JPH0669153B2 - Adaptive antenna receiver for mobile communication - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to an adaptive antenna receiver for mobile communication that automatically prevents reception interference due to an interference wave.

[Prior art and its problems]

In wireless communication, a received signal is often interfered with by an interfering signal, and it is important to eliminate the influence and obtain a received signal of good quality. Conventionally, various so-called adaptive antennas for automatically removing interference have been proposed in the fields of radar, satellite communication, mobile communication and the like. Although various types of adaptive antenna operation algorithms themselves have been proposed, the basic configuration thereof is to provide at least two or more antennas for receiving a desired wave signal and an interference wave signal together, and combine these antenna outputs. Sometimes, the parameters at the time of combining are automatically determined so that only the interference wave signal is canceled. An example of such a technique is U.S. Pat. No. 3,202,990 PW Howells, "Intermediate Frequency Canceller".
nceller) ”, which is used for radar.
This technology is RT Compton
By The Power Inversion Adaptive
Array: Concept and Performance
power inversion adaptive array: concept and perform
ance) ”(IEE transaction on Erowspace and Electronics
Systems (IEEE Transaction on Aerospace and Elec
tronics Systems), vol.AES-15, pp.803-814, November
In principle, it is included in the power inversion method that has been extended to 1979.).

However, the power inversion method has a drawback that it operates effectively only when the difference in power between the desired wave signal and the interference wave signal is large. Therefore, in mobile communication, the powers of the desired wave signal and the interference wave signal change due to fading, and there is no guarantee that the powers of these signals always have a predetermined level or more, so the power inversion method is suitable for mobile communication. Not a thing.

Also, B. Widraw's “Adaptive antenna Syst”
m) '' (proceedings of the IEEE, vol.55, pp.2143−215
9, December 1967) LMS (Least-Mean
-Squares) method does not have the above-mentioned drawbacks, but has a new problem that a reference signal is required on the receiving side. In analog voice communication by frequency modulation, which is usually performed in mobile communication, it is generally difficult to obtain such a reference signal.

In addition, “Interference Casselation System for Satellite Communication Earth Station (Interf) by T. Kaitsuka, T. Inoue
erence cancellation system for sattelite communica
tion earth station) "(vol.COM-32, pp.796-803, Ju
The method described in ly 1984) does not have the above-mentioned two drawbacks, but uses a highly directional antenna and assumes that the signal received by the backup antenna is only an interfering wave. It cannot be applied to mobile communications using omnidirectional antennas.

[Object of the Invention]

An object of the present invention is to use an omnidirectional antenna and not to require a reference signal on the receiving side, and to automatically prevent reception interference by an interference wave even when the power difference between the desired wave and the interference wave is small. An object of the present invention is to provide an adaptive antenna receiver for mobile communication, which can prevent and extract a desired signal.

[Structure of Invention]

The present invention provides first and second receiving antennas for receiving a composite wave of a desired wave and an interfering wave, and a variable gain amplifier and a variable phase shifter connected to at least one of the first and second receiving antennas. Controlling the amplitude and phase relationship between the received signals of the first and second receiving antennas by inputting the first and second control signals to the variable gain amplifier and the variable phase shifter, respectively. A complex multiplication circuit that outputs two output signals, a signal synthesis circuit that synthesizes the two output signals of the complex multiplication circuit, and a reception circuit that receives a part of the output of the signal synthesis circuit and performs demodulation. A first square-law detection circuit for inputting part of the output of the signal synthesis circuit, a high-pass filter for inputting the output of the square-law detection circuit, and a second for inputting the output of the high-pass filter. Square detection circuit of And a low frequency signal generation circuit for generating a second low frequency signal, a first correlation circuit for obtaining a correlation between the output signal of the second square wave detection circuit and the first low frequency signal, and the second A second correlation circuit for obtaining a correlation between the output signal of the square detection circuit and the second low frequency signal;
A first subtraction circuit for outputting a signal obtained by subtracting the first low-frequency signal from the output signal of the correlation circuit to the variable gain amplifier of the complex multiplication circuit as the first control signal; Second subtraction circuit for outputting a signal obtained by subtracting the second low-frequency signal from the output signal of the correlation circuit to the variable phase shifter of the complex multiplication circuit as the second control signal. Is characterized by.

