmission time slots and four reception time slots that are equally divided into eight.

The modem unit 60 modulates a signal input from the baseband unit 70 and demodulates a signal input from the signal processing unit 50. The modulation and demodulation method is π / 4 shift QPSK. The signal processing unit 50 (the DSP 50
The DSP 50) calculates a weight vector by executing a program. In particular, in the calibration processing, a correction value for compensating the transmission characteristics between the time of reception and the time of transmission of the wireless units 1 to 4 is calculated.

The front end units 11, 21, 3
1 and 41 convert each signal weighted by the DSP 50 into an RF signal at the time of array transmission, and
40, and when receiving the array, the antennas 10-4
0 to DS in the baseband
Output to P50. Hereinafter, a set of the antenna 10 and the front end unit 11 is referred to as a wireless unit 1. Similarly, the other three sets of the antenna and the front end unit are called radio units 2 to 4, respectively.

As shown in FIGS. 2A and 2B, the radio units 1 to 4 perform the DSP 50 in the calibration process.
, Each of which transmits a specific signal from the
A specific signal is array-received and array-transmitted by a set of two radio units. The control unit 80 controls the entire radio base station such as switching between transmission and reception of each radio unit. <1.2.1 Configuration of Signal Processing Unit> FIG.
FIG. 3 is a block diagram showing a detailed configuration of FIG. In the figure, DS
FIG. 3 is a block diagram illustrating functions realized by executing a program by P50. In FIG.
0 indicates user processing units 51a to 51d, adders 551 to 5
54, switches 561 to 564 for switching between transmission and reception, a correction value holding unit 570, and correction units 571 to 574.

The user processing units 51a to 51d are TDMA
It is provided corresponding to up to four user signals spatially multiplexed in each time slot in the / TDD frame.
Each user processing unit normally controls array reception and array transmission using all four wireless units (other than the calibration process). That is, at the time of reception, a weight vector is calculated from the received signals from the four radio units 1 to 4, and the received signals input from the radio units 1 to 4 via the switches 561 to 564 using the weight vectors. A user signal is extracted by synthesizing X1 to X4, and at the time of transmission, user signals S1 to S4 weighted using a weight vector calculated in the immediately preceding reception time slot.
S4 is output to each of the radio units 1 to 4.

On the other hand, in the calibration process, each user processing unit may control the array reception and array transmission of two antennas, and may perform the control of transmitting and receiving a specific signal independently instead of the array transmission and reception. DSP50
Performs a series of processes shown in FIGS. 2A and 2B by combining these cases, and performs a relative phase variation (Δθ
34, Δθ41, Δθ12, Δθ23), amplitude fluctuation (Amp34, A
mp41, Amp12, and Amp23), and the correction value (θ
_hosei_1 to θ_hosei_4, A_hosei_1 to A_hosei_4).

The adder 551 combines the weighted user transmission signals for the radio unit 1. However, FIG.
As shown in FIG.
Array transmission by two antennas as shown in FIG.
In this case, the transmission signal (specific signal) from one of the user processing units is output as it is without adding it to the other signals. The adders 552 to 554 are also the adders 551.
Except that they correspond to the radio units 2 to 4, respectively.

The correction value holding unit 570 stores the correction values (θ_hosei_1 to θ_hos_1) calculated in the calibration process.
ei_4, A_hosei_1 to A_hosei_4). Correction unit 57
1 indicates that θ_hosei_1 and A_hosei_ among the correction values held in the correction value holding unit 570 except for the calibration process.
According to 1, the transmission signal from the adder 551 is corrected and output to the wireless unit 1 via the switch 561. In the calibration process, the transmission signal from the adder 551 is output to the wireless unit 1 via the switch 561 as it is. I do. However, when the relative phase variation and the amplitude variation of the wireless unit 1 are to be measured in the calibration process, the phase compensation amount Δθ and the amplitude compensation amount Amp are given to the transmission signal while being gradually changed. .

The correction units 572 to 574 are the same except that the corresponding radio unit and the correction value stored in the correction value storage unit 570 are different. <1.2.2 Configuration of User Processing Unit> FIG. 5 is a block diagram showing a detailed configuration of the user processing unit 51a. Since the user processing units 51b to 51d have the same configuration,
Here, the user processing unit 51a will be described as a representative.

As shown in the figure, the user processing unit 51a includes a weight calculation unit 53, an adder 54, a memory 55, a switch 56, a switch 57, multipliers 521 to 524, and a multiplier 5.
81 to 584. The weight calculator 53 includes a first weight calculator 531, a response vector calculator 532, a virtual response vector generator 533, and a second weight calculator 534.
The weight vector for the four-array antenna is calculated by the first weight calculation unit 531 except for the calibration process, and the response vector calculation unit 532, the virtual response vector generation unit 533, and the second weight calculation unit are calculated in the calibration process. According to 534, a weight vector for two array antennas for turning null in the arrival direction of the specific signal is calculated (FIG. 2).

The circuit portion including the adder 54 and the multipliers 521 to 524 weights and combines the received signal with the weight vector calculated by the weight calculating section 53 at the time of reception, and switches the signal as a desired signal of a specific user. Output via. That is, except for the calibration processing, the reception signals X1 to X4
Are weighted and combined, and in the calibration process, weighting and combining are performed on signals received from any two wireless units.

Multipliers 581 to 584 weight transmission signals with the weight vectors calculated by weight calculation section 53 during transmission. That is, except for the calibration processing, the transmission signal input via the switches 56 and 57 is weighted for each of the four wireless units, and any two
Weighting is performed on a specific signal for one radio unit.

The reference signal generation section 55 generates the waveform data of the symbol sequence representing the reference signal required for calculating the weight vector in the weight calculation section 53. Specifically, the reference signal generation unit 55 demodulates (determines) the waveform data of the fixed symbol in the fixed symbol section in the reception time slot and the synthesized signal obtained from the adder 54 in the data section except for the calibration processing. ) Is generated and output to the first weight calculator 532. In the calibration process, the reference signal generation unit 55 generates waveform data of a symbol representing a specific signal, and outputs the waveform data to the second weight calculation unit 534. Here, the specific signal is
For example, a known symbol data sequence such as a PN (Pseudo random Noise) code may be used.
When the user processing unit controls the array reception of two antennas as shown in FIG. 4, the weight calculation unit 53 reads the reference signal (training signal) as a reference signal (training signal). When controlling transmission, the signal is read out as a transmission signal and supplied to multipliers 581 to 584 via switch 57. However, the outputs of the multipliers 581 to 584 are transmitted to only one corresponding to the radio unit that transmits independently. <1.2.2.1 First Weight Calculation Unit> The first weight calculation unit 531 transmits signals from the radio units 1 to 4 every symbol section of the reception time slot except for calibration processing (normal transmission and reception). A weight vector is calculated so that an error between the result of weighting and adding each of the received signals X1 to X4 and a reference signal generated by the memory 55 is minimized.

More specifically, the first weight calculator 53
1 adjusts the values of W1 (t-1) to W4 (t-1) to minimize the error e (t) in the following equation (19), and adjusts W1 (t-1)
To W4 (t-1) are weighting factors W1 (t) to W4 (t) of the symbol at time t.
And (19) e (t) = d (t)-(W1 (t-1) * X1 (t) + W2 (t-1) * X2 (t) + W3 (t-1)
* X3 (t) + W4 (t-1) * X4 (t)) where t is the timing in symbol units, d (t) is the symbol data of the reference signal, and W1 (t-1) to W4 (t- 1) is a weight vector for each antenna calculated for the previous symbol, or a weight vector calculated in the previous reception time slot, and X1 (t) to X4 (t) are reception signals of the radio units 10 to 40. It is. The weight vector is adjusted for each symbol as described above. In the symbol at the beginning of the section of the reference signal in the reception time slot, the error e (t) is large at the end of the section of the reference signal even if the error e (t) is large. ) Converges to a minimum (or converges to 0).

As a method of minimizing this error, for example, as described in “Adaptive signal processing by array antenna” (Nobuyoshi Kikuma, published by Science and Technology Publishing Co., Ltd. on September 20, 1998) Minimum Mean Square Error: Least Square Error Method, MSN (Maximum Signal-to-Noise Ratio: Maximum SNR Method), DCMP (Directionally Constrained Minimi
zation of Power: Direction-constrained output power minimization method) is known.

In the radio base station, LMS (Least Mean Square) is used as an algorithm for optimizing the MMSE.
And a sequential algorithm such as RLS (Recursive Least Square). According to this, the weight vector can be successively converged in the reception time slot for each symbol. <1.2.2.2 Response Vector Calculation Unit> The response vector calculation unit 532 calculates a response vector indicating the arrival direction of the specific signal based on the reception signals of the specific signal from the two wireless units in the calibration process. I do.

