JPS6359044A - System for reducing inteference wave noise - Google Patents

Abstract

PURPOSE:To reduce the effect of interference wave noise by using two carriers in phase and orthogonal to a biphase PSK wave so as to detect them individually and using a detection output of an opposite phase component so as to apply signal processing to the detection output of the in phase component. CONSTITUTION:A thermal noise and an interference wave noise subjected to band limit by a band pass filter 1 has equal power spectrum in the in phase and orthogonal components with respective signal. In utilizing the property above, a signal 11 enters a carrier recovery circuit 4 and a carrier omegaC is recovered. A cosomegaCt is outputted on a line 41 and a sinomegaCt is outputted to a line 31 via a pi/2 phase shifter 3. The two carriers in the orthogonal relation are multiplied with a signal 11 by two multipliers 2, 2'. Base band signals 51, 52 are given to A/D converters 6, 6', where the signals are digitized and given to FET circuits 7, 7', and they are converted from frequency axis signals into time base signals in real time. Output signals 71, 72 of the FET circuits 7, 7' enter a signal processing circuit 8, where they are processed.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a method for reducing interference wave noise in the same frequency band in a communication method using a wireless line.

(Prior Art) In recent years, with the development of wireless communications, the frequency bands allocated to wireless communications have become extremely congested.

In particular, 6/4 GHz is used in fixed satellite communications.

The 14/11GHz band is heavily congested because it shares the same frequency band as terrestrial microcommunications. In such frequency bands, countermeasures against co-frequency band interference noise have become a major problem. The interference wave noise reduction methods that have been considered so far can be broadly classified as follows.

(1) Low side lobe of the antenna (2) Shielding method (3) Interference wave compensation circuit method Here (11f21 is a method to reduce the interference waves input to the antenna as much as possible.In contrast, (3)
In this method, input interference waves are canceled by signal processing within the receiver.

(Problems to be Solved by the Invention) These conventional techniques are subject to limitations such as the large size of the receiving device and the fact that the direction of arrival of the interference wave is known for +11 and +21. In the method (3), it is necessary to independently separate only the interference wave added to the signal wave. Here, as a conventional method of independently separating only the interference waves, a method of providing an auxiliary antenna that receives only the interference waves in addition to the antenna that receives the signal waves to be received has been considered. However, there are restrictions such as the fact that the direction of arrival is known. Therefore, if there are multiple interference waves, it is necessary to provide as many auxiliary antennas as there are interference waves.

As described above, the conventional interference wave noise reduction method has the disadvantages that the receiving device becomes large and the direction of the interference wave source must be known in advance.

(Means for Solving the Problems) The present invention has been made in view of the above-mentioned prior art, and provides an interference wave noise reduction method using signal processing that does not require any external equipment such as an auxiliary antenna. . The feature is that a two-phase PSK wave (not a quadrature phase modulated wave) is detected separately using two carrier waves that are in phase and orthogonal to that signal, and the thermal rays contained in the two output signals obtained by the detection are Utilizing the property that the power spectra of audio and visual interference waves are equal to each other, signal processing can be performed by using the detected output of the negative phase component as the detected output of the in-phase component? The purpose of this addition is to completely reduce the influence of interference wave noise (there is no signal in the quadrature output, only noise equal to the in-phase output).

(Structure and operation of the invention) Hereinafter, the present invention will be explained in detail using the drawings.

FIG. 1 shows an embodiment of a receiver using the interference reduction method according to the present invention. In Figure 1, 10 input terminals include thermal noise expressed by a linear equation, interference wave noise, and two-phase PSK.
A signal obtained by combining signal waves is received.

rlft)=5it)tenn(t)+i(t)
-tl) However, s (t) is a two-phase PSK wave,
n(t) indicates thermal noise, and 1(t) indicates interference wave noise.

Here, even if i (t) is a composite wave of a plurality of interference wave noises, generality is not lost in the following operation. The received signal r(t) shown in equation (1) has a frequency characteristic B(ω) of 1
The signals shown below pass through a band pass filter 11. Let ω be the center frequency of B(ω). shall be.

Here ω. is the carrier frequency.

r2(t)=m(t) cosωct + n((t)
cqsωct + n8(t) sinω. t+i
(B(t) cosωct + 18(t) sin
ωct□(2) However, m(t) is information and takes ±it. Also, nc(
t) and n8(t) are narrowband pass stochastic processes of thermal noise band-limited by a bandpass filter of 1, and have the relationships shown in the following equations.

nc' (t) = n,z (t)
(3) pnctω)-pns(ω)=Pn(
ω) (4) However, -nc”(t), n3”(t) is 2
The root mean value is shown, and Pne(ωl, Pns(ω) is nc(
The respective power spectra of tl and n5(t) are shown. Similarly, the following relationship holds true for interference noise i (t).

l(:”(t)-Is”(t)
(51Pio(ω)-Pi8(ω)=Pifω)
-□+61 In this way, the thermal noise and the interference wave noise, which have been band-limited by the bandpass filter l, utilize the property that the power spectra of the respective signals, the in-phase component, and the orthogonal component are equal to each other.

Then, all in-phase components and orthogonal components are extracted as follows.

The signal 11 expressed by equation (2) enters the carrier wave regeneration circuit 4 and is converted into a carrier wave ω. is played. And in 41, eO8ω
ct is output, and sinωct is output to 31 via the π/2 phase shifter 3. These two orthogonal relationships
The two carrier waves are multiplied by the signal 11 by two multipliers 2, 2'. As a result, the following signals 21 and 22 are obtained at the output of the multiplier 2゜2'. Signal of 021: re(tl=r2(t) Signal of eO8ωct22: r3(tl-r2(t) sinω. Then 21, 2
The signals 2 pass through a 5.5' low-pass filter to remove harmonic components (2ω. components), and signals expressed by the following equations pass through 51 and 52, respectively.