[Action]

In the present invention, in the FM receiver for mobile communication, paying attention to the fact that the amplitude component fluctuates in the frequency-modulated wave when an interfering wave is present, that is, the beat component is generated, and this beat component becomes zero. The interference wave is removed by controlling by the perturbation method so that

〔Example〕

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

FIG. 1 is a block diagram of a receiver according to the present invention. In this figure, 1 is a complex multiplication circuit, and the complex multiplication circuit 1 is provided with two receiving antennas 2 and 3 via input terminals 101 and 102. This receiver is an FM receiver for mobile communication and has a receiving antenna.
2 and 3 are omnidirectional adaptive antennas. The receiving antennas 2 and 3 are assumed to receive the signals S ₁ (t) and S ₂ (t), respectively. Control signals α (t) and β (t) are input to the input terminals 103 and 104 from subtraction circuits 4 and 5 described later, respectively.

The complex multiplication circuit 1 is realized, for example, as shown in FIG. In FIG. 2, 6 is a variable gain amplifier, 7 is a variable phase shifter, and the control signal α (t) is the variable gain amplifier 6
And the control signal β (t) is input to the variable phase shifter 7. In the complex multiplication circuit 1 configured as described above, as an example, the signal S ₁ (t) input from the input terminal 101 is directly output to the output terminal 105 and the signal input from the input terminal 102.
The amplitude of S ₂ (t) is amplified by the variable gain amplifier 6, the phase is controlled by the variable phase shifter 7, and the signal is output to the output terminal 106 as the signal S ′ ₂ (t). Therefore, in this case the signal S
The amplitude and phase of ₂ (t) are control signals α (t) and β (t).
Controlled by the signals S ₁ (t) and S ₁
The amplitude and phase relationship of ₂ (t) will be controlled.

The signal S ₁ (t), S ′ output from the complex multiplication circuit 1
₂ (t) is input to the signal synthesizing circuit 8, is synthesized here, and is output as a signal Z (t). This output signal Z
(T) is given to the FM receiving circuit 9 and the square detection circuit 10. A part of the signal Z (t) is taken out to the output terminal 11 as a reception signal after passing through the FM reception circuit 9. Also the signal Z (t)
Part of the detection output signal D (t) by the square detection circuit 10.
Is input to the high pass filter 12. The output signal H (t) obtained by the high-pass filter 12 passes through the square detection circuit 13 to become the signal C (t), and the correlation circuit
Given to 14,15.

The above-mentioned signal C (t) and the output signal εcosωt of the low frequency oscillation circuit 16 are input to the correlation circuit 14, and the above-mentioned signal C (t) and the output signal εsinωt of the 90 ° phase shift circuit 17 are input to the correlation circuit 15. It The low frequency oscillation circuit 16 and the 90 ° phase shift circuit 17 constitute a low frequency signal generation circuit 18. The correlation circuits 14 and 15 are specifically realized by a multiplier 19, a low pass filter 20 and a loop filter 21, as shown in FIG. 3, for example. The multiplier 19 has two input terminals 19a and 19b, and the signal C is input to the input terminal 19a.
(T), εcosωt or εsinωt is input to the input terminal 19b, respectively. The signals α ₀ and β ₀ are output from the correlation circuits 14 and 15, respectively, and are input to the subtraction circuits 4 and 5. Further, the subtraction circuits 4 and 5 have the above εcosωt and εsinωt.
Respectively, and as a result, the control signal α
(T) and β (t) are output to the complex multiplication circuit 1.
In this case, α (t) and β (t) are expressed as α (t) = α ₀ −εcosωt β (t) = β ₀ −εsinωt. Where ω is the frequency of the low frequency signal, ε
Is a small positive number (ε << 1).

Next, the operation of the receiver having the above configuration will be described by expressing the signals of the respective parts of the circuit by mathematical expressions.

First, the received signals S ₁ (t) and S ₂ (t) are expressed as follows.

S ₁ (t) = D ₁ (t) + U ₁ (t) (1) S ₂ (t) = D ₂ (t) + U ₂ (t) (2) where D ₁ (t), D ₂ (t) is the receiving antenna 2, 3 respectively
The desired wave signals, U ₁ (t) and U ₂ (t), received by the are the interfering wave signals. D ₁ (t), D ₂ (t), U ₁ (t) and U ₂ (t) are further expressed as follows.

D ₁ (t) = A ₁ exp {j (ωct + θ (t) + θ ₁ )} ...
(3) D ₂ (t) = A ₂ exp {j (ωct + θ (t) + θ ₂ )} ...
(4) U ₁ (t) = B ₁ exp {j (ωct + φ (t) + φ ₁ )} ...
(5) U ₂ (t) = B ₂ exp {j (ωct + φ (t) + φ ₂ )} ...
(6) In the above equation, A ₁ , A ₂ , B ₁ , and B ₂ are amplitudes, ωc is a carrier frequency, and θ (t) and φ (t) are phase signals determined by modulation signals of a desired wave and an interfering wave, respectively. θ ₁ , θ ₂ , φ ₁ and φ ₂ are phase constants. These amplitude and phase constants are assumed to change due to fading, and the changing speed is sufficiently slow.