Specifically, the response vector is calculated as follows. However, for convenience of explanation, a case where a signal transmitted from Ant1 is multiplexed in addition to a specific signal transmitted from Ant2 will be described below. An
Each signal transmitted from t1 and Ant2 is multiplexed,
Signals received by Ant3 and Ant4 are X1 (t), X2 (t), Ant1, An
Sa (t), Sb (t), Ant3, Ant4
Let n1 (t) and n2 (t) be the noise components included in each received signal. X1 (t) = ha1 · Sa (t) + hb1 · Sb (t) + n1 (t) X2 (t) = ha2 · Sa (t) + hb2 · Sb (t) + n2 (t) Here, ha1 represents a propagation path from Ant1 to Ant3, and is a parameter expressed in complex notation including amplitude and phase. The same applies to ha2, hb1, hb2.

The response vectors Ha and Hb of the signals Sa (t) and Sb (t) are represented by the following equations. Ha = [ha1, ha2] T (T is transposed) Hb = [hb1, hb2] T where Sa (t), Sb (t), and n1 (t), n2 (t) Assume no correlation. Also, the reference signal generator 5
5 generates a signal Sb (t) as a reference signal rb (t), that is, rb
(t) = Sa (t).

The response vector calculator 532 calculates the received signal X2
By multiplying (t) by a reference signal rb '(t) (' is a complex conjugate) and taking an ensemble average, hb1 is calculated based on the following equation. E [X1 (t) • rb '(t)] = E [X1 (t) • Sb' (t)] = E [ha1 • Sa (t) • Sb '(t)] + E [hb1 • Sb (t ) .Sb '(t)] + E [n1 (t) .Sb' (t)] = ha1.E [Sa (t) .Sb '(t)] + hb1.E [Sb (t) .Sb' (t )] + E [n1 (t) · Sb '(t)] ≒ hb1 Here, the ensemble average between identical signals is 1 and the ensemble average between uncorrelated signals is approximately 0, so E [S
a (t) Sa '(t)] = 1, E [Sb (t) Sa' (t)] ≒ 0, E [n3 (t) Sa '
(t)] ≒ 0.

Similarly, the response vector calculation unit 532
2 is calculated by multiplying the received signal X2 (t) by the reference signal rb (t) 'and taking an ensemble average. This allows An
The response vector Hb of the specific signal Sb (t) from t2 is calculated. Similarly, the response vector Ha of the signal Sa (t) from Ant1
Can also be calculated. However, in this embodiment, Ant1
Is not present (Sa (t) = 0), the response vector calculation unit 532 calculates only the response vector Hb of the specific signal Sb (t) from Ant2.

As described above, since the response vector is calculated by the ensemble average (time average), the longer the period, the more the influence of noise and fading is removed, and the optimum response vector can be obtained. In this regard, the response vector calculation unit 532 may calculate the response vector by averaging the ensemble over a number of symbol periods of the specific signal, for example, although the ensemble average may be one symbol period. <1.2.2.3 Virtual Response Vector Calculation Unit> The virtual response vector generation unit 533 generates a virtual response vector with low correlation (for example, orthogonal) to the response vector calculated by the response vector calculation unit 532. I do. This virtual response vector corresponds to the arrival direction of the specific signal (Ant2 in FIG. 2).
) Means a direction that is orthogonal, and is used by the second weight calculator 534 as a virtual desired wave arrival direction.

The virtual response vector generator 533 calculates a virtual response vector that satisfies the following equation. Specifically, the virtual response vector is calculated as follows. Let the virtual response vector be Hv. Hv = [hv1, hv2] T The correlation value Cvb between the virtual response vector Hv and the response vector Hb is given by the following equation. Cvb = (hv1 · hb1 ′ + hv2 · hb2 ′) / (| Hv | · | Hb |) The correlation value Cvb indicates that the closer the value is to 1, the closer the direction represented by the two response vectors is. If 0, 2
This means that the directions represented by two response vectors are orthogonal.

The virtual response vector generation unit 533 calculates the above Cv
A virtual response vector Hv such that b = 0 is generated. As described above, since the virtual response vector is calculated from only the above-described optimal response vector, the virtual response vector generation unit 53
No. 3 can obtain an optimal response vector with high orthogonality. <1.2.2.4 Second Weight Calculation Unit> The second weight calculation unit 534 sets the direction indicated by the response vector calculated by the response vector calculation unit 532 as the direction of arrival of the interference wave, and sets the virtual response vector generation unit 533. The weight vector is calculated with the direction indicated by the virtual response vector calculated by the above as the direction of the virtual desired wave. That is, this weight vector is directed to Ant2 in FIG.
This is a weight vector for t4.

Specifically, this weight vector is calculated as follows. Weight vector W for two antennas
Is represented by the following equation. The following relationship holds between the W = [w1, w2] T weight vector, the response vector Hb, and the virtual response vector Hv. w′1 · hb1 + w′2 · hb2 = 0 w′1 · hv1 + w′2 · hv2 = 1 The second weight calculation unit 534 should easily calculate the weight vector W so as to satisfy these two equations. Can be. As described above, since the response vector and the virtual response vector are free from the effects of noise and fading, optimal values can be obtained for the weight vector calculated from only the response vector and the virtual response vector.

Further, a more optimal weight vector Wopt
In order to obtain MMSE, SMI (Sa
(mple Matrix Inversion: direct solution using sample values)
It is desirable to calculate the optimal weight by SMI has a large computational load such as inverse matrix calculation, but the calibration process does not require real-time communication, so the computational load does not matter.

The second weight calculator 534 calculates an optimum weight vector Wopt as follows by SMI, which is an optimized method for obtaining a weight vector called a Wiener solution. The optimum weight vector is obtained by the following equation. Wopt = inverse (CM) · CV Here, CM is a correlation matrix, and CV is a correlation vector of a desired signal, and is calculated as follows using the response vector Hv of the desired signal and the response vector Hb of the interference signal. CM = CM11, CM12 CM21, CM22 This correlation matrix CM is a 2-by-2 complex matrix.

The correlation vector is represented by the following equation. CV = [hv'1, hv'2] T CM11, CM22 are represented by the following equations. CM11 = hv'1, hv1 + hb'1, hb1 + α CM22 = hv'2, hv2 + hb'2, hb2 + α
And an appropriate positive number (for example, 512/2048) is substituted for α.

CM12 and CM21 are represented by the following equations. CM12 = hv′1 · hv2 + hb′1 · hb2 CM22 = CM′12 Here, the inverse matrix of the correlation matrix CM is expressed by the following equation. Then, each component of the optimal weight vector Wopt is calculated by the following equation. w′1 = CMI11 · hv′1 + CMI12 · hv′2 w′1 = CMI21 · hv′1 + CMI22 · hv′2 In this way, the second weight calculator 534 calculates the optimum weight vector.

The user processing units 51a to 51d may have the same configuration, but for convenience of explanation, it is assumed that each user processing unit performs fixed processing in the calibration processing. FIG. 6 shows a list of processing contents of each user processing unit in the calibration processing. Ant1 in the figure
To Ant4 means a logical wireless unit that is associated with the physical wireless units 1 to 4 on a one-to-one basis. This correspondence is assumed to be stored in advance in the DSP 50 as a table. FIG. 7 shows an example of this table. Although there may be many such correspondences, in the present embodiment, there are at least four correspondences as in cases 1 to 4 shown in FIG.

In FIG. 6, in the first half of the calibration process (that is, in the case as shown in FIG.
With the control of 0, the three radio units use the same frequency,
Ant2 is transmitting, Ant3 and Ant4 are receiving. In this case, as shown in the “first half” column of FIG.
b is an idle state. The user processing unit 51b causes Ant2 to transmit a specific signal alone, that is, generates a specific signal and supplies it to Ant2. The user processing unit 51c includes Ant3 and
It controls the array reception of the two antennas for each received signal from Ant4, that is, calculates the response vector of the specific signal, calculates the virtual response vector, and calculates the weight vector for the two antennas pointing null to Ant2.