51 signal: 52 signal; Baseband signal 5 shown by formula +9+, (10)
1 and 52 pass through a 6.6' A/D converter, are digitized, and are converted from time-domain signals to frequency-domain signals in real time by a 7.7' FFT circuit.

The output signals 71 and 72 of the FFT circuits 7 and 7' enter a signal processing circuit 8, where the following signal processing is performed.

The signal processing circuit No. 8 realizes frequency characteristics such that the square error between the baseband signal 51 shown by equation (9) and the signal wave m(t) is minimized. The role of the signal processing circuit 8 will be explained in detail below.

The amplitude spectrum of the output of the FFT circuit 71 is expressed by the following equation.

Ro(ω) − M(ω) ten Nc(ω) + IC(ω)
(11) However, M(ωl, No(
ω) and Ic(ω) are respectively m(t).

The frequency spectrum of no(t) and 1o(t) is shown.

The frequency characteristics realized by the signal processing circuit 8 are expressed by a full-order equation.

W(ω)−A(ω)eJβ((′)
(12) However, A(ω) indicates the amplitude characteristic and β(ω) indicates the phase characteristic.

Therefore, the output signal when the signal in equation 01) passes through a characteristic having a constant transfer function is expressed by the following equation.

R, (a+1=R0fω)-W(ωi -□
13+Next, let the time axis waveform kTo(t) of equation Q3) and generally allow some time delay and calculate r. (11) is made to approximate the signal waveform m (-α) as much as possible. That is, the error e (t
l is expressed by the following formula.

e(t)=r. (t)-m(-α)□(矧e (tl
The amplitude spectrum of is expressed by the following equation.

E(ω)-R6(ω)-M(ω) e-Jo)a(15
) However, E(ω) indicates the amplitude spectrum of e (t).

Therefore, the root mean square value of the error is P + l NO(ω)121W(ω)12.

Here, the power spectrum of thermal noise and interference waves is expressed as Pn(ω
), Pi[ω), then Nc(ω), IC(ω)
By using the relationships of equations (4) and (6), the following relationship holds true.

Similarly, the power spectrum 2 p of the signal wave m [tl,
(If (=), then holds true.

By using the formula αη~09), the formula (161) becomes the following formula.

Next, from the equation qzω, A(ω) and β(ω) when the least square error eo(t)2 is minimum are given by the following equations.

β(ω)−1ωα □ (22) Therefore, .

Therefore, if the frequency characteristic shown by equation (23) is realized by 8 signal processing circuits, 71 and 81 in FIG. 1 are multiplied by 9 multipliers, and the output of 91 has the amplitude spectrum shown by equation Q31. R8(ω) is obtained. R8 obtained here (
ω) is a signal wave in which the square error between tl and r of 61 (1) is the smallest, that is, the influence of thermal noise and interference wave noise is the smallest.

Here, in the signal processing circuit 8, the following operation is performed in order to fully realize the frequency characteristic of equation (23).

From the amplitude spectra of the signals 71 and 72 in FIG. 1, the power spectra are determined as shown below.

71 signal=M(ω)+Nc(ω)+I. (ω)
(24) 72 signal −NS(ω)+Is(ω)
(25) However, Ns(ω
), I s(ω) is n 5(tl , i 5(t)
shows the amplitude spectrum of

When the amplitude spectra of equations (24) and (25) are squared, the power spectrum is expressed by the following equation using the fact that the signal, thermal noise, and interference wave are uncorrelated, and the relationship of equation +4)+6+.

P71(ω)=Ps(ωl+Pnfω)+Pi(ω)
-(26)P? 2 [ω)=Pn(ω)+Pi(a
l (27) Therefore, equation (26)
(27) '& use ps(ω1-P7+(
ω) −Pn (ωl −(28)).

From equations (26) and (28), equation (23) can be realized. Here, α in equation (23) is a constant,
It can be set to give the best signal quality.

As described above, in the signal processing circuit 8, Equations (26) and (27
) and (28), the frequency characteristic of equation (23) will be output. The signal processing here is
This must be done in real time with respect to the received signal.

Since these are signal processings of baseband signals, it is sufficient to perform sampling at least twice the transmission rate of two-phase PSK. Since the output signal 91 of the multiplier 9 is an amplitude spectrum, the IFFT circuit 10 (effect of the invention) As described above, according to the present invention, by performing quadrature phase detection by fully utilizing the properties of the two-phase PSK wave, Interference can be easily removed.

Claims (1)

[Scope of Claims] An interference wave noise reduction method that receives a two-phase PSK wave including an interference wave noise component and removes the interference wave noise component by applying signal processing to the received signal, comprising: is detected with two carrier waves that are in phase and orthogonal to the carrier wave of the two-phase PSK wave to obtain an in-phase component and a quadrature component of the signal, and then convert these two components into power spectrum signals, respectively, and Using the property that the power spectra of interference wave noise components appearing in two power spectrum signals are equal to each other, a frequency characteristic that suppresses the power spectrum of the interference wave noise component is obtained from the two power spectrum signals, and the frequency characteristic is An interference wave noise reduction method characterized in that the interference wave noise component is reduced by multiplying the power spectrum signal of the in-phase component by the power spectrum signal of the in-phase component.