A signal S ′ ₂ which is controlled in amplitude and phase by the complex multiplication circuit 1.
(T) is represented by α (t) and β (t) as follows.

S ′ ₂ (t) = α (t) {D ₂ (t) + U ₂ (t)} exp (jβ (t)) (7) As a result, the output signal Z (t) of the signal synthesis circuit 8 is Z (t) = S ₁ (t) + S ′ ₂ (t) = [A ₁ exp (jθ ₁ ) + α (t) A ₂ exp {j (θ ₂ + β (t))}] exp (jθ (t) ) + [B ₁ exp (jθ ₁ ) + α (t) B ₂ exp {j (θ ₂ + β (t))}] exp (jφ (t)) (8). The detection output signal D (t) is, D (t) = | Z (t) | 2 = | A (t) | 2 + | B (t) | 2 +2 | A (t) || B (t) | cos {φ (t) −θ (t) + ψa (t) −ψb (t)} (9), where | A (t) | ² = A ₁ ² α ² (t) A ₂ ² + 2α (t) A ₁ A ₂ cos {θ ₂ −θ ₁ + β (t)} (10) | B (t) | ² = B ₁ ² + α ² (t) B ₂ ² + 2α (t) B ₁ B ₂ cos {θ ₂ −θ ₁ + β (t)} (11) Is.

The output signal D (t) of the square-law detection circuit 10 is passed through a high pass filter.
If the cutoff frequency of the high-pass filter 12 is higher than the frequency ω of the low-frequency signal, then the output signal H (t) obtained by inputting to 12 is H (t) = | A (t) || B (T) | cos {φ (t) −θ (t) + φa (t) −φb (t)} (14) This is the beat component of the desired wave and the disturbing wave, and cos
{Φ (t) −θ (t) + φa (t) −φb (t)} includes a modulation component, and for example, in the case of frequency modulation by a voice signal, the spectrum extends up to several kHz. Also signal | A
(T) | ² and | B (t) | ² are low frequency signals.

Next, the signal C (t) obtained by inputting the signal H (t) into the square detection circuit 13 is C (t) = | A (t) | ² | B (t) | ² (15) If | A | ² , | B | ² is approximated under the condition of ε << ₁ , | A (t) | ² ≈A ₁ ² + α ₀ A ₂ ² + 2α ₀ A ₁ A ₂ cos (θ ₂ − θ ₁ + β ₀ ) −2 {α ₀ A ₂ + A ₁ A ₂ cos (θ ₂ −θ ₁ + β ₀ )} ε cosωt + 2α ₀ A ₁ A ₂ εsinωt (16) | B (t) | ² ≈B ₁ ² + α ₀ B ₂ ² + 2α ₀ B ₁ B ₂ cos (θ ₂ −θ ₁ + β ₀ ) +2 {α ₀ B ₂ ² + B ₁ B ₂ cos (θ ₂ −θ ₁ + β ₀ )} ε cosωt + 2α ₀ B ₁ B ₂ εsinωt ... (17) As a result, C (t) ≈A ₀ ² B ₀ ² + 2A ₀ [{α ₀ B ₂ ² + B ₁ B ₂ cos (θ ₂ −θ ₁ + β ₀ )} εcosωt−2α ₀ B ₁ B ₂ εsinωt] + 2B ₀ [{Α ₀ A ₂ ² + A ₁ A ₂ cos (θ ₂ −θ ₁ + β ₀ )} εcosωt-2α ₀ A ₁ A ₂ εsinωt] (1
8) Here, A ₀ = A ₁ ² + α ₀ ² A ₂ ² + 2α ₀ A ₁ A ₂ cos (θ ₂ −θ ₁ + β ₀ ) ... (19) B ₀ = B ₁ ² + α ₀ ² B ₂ ² + 2α ₀ B ₁ B ₂ cos (θ ₂ −θ ₁ + β ₀ ) ... (20).

The signal C (t) obtained as described above is used for the correlation circuits 14 and 15
To the low frequency signals ε cos ωt and ε sin ωt. The correlation signal Cc between the signal C (t) and ε cosωt and the correlation signal Cs between the signal C (t) and εsin ωt are given as follows.