In the latter half of the calibration process (that is, in the case as shown in FIG. 2B), under the control of the control unit 80, the three radio units use the same frequency, and Ant2 receives, Ant
t3 and Ant4 perform array transmission. In this case, as shown in the “second half” column of the figure, the user processing unit 51 c
The specific signal from Ant4 is controlled as two array antennas, that is, the specific signal is weighted using the calculated weight vector and supplied to Ant3 and Ant4. At this time, the user processing unit 51c changes the phase compensation amount Δθ as shown in FIG.
The amplitude compensation amount Amp_coef is changed as shown in FIG. The user processing unit 51a acquires a single received signal at Ant1. Each time the phase compensation amount Δθ and the amplitude compensation amount Amp_coef change, the user processing unit 51b acquires a single received signal from Ant2 and the received signal level from Ant2. <1.2.3 Calibration Processing> FIG. 8 is a flowchart showing more detailed contents of the calibration processing. N in the figure represents cases 1 to 4 shown in FIG.
It is a variable for counting up to.

As shown in FIG. 8, the DSP 50
Is initialized (N = 1) (step 81), and referring to case N in the table shown in FIG.
, Ant2 as a logical wireless unit is determined (step 82), and Ant3 and Ant4 are similarly determined (step 83). As shown in FIGS. 2A and 2B and FIG.
Ant2 is for single signal transmission / reception, and Ant3 and Ant4 are for array reception and array transmission.

The DSP 50 causes Ant2 to transmit a specific signal (step 84), and at the same time starts receiving Ant3 and Ant4 as an adaptive antenna with two antennas (step 85). A reception response vector indicating the arrival direction of the signal is estimated (calculated) (step 86). This response vector is calculated by the response vector calculator 532 as described above. Further, the DSP 50 generates a virtual response vector orthogonal to the calculated response vector (Step 87). This virtual response vector is generated by the virtual response vector generation unit 533 as described above.

The DSP 50 then directs a null in the direction of Ant2 and a beam in the direction of the virtual response vector.
A weight vector for Ant3 and Ant4 is calculated (step 88). The calculation of the weight vector is performed by the second weight calculator 534 as described above. At this time, as shown in FIG. 6, the specific signal to Ant2 is supplied from the user processing unit 51b, and a response vector, a virtual response vector, and a weight vector based on the received signals from Ant3 and Ant4 are output by the user processing unit 51c. Is calculated.

Using a weight vector pointing null in the direction of Ant2, the DSP 50 transmits a specific signal in an array using Ant3 and Ant4 as an adaptive array of two antennas, and switches Ant2 to single reception (step 89). At this time, weighting by the weight vector is performed by the user processing unit 51c. In this state, the DSP 50
The relative movement fluctuation characteristics and the amplitude fluctuation characteristics of Ant4 with respect to Ant3 are measured (step 90).

In this way, the case 1 shown in FIG. 7, that is, the relative movement variation characteristic θ34 and the amplitude variation characteristic Amp34 of the radio unit 4 with respect to the physical radio unit 3 are measured. Further, the DSP 50 performs the same processing for the cases 2 to 4 in FIG. 7 by the loop processing in steps 92 and 93, thereby the logical wireless unit selected from the physical wireless units 1 to 4. While changing the combination of Ant1 to Ant4 as above, Δθ41 and Amp41 of the wireless unit 1 with respect to the wireless unit 4 in the second loop, and Δθ12 and Am of the wireless unit 2 with respect to the wireless unit 1 in the third loop.
In the fourth loop, the radio unit 3
Δ23 and Amp23 are measured.

After that, the DSP 50 determines the validity of each of the measured phase variation and amplitude variation as described in (17) and (18) above.
The verification is performed according to the formulas, and if valid, the phase correction value and the amplitude correction value for each of the other three wireless units are calculated based on one of the wireless units based on them, and the correction values corresponding to the three wireless units are calculated. This is set in the holding unit 570, and the phase correction value and the amplitude correction value of one reference wireless unit are set to 0 and 1, respectively. <1.2.3.1 Δθ Measurement Processing> FIG. 9 is a flowchart showing the measurement processing of the phase variation Δθ34 in step 90 of FIG.

Δθ34 in FIG.
, Min_RSSI is a variable for obtaining the minimum value of the received signal level RSSI of the specific signal by Ant2, and min_Δθ34 is a variable for obtaining Δθ34 corresponding to min_RSSI. Also, during the processing of the figure, the phase and amplitude of the transmission signal of Ant3 and the amplitude of the transmission signal of Ant4 are
It is not changed except by weighting with a weight vector.

In the array transmission of the specific signal by Ant3 and Ant4, the DSP 50 sets the initial value Δθ34 to 0 (step 901), and at that time, sets the value of the received signal level RSSI of the specific signal by Ant2 to min_RSSI and the value of Δθ34 to min_Δθ34.
(Step 902). Next, the DSP 50 calculates Δ
θ34 is changed by a small amount (in the figure, once) (step 90
3), the received signal level RSSI at Ant2 at that time is min_RS
If smaller than SI, min_RSSI and min_Δθ34
And the value of Δθ34 (steps 905 and 90).
6). The DSP 50 repeatedly executes this processing to obtain Δθ34
Has reached the upper limit value (360 degrees in the figure) (step 904).

In this way, the DSP 50 measures the phase compensation amount min_Δθ34 when the reception signal level measured at Ant2 is the minimum. This final min_Δθ34 is An
Phase variation Δθ34 of Ant4 relative to t3 = (θTX3-
Measure θRX3)-(θTX4-θRX4)). In FIG. 9, the amount of phase change is changed from 0 degree to 360 degrees, but it is -18
It may be 0 degrees to 180 degrees. In step 903, the amount of change is set to 1 degree, but may be set to 0.05 degree or other values. <1.2.3.2 Amp Measurement Processing> FIG. 10 is a flowchart showing the measurement processing of the amplitude fluctuation Amp34 in step 91 of FIG.

In the figure, Amp 34 is Ant4
Min_RSSI is a variable for obtaining the minimum value of the received signal level RSSI of the specific signal by Ant2, and min_AMP34 is a variable for obtaining Amp34 corresponding to min_RSSI. AMP_min, AMP_max, AMP_step is A
A lower limit (for example, 0.5), an upper limit (for example, 1.5), and a step of change (for example, 0.01) of mp34 are shown. Also, during the processing in the figure, it is assumed that the phase and amplitude of the transmission signal of Ant3 and the phase of the transmission signal of Ant4 are not changed except for weighting with the weight vector.

In the array transmission of the specific signal by Ant3 and Ant4, the DSP 50 sets the initial value Amp34 to AMP_min (for example, 0.5) (step 911), and at that time, sets the value of the reception signal level RSSI of the specific signal by Ant2 to min_RSSI, Amp3
The value of 4 is set to min_AMP34 (step 912). Next, the DSP 50 changes Amp34 by a small amount AMP_step (step 913). If the received signal level RSSI at Ant2 at that time is smaller than min_RSSI, min_RSSI and min_RSSI are changed.
Amp34 is updated to the value of the RSSI and the value of Amp34 (steps 915 and 916). The DSP 50 repeatedly executes this processing, and ends when the Amp 34 reaches the upper limit value AMP_max (step 904).

In this way, the DSP 50 measures the amplitude compensation amount min_Amp34 when the reception signal level measured at Ant2 is the minimum. This final min_Amp34 is Ant3
Amp4 = Amplitude variation relative to Ant4 relative to (ATX3 / ARX
3) / (ATX4 / ARX4)). <1.2.3.3 Verification and Correction Value Calculation Processing> FIG.
9 is a flowchart showing a process of verifying Δθ and Amp and calculating a correction value in step 91 of FIG.

As shown in the figure, the DSP 50 controls the relative phase variation (Δ
θ34, Δθ41, Δθ12, Δθ23) and the amplitude variation (Am
It is determined whether p34, Amp41, Amp12, and Amp23) are appropriate (steps 98 and 99). This determination is based on whether both of the equations (17) and (18) described above are satisfied. If any of the conditions is not satisfied, the calibration process may be terminated, the specific signal, the frequency, or the like may be changed and the process may be restarted from the beginning.

When both equations (17) and (18) are satisfied, the DSP 50 determines the phase correction values θ_hosei_1 to θ_hosei_
4. Calculate the amplitude correction values A_hosei_1 to A_hosei_4 according to the equations (9 ′) to (16 ′) described above (step 10).
0, 101). The calculated correction value is stored in the correction value holding unit 57.
It is written to 0 and is used for correcting the transmission signal of each radio unit at the time of normal array transmission other than calibration.

As described above, according to the adaptive array apparatus of the present embodiment, one radio section selected from a plurality of radio sections and the other two radio sections have
Since the array transmission is performed and the transmission characteristics of the radio unit selected based on the received signal are measured, the relative transmission characteristics of each radio unit can be calculated without providing an additional device. <2. Second Embodiment (Wireless Base Station)><2.1.Overview> In the present embodiment, a description will be given of a wireless base station that switches a method of calculating a weight vector for turning null to Ant2 in the calibration process according to the situation. That is, the radio base station in the present embodiment is:
When calculating a weight vector for turning null to Ant2 in the same manner as in the first embodiment (hereinafter referred to as a first mode), Ant
1. When calculating a weight vector for pointing null to Ant2 based on different signals transmitted from Ant2 (hereinafter referred to as second
Mode).

FIGS. 12A and 12B are explanatory diagrams showing an outline of the calibration processing in the radio base station of the present embodiment. In FIG. 2A, Ant1 alone transmits a desired signal and Ant2 independently transmits an interference signal on the same frequency (in the figure). The desired signal and the interference signal represent different known data strings.

On the other hand, Ant3 and Ant4 form an array antenna pattern in which a directional beam is directed to Ant1 and a directional null is directed to Ant2 as a two-antenna adaptive array device, and receive a desired signal. Yes (). That is, the radio base station calculates a weight vector for separating a desired signal from a received wave in which a desired signal wave and an interference signal wave are multiplexed. As a result, a weight vector for turning null to Ant2 is calculated. The process up to this point is the weight vector calculation in the second mode.

Further, the radio base station determines the validity of the weight vector calculated in the second mode, that is, whether the correlation between the response vector of the desired signal and the response vector of the interference signal is lower than a threshold value. Determine whether or not. This determination is made for the following two reasons (a) and (b).
(A) Depending on the installation environment of the radio base station, the desired signal from Ant1 and the interference signal from Ant2 are, for example, superimposed on Ant3 and Ant4, a reflected wave and a direct wave from a building or wall near the radio base station, Apparently they can come from almost the same direction. In this case, the accuracy of the weight vector deteriorates, and the null directed to Ant2 becomes shallow, resulting in the result.
Calibration accuracy deteriorates. (B) already mentioned
When the weight vector is calculated by a sequential algorithm such as RLS or LMS, the directivity pattern changes each time the weight vector converges, so that the measurement value is likely to change each time the calibration process is performed.

If the correlation determination shows that the correlation between the response vector of the desired signal and the response vector of the interference signal is higher than the threshold, the radio base station calculates the weight vector in the first mode. Switch to. FIG.
In (b), Ant2 receives the interference signal, and
Transmits an array of desired signals using a weight vector calculated at the time of array reception as an adaptive array device having two antennas (). The subsequent measurement processing of Δθ34 and Amp34 is the same as in the first embodiment. <2.2 Configuration of Radio Base Station> FIG. 13 is a block diagram showing the overall configuration of the radio base station in the second embodiment. This figure is different from FIG. 3 of the first embodiment in that the DSP 50
In that a signal processing unit 500 (hereinafter, referred to as DSP 500) is provided instead of the above. Hereinafter, description of the same components as those in the first embodiment will be omitted, and different points will be mainly described.

The DSP 500 is the DSP 5 of the first embodiment.
0, as shown in FIG. 14, the user processing unit 51a
The difference is that user processing units 551a to 551d are provided in place of the processing units 551a to 551d. FIG. 15 is a block diagram illustrating a detailed configuration of the user processing unit 551a. User processing unit 55
Since the configuration is the same for 1b to 551d, the user processing unit 551a will be described as a representative here.

The user processing section 551a is different from the user processing section 51a in FIG. 5 in that a weight calculation section 553 is provided instead of the weight calculation section 53 and a reference signal generation section 550 is provided instead of the reference signal generation section 55. different. The weight calculation unit 553 includes the weight determination unit 53 with the correlation determination unit 53
The response vector calculation unit 532 calculates respective response vectors of the desired signal from Ant 1 and the interference signal from Ant 2 shown in FIG. 12, and determines the correlation between the two calculated response vectors. The correlation determination unit 535 performs the determination, and switches from the second mode to the first mode according to the determination result.

The correlation determining section 535 calculates a correlation value between the response vector of the desired signal and the response vector of the interference signal calculated by the response vector calculation section 532, and the correlation value is determined by a threshold value (for example, 0.7). ) It is determined whether it is smaller than. The reference signal generation unit 550 generates, in addition to the function of the reference signal generation unit 55, waveform data of a symbol sequence representing a desired signal and waveform data of a symbol sequence representing an interference signal in the second mode. The desired signal and the interference signal may be a known symbol data string such as a PN (Pseudo random Noise) code, for example, and are desirably orthogonal to each other. This is because if they are orthogonal to each other, the weight vectors can be converged more quickly and can be accurately calculated. When the same PN code or the same fixed symbol is used, the timing (for example, 0.5 symbol time) may be shifted.

The user processing units 551a to 551d may have the same configuration, but for convenience of explanation, each user processing unit in the calibration process is shown in FIG. 6 in the first mode and shown in FIG. 16 in the second mode. As described above, fixed processing is performed. Also, the logical radio parts Ant1 ~
The correspondence between Ant4 and the physical radio units 1 to 4 is as follows:
It is assumed that the cases correspond to cases 1 to 4 shown in FIG. <2.2.1 Calibration Processing> FIG. 18 is a flowchart showing more detailed contents of the calibration processing. This figure is different from FIG.
The difference is that steps 800 to 807 are used instead of 88. Steps 800 to 803 are the second mode weight vector calculation processing, and step 807 is the first mode weight vector calculation processing. Step 8
07 is the same processing as steps 82 to 88 in FIG.

Referring to FIG. 18, the DSP 500
, Ant2 and Ant2 as logical wireless units from among the physical wireless units 1 to 4 with reference to case N in the table shown in FIG.
Is determined (step 800), and Ant3 and Ant4 are similarly determined (step 801). As shown in FIGS. 12A and 12B and FIG. 6, Ant1 is for single transmission of a desired signal,
t2 is for single transmission and reception of interference signals, and Ant3 and Ant4 are for array reception and array transmission.

Further, DSP 500 causes Ant1 to transmit a desired signal and Ant2 to transmit an interference signal, respectively (step 802). Simultaneously, Ant3 and Ant4 are used as an adaptive array device having two antennas, and an array antenna is provided for the desired signal from Ant1. Formation of pattern, that is, DSP50
0 is from a received wave in which the desired signal and the interference signal are multiplexed,
A response vector indicating the arrival direction of the desired signal and a response vector indicating the arrival direction of the interference signal are calculated (step 8).
03).

Next, the DSP 500 compares the response vector of the desired signal and the response vector of the interference signal calculated by the response vector calculation section 532 with the correlation determination section 5.
At 35, the correlation is determined (step 80).
4). If the correlation determining unit 535 determines that the correlation between the two response vectors is smaller than the threshold value (step 805: yes), the first weight calculating unit 531 directs the directional beam to the desired signal and causes interference. A weight vector for directing nulls of directivity to the signal is calculated. Thus, the weight vector in the second mode is calculated.

On the other hand, when the correlation determining unit 535 determines that the correlation between the two response vectors is equal to or larger than the threshold value (step 805: no), step 8 shown in FIG.
A weight vector in the first mode is calculated in the same manner as in steps 4 to 88 (step 807). Subsequent processing is the same as in the first embodiment, and a description thereof will not be repeated.

As described above, according to the radio base station of the present embodiment, the second mode and the first mode are selectively used in the calibration process according to the situation, so that the optimum mode is set according to the installation environment of the radio base station. A correction value can be obtained. <3. Third Embodiment (Mobile Phone)> In the present embodiment,
A case where the calibration processing in the second embodiment is applied to a mobile phone will be described.

The mobile phone according to the present embodiment is an adaptive array device that transmits and receives an array by forming an array antenna pattern using two antennas, and cannot perform the above-described calibration processing by itself.
In such a mobile phone, the correction value is measured in cooperation with another measuring device. Further, the mobile phone is configured to hold the measured correction value and correct only the transmission signal of an antenna other than the reference antenna with the correction value.

Hereinafter, the configuration in the case where the adaptive array device of the present invention is a mobile phone of a mobile communication network will be described first, and then the above-mentioned measuring device will be described. <3.1 Configuration of Mobile Phone> FIG. 19 is a block diagram showing a configuration of a main part of a mobile phone according to the third embodiment. As shown in FIG.
0, changeover switch 213, transmission circuit 211, reception circuit 2
12, a radio unit (hereinafter, referred to as a radio unit A), a radio unit including an antenna 220, a changeover switch 223, a transmission circuit 221, and a reception circuit 222 (hereinafter, referred to as a radio unit B).
This is an adaptive array device that includes a DSP 260 (dashed frame in the figure) and an external I / F 250 and forms an array antenna pattern with two antennas and transmits and receives.

The two antennas 210 and 220 may be rod-shaped rod antennas, planar pattern antennas, helical antennas at the rod ends, chip antennas (antennas mounted as chip parts on a substrate), and the like. Here, the antenna 210 is a rod antenna, and the antenna 220 is a chip antenna. D shown by a broken line frame
The SP 260 actually operates according to a program, but the operation is illustrated in FIG. The DSP 260 includes multipliers 214, 224, 21
5, 225, adder 230, demodulation circuit 231, remodulation circuit 232, memory 233, counter 234, switch 2
35, a weight calculation unit 236, a memory 237, a weight control unit 238, a correction control unit 239, a phase shifter 240, an amplifier 241, and a modulation circuit 242.

Multipliers 214 and 224 weight the reception signals input from reception circuits 212 and 222 by multiplying them by weight vectors W 1 and W 2 from weight calculation section 236. The multipliers 215 and 225
Each of the transmission signals input from the modulation circuit 242 is weighted by multiplying it by weight vectors W1 and W2 from the weight control unit 238, and is output to the transmission circuit 211 and the phase shifter 240.

The adder 230 adds the received signals weighted by the multipliers 214 and 224. Demodulation circuit 23
1 demodulates the received signal after addition by the adder 230. The demodulation result is output as a received bit string. The remodulation circuit 232 remodulates the received bit string input from the demodulation circuit 231 into symbol data (symbol waveform data).

The memory 233 holds a reference signal table. The reference signal table represents symbol data (symbol waveform data) representing a reference signal used in a process other than the calibration process (normal reception from the radio base station), a desired signal used in the calibration process, and an interference signal. The symbol data is stored. The reference signal and the desired signal are the same as those described in the wireless base station.

In normal reception, the counter 234 indicates the number of symbols (0 in PHS) in synchronization with the symbol timing from the first symbol to the last symbol in the reception time slot.
To 120). This count value is
It is used to distinguish between a fixed bit pattern symbol period and a non-symbol period. In normal reception, the symbol period from the third symbol to the sixteenth symbol is SS,
This corresponds to the period of the fixed bit pattern of PR and UW.

When the count value of the counter 234 indicates a symbol period of a fixed bit pattern in normal reception, the switch 235 selects (waveform data of) symbol data representing a reference signal read from the memory 233, and In periods other than the above, the symbol data from the remodulation circuit 232 is selected, and in the calibration process, the symbol data representing the desired signal read from the memory 233 is selected.

The weight calculator 236 has the same function as the weight calculator 553 shown in FIG. Memory 23
Reference numeral 7 includes a RAM and a ROM, and stores the weight vector calculated by the weight calculator 236 and the relative correction value of the radio unit B with respect to the radio unit A. This weight vector may be a weight vector calculated for a symbol at the end of the reception time slot in normal reception, used in a transmission time slot immediately after the reception time slot, and calculated in reception of a desired signal in calibration processing. The weight vector is stored and used in subsequent transmission of the desired signal. Radio unit A,
The weight vectors of B are W1 and W2.

The correction value is expressed by the following equations (20) and (21), and the value measured in the calibration process is written in the storage area of the ROM in the memory 237 before shipment from the factory. (20) Δθ12 = ((θTX1-θRX1)-(θTX2-θRX2)) (21) Amp12 = ((ATX1 / ARX1) / (ATX2 / ARX2)) FIG. 20 is an explanatory diagram of the correction values. ΘRX1, ARX1 in the figure
Indicates the amount of phase variation and the amount of amplitude variation caused by the signal passing from the antenna 210 through the changeover switch 213 and the receiving circuit 212, respectively. θTX1 and ATX1 are transmitted from the transmission circuit 211 and the changeover switch 213 to the antenna 210
The amount of phase fluctuation and the amount of amplitude fluctuation caused by the signal passing through are shown. θRX2 to θRX4 and ARX2 to ARX4 also indicate the same phase variation and amplitude variation in the respective radio units. In the above (20) and (21), Δθ12 and Amp12 mean the relative phase variation and amplitude variation of the radio unit B with respect to the radio unit A, respectively.

The weight control unit 238 reads the weight vectors W1 and W2 from the memory 237 in the transmission time slot in normal transmission, and
6 is output. The same applies to the transmission of a desired signal in the calibration process. The correction control unit 239 reads the correction values Δθ12 and Amp12 from the memory 237 in the transmission time slot during normal transmission, and
41, respectively. In addition, the correction control unit 239
In the calibration process, when the desired signal is transmitted, Δθ is changed from −180 degrees to +180 degrees, for example, by 1 degree, is output to the phase shifter 240, and Amp is gradually (for example, 0.5 degrees).
Amplifier 241 while changing it within 0.05 to 2).
Output to

The phase shifter 240 corrects the phase of the transmission signal input from the multiplier 225 by the correction value Δθ12 input from the correction controller 239. The amplifier 241 corrects the amplitude of the transmission signal input from the phase shifter 240 by the correction value Amp12 input from the correction control unit 239, and outputs the result to the transmission circuit 221. The modulation circuit 242 modulates a bit sequence to be transmitted in normal transmission to generate a transmission signal (symbol data).

The external I / F 250 includes an input / output port of the DSP 260 and a memory (the memory 233,
237 (including 237).
It is provided on a substrate of a mobile phone. External I / F250
Is connected to an external measuring device in the calibration process, and is used for input / output of various commands and their responses, programs, and data.

According to the portable telephone configured as described above, in normal transmission / reception, an array antenna pattern is formed and received using the weight vector calculated in the reception time slot, and the weight vector is stored in the memory 237. Then, an array antenna pattern is formed using the weight vector stored in the immediately following transmission time slot, and the transmission is performed.

At the time of this transmission, the correction control section 239 corrects the transmission signal to the radio section B using the correction value Δθ12 and Amp12 stored in the memory 237. As a result, correction can be made so that the array antenna pattern during reception and the array antenna pattern during transmission do not shift. In other words, the difference between the phase and amplitude fluctuation characteristics between the radio unit A and the radio unit B is corrected by simply correcting the transmission signal of the radio unit B without correcting the reference transmission signal of the radio unit A. The directivity and the directivity at the time of transmission can be matched.

Further, by providing the external I / F 250 and performing the calibration process under the control of the external measuring device, the above-mentioned correction value can be easily measured. Note that Δθ12 and Amp12 are the same physical quantities as the weight vector,
At 0, the correction weight vector representing Δθ12 and Amp12 may be stored in the memory 237, and a multiplier may be provided instead of the phase shifter 240 and the amplifier 241. The correction units 571 to 574 shown in FIG. 4 are also circuits equivalent to the phase shifter 240 and the amplifier 241 or circuits equivalent to the multiplier.

When the antennas 210 and 220 have different antenna gains, such as a rod antenna and a chip antenna, Δθ12 may be a value that takes into account the antenna gain compensation value A_cmp as in the following equation. . (21 ′) Amp12 = A_cmp · ((ATX1 / ARX1) / (ATX2 / ARX2)) <3.2 Configuration of Measurement Apparatus> FIG. 21 shows a measurement for measuring (calibrating) the correction value of the mobile phone shown in FIG. FIG. 2 is a block diagram illustrating a configuration of the device and a mobile phone.

As shown in the figure, the measuring device is composed of a transmitting and receiving device 30.
1, transmitting apparatus 302, timing adjuster 331, control P
C330, a clock generation circuit 332, and an I / F unit 333. The transmitting / receiving device 301 plays the role of Ant2 shown in FIG.
Signal selection unit 312, reception circuit 313, level measurement unit 31
4. A switch 315 is provided to receive a desired signal transmitted from the mobile phone 200 after transmitting the interference signal.

The transmission circuit 311 transmits the interference signal input from the signal selection section 312 from the antenna 310 via the switch 315. The signal selection unit 312 stores the symbol data strings of the plurality of interference signals, selects one of them, and outputs the selected one to the transmission circuit 311. The plurality of interference signals include a first interference signal composed of a PN code and a second interference signal composed of a known code string including the same fixed bit pattern (SS, PR, UW) as a normal transmission time slot. including. The selection of the interference signal is based on an instruction from the control PC 330.

[0107] The receiving circuit 313 receives a transmission signal with the null directed from the mobile phone 200 to the transmitting / receiving device 301 via the antenna 310 and the switch 315. The level measurement unit 314 measures the reception signal level of the reception signal by the reception circuit 313 and notifies the control PC 330 of the measured reception signal level. The transmission device 302 includes an antenna 320, a transmission circuit 321, and a signal selection unit 322 to perform the function of Ant1 shown in FIG. 2, and transmits a desired signal.

The transmitting circuit 321 transmits a desired signal input from the signal selecting section 322 from the antenna 320 via the switch 325. The signal selection unit 322 stores a symbol data sequence of a plurality of desired signals, selects one of them, and outputs it to the transmission circuit 321. The plurality of desired signals include a first desired signal composed of a PN code orthogonal to the first interference signal and a fixed bit pattern (S
S, PR, UW).
A desired signal. The selection of the desired signal is based on an instruction from the control PC 330.

The timing adjuster 331 is connected to the signal selector 3
12, the first interference signal by the signal selection unit 322,
When the first desired signal is selected, the clock signal (symbol clock) input from the signal selecting unit 322 is directly output to the transmitting / receiving device 301, and the signal selecting unit 322 generates the second interference signal and the second desired signal, respectively. If selected, the clock signal input from the signal selection unit 322 is delayed by, for example, 0.5 symbol time, so that the transmission / reception device 301
Is output to the transmitting / receiving device 301. The reason for the delay is that the second interference signal and the second desired signal include the same fixed bit pattern (SS, PR, UW, etc.). That is,
This is to facilitate separation of a desired signal in the mobile phone 200. When the first interference signal and the first desired signal are selected, the timing adjuster 331 does not delay, but it may be delayed to simplify the configuration.

The control PC 330, similarly to the calibration processing shown in FIG. 12, controls the transmission / reception device 301, the transmission device 302, and the Table 33
1. Control the mobile phone 200. Clock generation circuit 3
32 outputs a clock signal indicating the symbol timing to the transmission device 302 and the timing adjuster 331.

The I / F unit 333 is connected to the external I / F 250 in the mobile phone 200, and is an interface for inputting and outputting commands and data to and from the mobile phone 200. FIG. 22 shows the present measuring device and the mobile phone 20.
0 shows an external appearance and a physical connection example. In the figure, the mobile phone 200 shows only the substrate excluding the housing,
The / F section 333 is a connector that fits into the external I / F 250 on the board. Also, the transmitting / receiving device 301, the transmitting device 30
2 can be constituted by a general signal generator. Alternatively, the transmitting / receiving device 301 and the transmitting device 302
May be modified from a wireless base station or a mobile phone.

The external I / F 250 need not be a connector but may be a plurality of pads provided on a substrate. In this case, the I / F unit 333 may be a probe connected to a plurality of pads. It is desirable that the measuring device and the mobile phone shown in FIG. 22 be placed in an electromagnetically shielded environment such as an anechoic chamber during the calibration process. <3.3 Calibration process> FIGS. 23 and 24
9 is a flowchart showing a calibration process executed under the control of the control PC 330. This figure basically shows the same processing as that of FIG.

As described above, according to the present measuring device, the relative correction value of the radio unit B with respect to the radio unit A in the mobile phone 200 is measured, and the correction value is set in the mobile phone 200. I do. <4 Other Modifications> Modifications to the configuration shown in the above embodiment will be described below. (1) In the above radio base station, relative phase variation and amplitude variation were measured for all four radio units.
In order to calculate the correction value of each wireless unit, it is sufficient to measure the relative phase variation and amplitude variation for one wireless unit less than the number of all wireless units. For example, it is sufficient to measure cases 1 to 3 shown in FIG. This is because the correction value is a relative value based on one wireless unit, and the wireless unit serving as a reference need not be corrected.

In the above embodiment, the relative phase variation and amplitude variation are measured for all the radio units because the phase variation and the amplitude variation by the equations (17) and (18) are used. This is to determine the validity. (2) The weight vector used in the array transmission by Ant3 and Ant4 in FIG. 2B may not be the one calculated in the array reception in FIG. For example,
The weight vector used in the previous calibration process may be stored in memory and used, or Ant2
A weight vector having a property of turning null into a vector may be acquired from the outside, or may be stored in advance. In this case, the processing of FIG. 2A can be omitted.

As a weight vector having a property of turning null to Ant2, a weight vector for turning forced null to Ant2 in FIG. 2A may be calculated. Forced null refers to pointing null in a specific direction. (3) In the above embodiment, Δθ, A_Amp when the reception signal level at Ant2 is minimized as shown in FIG.
Was determined as Δθ34 and Amp34. Alternatively or together with this, Δθ and A_Amp when the received signal level at Ant1 becomes the maximum may be set to Δθ34 and Amp34. This is because the array antenna pattern in FIG. 2B is formed so that Ant1 has the maximum gain. (4) In the calibration process of FIGS. 8 and 9, the relative phase variation and amplitude variation are measured for all wireless units, but for one wireless unit as shown in FIGS. Alternatively, the relative phase variation and the amplitude variation may be simply measured for the two radio units. For example, when the correction value holding unit 570 already holds the correction value of each wireless unit, the amount of phase variation necessary for calculating the correction value of the wireless unit,
It is sufficient to measure the amplitude fluctuation. (5) It is preferable that the calibration process is periodically performed in the wireless base station. This is because the characteristic difference between the time of transmission and the time of reception changes depending on the installation environment of the wireless base station or aging.

In this case, the correction value holding unit 570 also holds the phase fluctuation amount and the amplitude fluctuation amount of each wireless unit, and partially compares / updates the newly measured phase fluctuation amount and the amplitude fluctuation amount. It may be. If the comparison result is significantly different (more than the threshold value), the calibration process may be performed for all wireless units. The correction units 571 to 574 change the phase compensation amount Δθ and the amplitude compensation amount Amp little by little in the calibration process.
The configuration may be such that θ and the amplitude compensation amount Amp are held, and are changed based on the held phase compensation amount Δθ and amplitude compensation amount Amp. That is, the DSP 50 performs the correction value holding unit 5
The phase compensation amount Δθ and the amplitude compensation amount Amp of 70 may be updated while being gradually changed. (6) In the above embodiment, the DSP 50 controls almost all of the calibration processing. However, the DSP 50 may share the control with the control unit 80. (7) In the above embodiment, 2 is used to direct null to Ant2.
The description has been given on the assumption that the array transmission is performed by one of the radio units Ant3 and Ant4. May be directly obtained. In this case, a known signal such as an unmodulated signal may be transmitted from the transmitting side to the receiving side, and the amount of phase variation and the amount of amplitude variation may be measured from the signal input to the DSP 50 from the wireless unit on the receiving side. (7) As shown in the above embodiment, the main part of the present invention in the adaptive array device as a radio base station is:
The DSP 50 provided in the adaptive array device, that is, the digital signal processor executes the program to execute the program. This program includes PROM, EE
It is stored in PROM or RAM, upgraded by ROM exchange, downloaded to EEPROM or RAM via a program recording medium, a network or a telephone line, and can be read by a digital signal processor. (8) In the mobile phone 200, the correction control unit 23
9. The configuration may be such that the functions are realized by the weight control unit 238 and the multiplier 225 without the phase shifter 240 and the amplifier 241. In this case, the weight control unit 238 stores the weight vector W2 from the memory 237.
In addition, a weight vector taking into account the correction values Δθ12 and Amp12 is calculated, and the calculated weight vector
5 may be weighted. this is,
This is because the weight vector is a physical quantity equivalent to the phase and the amplitude in the first place. Further, in this case, any one of the wireless units A and B may be used as a reference. In addition, the dashed line in FIG.
Since the functions realized by 260 are shown, the configuration of the embodiment and the above configuration are substantially the same, and can be easily realized. (9) Steps 87 and 88 in FIG. 8 and 187 and 1 in FIG.
In step 88, the phase and the amplitude are changed at a fixed step size (the phase is changed by 1 degree in the range of -180 to 180 degrees, and the magnification of the amplitude is changed by 0.05 in the range of 0.50 to 2.00). The received signal level is sequentially measured. However, the received signal level is measured in large steps (for example, 90 degrees in phase and 0.5 in amplitude), and the amount of phase and amplitude at which the received level is minimized are measured. After finding the magnification, the received signal level may be measured while changing the second range including the found phase amount and amplitude magnification with a small step size (for example, 1 degree, 0.05). Thus, the time for the calibration process can be reduced.

Further, steps 87 and 88 in FIG.
In steps 187 and 188, the step may be stopped when the minimum amount of phase and amplitude is found. (11) In the above embodiment, the mobile phone 200 includes two wireless units, but may be configured to include three or more wireless units. In that case, the antenna may be mounted selectively from a rod antenna, a pattern antenna, and a chip antenna. Further, the measuring device is configured to measure a correction value with respect to the reference radio unit for each of the radio units other than the reference one radio unit, and the mobile phone is configured to correct each transmission signal other than the reference radio unit. do it. In this case, for any of the reasons (8), any wireless unit can be used as a reference. In the calibration processing of FIGS. 14 and 15, correction values are measured in steps 182 to 192 for each of the reference wireless unit and the wireless unit to be measured, and thereafter, steps 98 and 99 in FIG.
The validity of the measured correction value may be determined in the same manner as described above.

Further, when the mobile phone has four or more radio units, the calibration process is performed by the mobile phone alone, similarly to the radio base station in the embodiment, without providing an external measuring device. It can be. In this case, a configuration may be adopted in which a program for the calibration process is downloaded from an external device to the memory in the mobile phone via the external I / F 250, and is deleted after the measurement. Also, the program is left in the memory (RO
M (stored in M). When the data is stored in the ROM, the calibration process can be performed by a user operation after shipment, and a change with time of the wireless unit can be absorbed. (12) In the above embodiment, the control PC 330 controls the mobile phone 200, the transmitting / receiving device 301, and the transmitting device 302 as the main body of the calibration process.
A program for performing the calibration process may be downloaded to a memory inside the mobile phone 200 via the mobile phone 200, and the mobile phone 200 may be configured to be the main control unit. (13) In the above embodiment, commands and data are input / output to / from the control PC 330 via the external I / F 250, but commands / data / programs are input / output via the wireless unit, and the DSP 260 , May be configured to execute a program. In this case, the cost can be reduced because the external I / F 250 does not need to be provided.

[0119]

According to the adaptive array apparatus of the present invention,
An adaptive array device that includes at least three radio units each including a transmission unit, a reception unit, and an antenna, and measures a phase and amplitude characteristic of a third radio unit relative to a second radio unit. The first and second wireless units receive the transmission signal of the first and second wireless units, and, from the received signals of the second and third wireless units, a first calculating unit that calculates a response vector indicating a propagation path of the transmission signal, A second calculating unit configured to calculate a weight vector indicating a directivity pattern in which a null is directed to the antenna of the first wireless unit based on the response vector;
Control means for transmitting signals by forming directivity patterns in the second and third radio units using the calculated weight vector, and the first radio unit receiving transmission signals from the second and third radio units Based on the received signal level at
Measuring means for measuring a phase variation and an amplitude variation of the third wireless unit relative to the second wireless unit.

According to this configuration, the third and fourth wireless units are caused to perform array transmission without adding an additional circuit, and relative phase variation characteristics and amplitude variation specification are measured based on the received signals. There is an effect that can be. further,
Since the weight vector is calculated after obtaining the response vector obtained by the ensemble averaging from the transmission signal from the first wireless unit, the influence of noise and fading can be removed as compared with the weight vector obtained by the sequential processing. , The amount of phase fluctuation and the amount of amplitude fluctuation can be measured with higher accuracy.

Further, as compared with the case where the weight vector is calculated from the desired signal and the interference signal by the second and third radio units, the accuracy of the weight vector is less likely to deteriorate due to the superposition of the direct wave and the indirect wave. The phase fluctuation amount and the amplitude fluctuation amount can be measured with higher accuracy. Here, the second calculator calculates a virtual response vector having no correlation with the response vector calculated by the first calculator, and calculates a beam in a direction indicated by the virtual response vector by the first calculator. It may be configured to calculate a weight vector that turns null in the direction indicated by the response vector.

According to this configuration, the weight vector is calculated by assuming a virtual desired wave in a direction having no correlation with the direction indicated by the response vector by the virtual response vector.
The accuracy of the direction in which the null is directed can be improved, and the null can be made deeper. Here, the control unit changes the phase and the amplitude of the transmission signal of the third radio unit during the transmission by the second and third radio units, and the measurement unit changes the phase of the second and third radio signals during the change. The first that received the transmission signal from the radio unit
The configuration may be such that the phase variation and the amplitude variation are measured based on the values of the phase and the amplitude when the reception signal level in the wireless unit is minimized.

According to this configuration, it is detected that the null is correctly directed to the first radio section by detecting that the received signal level has become minimum, and thus the relative phase fluctuation and amplitude fluctuation can be determined. Can be measured. Here, the control unit further activates the first and second calculation units by relatively exchanging the first to third wireless units, and performs a relative phase change of each of the other wireless units with respect to the one wireless unit. A configuration in which the amount and the amplitude variation are measured by the measuring means, and the phase correction value and the amplitude correction value for each wireless unit other than the one wireless unit are calculated based on the measured phase variation and each amplitude variation. It may be.

According to this configuration, it is possible to calculate the relative phase correction value and amplitude correction value of each of the other radio units with reference to one of the radio units. Further, the control means further determines the validity of the sum of the relative phase variation measured for each radio unit and the product of the amplitude variation is within a predetermined range. It may be configured to calculate the phase correction value and the amplitude correction value when it is determined.

According to this configuration, the validity of the measured phase variation and amplitude variation for each radio unit can be easily determined by utilizing the relativity of the transmission characteristics for each radio unit. As a result, it is possible to avoid using a correction value having a large measurement error. The calibration method of the present invention is a calibration method for measuring a correction value of a third antenna with respect to a second antenna using first to third antennas.
A first step of transmitting a signal, a second step of calculating a response vector indicating a propagation path of the first signal based on a received signal of the first signal at each of the second and third antennas,
A third method for calculating a weight vector indicating directivity with a null pointing toward the first antenna based on the calculated response vector
Step 2 and the second using the calculated weight vector
And a fourth step of array transmitting the second signal from the third antenna and causing the first antenna to receive the second signal, and changing the amplitude and phase of the second signal transmitted from the third antenna, Measuring a correction value of the third antenna with respect to the second antenna based on the receiving signal level of the changing first antenna.

Further, the program recording medium of the present invention is a recording medium for storing a computer readable program of an adaptive array device having first to third antennas, wherein the first to fifth steps are performed by a computer. A program recording medium characterized by recording a program to be executed by a computer. Here, the third step is the second step
A sub-step of calculating a virtual response vector having no correlation with the response vector calculated in the step; a beam in a direction indicated by the virtual response vector; and a null in a direction indicated by the response vector calculated by the first calculation unit. And a sub-step of calculating an aimed weight vector.

According to this configuration, the third and fourth radio units are caused to perform array transmission without adding an additional circuit, and relative phase variation characteristics and amplitude variation specification are measured based on the received signals. There is an effect that can be. further,
Since the weight vector is calculated after obtaining the response vector obtained by the ensemble averaging from the transmission signal from the first wireless unit, the influence of noise and fading can be removed as compared with the weight vector obtained by the sequential processing. , The amount of phase fluctuation and the amount of amplitude fluctuation can be measured with higher accuracy.

In addition, compared to the case where the weight vector is calculated from the desired signal and the interference signal by the second and third radio units, the accuracy of the weight vector is less likely to deteriorate due to the superposition of the direct wave and the indirect wave. The phase fluctuation amount and the amplitude fluctuation amount can be measured with higher accuracy. Also,
The computer that has read the program can measure the relative phase correction value and amplitude correction value of the wireless unit of the wireless terminal without any additional device.

[Brief description of the drawings]

FIG. 1 is a diagram showing a schematic configuration of a main part of an adaptive array device according to an embodiment of the present invention.

FIG. 2 shows a relative phase variation Δθ34 and an amplitude variation Amp34.
FIG. 4 is an explanatory diagram illustrating a schematic operation of the adaptive array device when measuring the distance.

FIG. 3 is a block diagram illustrating an overall configuration of a wireless base station.

FIG. 4 is a block diagram showing a detailed configuration of the DSP 50.

FIG. 5 is a block diagram illustrating a detailed configuration of a user processing unit 51a.

FIG. 6 is a diagram showing a list of processing contents of each user processing unit.

FIG. 7 shows a physical wireless unit 1 to a wireless unit 4 and a logical wireless unit.
It is a figure which shows the correspondence with Ant1-Ant4.

FIG. 8 is a flowchart showing the contents of a calibration process.

FIG. 9 is a flowchart showing a process for measuring a phase variation Δθ34.

FIG. 10 is a flowchart illustrating a measurement process of an amplitude variation Amp34.

FIG. 11 is a flowchart showing a process of verifying Δθ and Amp and calculating a correction value.

FIG. 12 is an explanatory diagram illustrating an outline of a calibration process in the wireless base station according to the second embodiment.

FIG. 13 is a block diagram illustrating an overall configuration of a wireless base station according to the embodiment.

FIG. 14 is a block diagram showing a detailed configuration of a DSP 500.

FIG. 15 is a block diagram illustrating a detailed configuration of a user processing unit 551a.

FIG. 16 is a diagram showing a list of processing contents of each user processing unit.

FIG. 17 is a diagram showing a correspondence between physical wireless units 1 to 4 and logical wireless units Ant1 to Ant4.

FIG. 18 is a flowchart showing more detailed contents of a calibration process.

FIG. 19 is a block diagram illustrating a configuration of a main part of a mobile phone according to a third embodiment.

FIG. 20 is an explanatory diagram of a correction value of the mobile phone.

FIG. 21 is a block diagram illustrating a configuration of a measurement device that measures a correction value of a mobile phone and a mobile phone.

FIG. 22 shows an appearance and a physical connection example of a measuring device and a mobile phone.

FIG. 23 is a flowchart showing a calibration process executed under the control of the control PC 330 (part 1).

FIG. 24 is a flowchart showing a calibration process executed under the control of the control PC 330 (part 2).

[Explanation of symbols]

 1-4 Radio unit 10-40 Antenna 11-14 Front end unit 50 Signal processing unit 51a-51d User processing unit 53 Weight calculation unit 54 Adder 55 Memory 56, 57 Switch 60 Modem unit 70 Baseband unit 80 Control unit 111 Transmission Unit 112 reception unit 113 antenna switch 521 to 524 multiplier 551 to 554 adder 561 to 564 switch 564 switch 570 correction value holding unit 572 to 574 correction unit 581 to 584 multiplier 200 mobile phone 210, 220 antenna 211, 221 transmission circuit 212, 222 Receiving circuit 213, 223 Changeover switch 214, 215, 225 Multiplier 230 Adder 231 Demodulation circuit 232 Remodulation circuit 233 Memory 234 Counter 235 Switch 236 Weight calculation unit 23 7 memory 238 weight control section 239 correction control section 240 phase shifter 241 amplifier 242 modulation circuit 250 external I / F 260 DSP

Continued on the front page (72) Inventor Shogo Nakao 2-5-5 Keihanhondori, Moriguchi-shi, Osaka Sanyo Electric Co., Ltd. F-term (reference) 5J021 AA03 AA04 AA05 AA11 DB02 DB03 EA04 FA06 FA13 FA14 FA16 FA26 FA29 FA30 FA31 GA02 HA05 JA10 5K052 AA01 AA12 BB01 DD04 EE38 FF32 FF33 FF34 GG11 GG31 GG57 5K059 CC01 CC04 DD33 DD35 EE02

Claims (14)

[Claims] An adaptive array device including at least three radio units each including a transmission unit, a reception unit, and an antenna, and measuring a phase and amplitude characteristic of a third radio unit relative to a second radio unit, A first calculating unit that receives a transmission signal from the first wireless unit by the second and third wireless units and calculates a response vector indicating a propagation path of the transmission signal from each of the received signals of the second and third wireless units; And second calculating means for calculating a weight vector indicating a directivity pattern in which a null is directed to the antenna of the first radio unit based on the calculated response vector; and using the calculated weight vector, Control means for forming a directivity pattern in the third radio section to transmit a signal; and a second radio section based on a reception signal level in the first radio section which has received transmission signals from the second and third radio sections. Adaptive array apparatus comprising: a measuring means for measuring the relative phase fluctuation amount and amplitude fluctuation amount of third wireless unit for the line unit. 2. The second calculating means calculates a virtual response vector having no correlation with the response vector calculated by the first calculating unit, and outputs a beam in a direction indicated by the virtual response vector by the first calculating unit. 2. A weight vector for nulling in a direction indicated by the calculated response vector.
An adaptive array device as described. 3. The control unit changes a phase and an amplitude of a transmission signal of a third radio unit during the transmission by the second and third radio units. And measuring the phase variation and the amplitude variation based on the phase and amplitude values when the received signal level in the first wireless unit that has received the transmission signal from the third wireless unit is minimized. The adaptive array device according to claim 2, wherein 4. The control means further comprises: firstly and thirdly replacing the first to third wireless units,
Activating the calculating unit, and causing the measuring unit to measure the relative phase variation and the amplitude variation of each of the other radio units with respect to the one radio unit, 4. The adaptive array apparatus according to claim 3, wherein the correction value is calculated based on each measured phase variation and each amplitude variation. 5. The control means further determines the validity of the sum of the relative phase fluctuations and the product of the amplitude fluctuations measured for each radio unit, based on whether or not the respective products are within predetermined ranges. 5. The adaptive array apparatus according to claim 4, wherein the phase correction value and the amplitude correction value are calculated when it is determined to be valid. 6. A calibration method for measuring a correction value of a third antenna with respect to a second antenna using the first to third antennas, wherein a first step of transmitting a first signal from the first antenna; A second step of calculating a response vector indicating a propagation path of the first signal based on the received signal of the first signal at each of the second and third antennas; and setting a null to the first antenna based on the calculated response vector. 3rd to calculate weight vector indicating directivity toward
Step, array transmitting the second signal from the second and third antennas using the calculated weight vector, and transmitting the second signal to the first signal.
A fourth step of causing the antenna to receive, changing the amplitude and phase of the second signal transmitted from the third antenna, and changing the third antenna with respect to the second antenna based on the changing received signal level of the first antenna And a fifth step of measuring the correction value of. 7. The sub-step of calculating a virtual response vector having no correlation with the response vector calculated in the second step, and the first calculation of a beam in a direction indicated by the virtual response vector. 7. The calibration method according to claim 6, further comprising a sub-step of calculating a weight vector for turning null in a direction indicated by the response vector calculated by the unit. 8. The method of claim 5, wherein the step (c) comprises:
8. The calibration method according to claim 7, wherein the amount of phase variation and the amount of amplitude variation are set based on the values of the phase and the amplitude when the signal level received by the antenna is minimized. 9. The control unit further includes: a sixth step of relatively replacing the first to third wireless units to execute the first to fifth steps; 9. The calibration method according to claim 8, further comprising: a seventh step of calculating the phase correction value and the amplitude correction value based on each of the measured phase variation and each amplitude variation. 10. The seventh step further comprises determining whether the sum of the relative phase fluctuations and the product of the amplitude fluctuations measured for each radio unit are within a predetermined range. 10. The calibration method according to claim 9, wherein the phase correction value and the amplitude correction value are calculated when it is determined to be valid. 11. A recording medium for storing a computer-readable program of an adaptive array device having first to third antennas, wherein the first step is to transmit a first signal from the first antenna; A second step of calculating a response vector indicating the propagation path of the first signal based on the received signal of the first signal at each of the third antennas; and pointing the null to the first antenna based on the calculated response vector. 3rd to calculate the weight vector indicating the property
Step, array transmitting the second signal from the second and third antennas using the calculated weight vector, and transmitting the second signal to the first signal.
A fourth step of causing the antenna to receive, changing the amplitude and phase of the second signal transmitted from the third antenna, and changing the third antenna with respect to the second antenna based on the changing received signal level of the first antenna Recording a program for causing a computer to execute a fifth step of measuring the correction value of the program. 12. The third step includes a sub-step of calculating a virtual response vector having no correlation with the response vector calculated in the second step; and a first calculation of a beam in a direction indicated by the virtual response vector. 12. The program recording medium according to claim 11, further comprising: a sub-step of calculating a weight vector for turning null in a direction indicated by the response vector calculated by the unit. 13. A calibration apparatus having a first antenna for a wireless terminal having first and second wireless units including a transmitting unit, a receiving unit, and an antenna for wireless communication by forming an array antenna pattern. A calibration method, comprising: a first step of receiving a transmission signal from a first antenna in a first and a second wireless units, and calculating a response vector indicating a propagation path of the transmission signal from the received signal; A second step of calculating a weight vector indicating directivity with a null pointing toward the first antenna based on the response vector obtained, and using the calculated weight vector to calculate an array antenna pattern in the first and second radio units. A third step of forming and transmitting a signal, wherein the first antenna receives the transmission signals of the first and second radio units. Based on the signal the signal level, the fourth calibration method characterized by a step of measuring the relative correction value of the second radio unit to the first radio unit. 14. The third step includes a sub-step of calculating a virtual response vector having no correlation with the response vector calculated in the second step; and a first calculation of a beam in a direction indicated by the virtual response vector. 12. The calibration method according to claim 11, further comprising: a sub-step of calculating a weight vector for turning null in a direction indicated by the response vector calculated by the unit.