Cc = −A ₀ ε ² {α ₀ B ² + B ₁ B ₂ cos (θ ₂ −θ ₁ + β ₀ )} − B ₀ ε ² {α ₀ A ₂ ² + A ₁ A ₂ cos (θ ₂ −θ ₁ + β ₀ )} (21) Cs = + 2A ₀ ε ² α ₀ B ₁ B ₂ + 2B ₀ ε ² α ₀ A ₁ A ₂ (22) In the above, the correlation signals Cc and Cs are α of the signal C (t).
(T) and β (t) are shown by inverting the signs of the partial differential coefficients in the vicinity of α ₀ and β _0, respectively. The signal C (t) has a downwardly convex shape as shown in FIG. 4 as a function of β ₀ , and β m exists that gives a minimum value. The sign of the partial differential coefficient differs depending on whether β ₀ is larger or smaller than β m. By inputting the signal Cs to the loop filter 21, the control signal β ₀ gradually approaches the optimum value β m. Similarly, the value of α ₀ is also the optimum value α m
Approach. When both α ₀ and β ₀ are optimal values,
A ₀ ² B ₀ ² is zero, that is, A ₀ or B ₀ is zero. As is clear from the defining equations (19) and (20), these A ₀ and B ₀ are the desired wave (first
2) and the square of the amplitude of the interfering wave (second term). Therefore, in the above receiver, the signal combining circuit 8
It has been shown that the output signal Z (t) of 1 operates so that the interference wave or the desired wave is canceled. As a result, only the desired wave can be taken out at the output terminal 11 of the FM receiving circuit 9. There is no problem when the interference wave is canceled, but when the desired wave is canceled, a means for forcibly changing the system to another state is required. For that purpose, the most effective method is to invert the sign of the phase control amount β ₀ as conventionally known.

As described above, a method of obtaining a correlation by superimposing a sinusoidal minute signal in order to obtain the minimum value of the evaluation function is known as a perturbation method. The frequency ω of the perturbation signal is preferably set to 100 Hz to 200 Hz in the case of an analog voice signal. This value is determined by the condition that the fading frequency that normally occurs in mobile communication is several tens of Hz, so that it is higher than this and that it is lower than the lower limit 300 Hz of voice.

Further, in the above embodiment, it was assumed that the high-pass filter 12 can sufficiently attenuate the signal component of the frequency ω, but when this is not satisfied, it is desirable to provide a band-stop filter of the frequency ω together with the high-pass filter 12. .

〔The invention's effect〕

As described above, according to the present invention, attention is paid to the fact that a beat component is generated in the amplitude when an interference wave interferes with a frequency-modulated desired wave, and the amplitudes of two received signals are adjusted so that the beat component becomes zero.・ Because the phase relationship is controlled to obtain a desired signal, an omnidirectional receiving antenna is used, and no reference signal is required.
In addition, even if the level difference between the desired wave and the interfering wave is small, the desired wave can be taken out, which is particularly effective as a receiver for mobile communication.

[Brief description of drawings]

 1 is a block diagram showing a configuration of a receiver according to the present invention, FIG. 2 is a block diagram showing a specific configuration of a complex multiplication circuit, FIG. 3 is a block diagram showing a specific configuration of a correlation circuit, and FIG. The figure is a conceptual diagram for explaining the operation. 1 …… Complex multiplication circuit 2,3 …… Reception antenna 4,5 …… Subtraction circuit 8 …… Signal synthesis circuit 9 …… FM reception circuit 10,13 …… Square detection circuit 12 …… High-pass filter 14, 15 …… Correlation circuit 18 …… Low frequency signal generation circuit

Claims (1)

[Claims] 1. A first receiving a composite wave of a desired wave and an interfering wave
And a second receiving antenna, and a variable gain amplifier and a variable phase shifter connected to at least one of the first and second receiving antennas. The variable gain amplifier and the variable phase shifter have first and second variable gain amplifiers. A complex multiplication circuit for controlling the amplitude and phase relationship between the reception signals of the first and second reception antennas by inputting two control signals, and outputting two output signals, and this complex multiplication circuit A signal synthesizing circuit for synthesizing the two output signals, a receiving circuit for inputting part of the output of the signal synthesizing circuit for demodulation, and a first square inputting part of the output of the signal synthesizing circuit. A detection circuit and a high-pass filter for inputting the output of the squared detection circuit,
A second square-law detection circuit that inputs the output of the high-pass filter, a low-frequency signal generation circuit that generates first and second orthogonal low-frequency signals, and an output of the second square-law detection circuit. A first correlation circuit for obtaining a correlation between a signal and the first low frequency signal; an output signal of the second square detection circuit; and the second
Second correlation circuit for obtaining the correlation of the low-frequency signal and the signal obtained by subtracting the first low-frequency signal from the output signal of the first correlation circuit, as the first control signal, and the complex multiplication. A first subtraction circuit for outputting to a variable gain amplifier of the circuit;
A second subtraction circuit for outputting a signal obtained by subtracting the second low frequency signal from the output signal of the second correlation circuit to the variable phase shifter of the complex multiplication circuit as the second control signal; An adaptive antenna receiver for mobile communication, comprising: