JPH05136759A - Spread spectrum demodulator - Google Patents

Abstract

PURPOSE:To eliminate an interference wave by passing one of halved reception signals through a variable phase shifter, passing the other through a circuit whose phase is unchanged, synthesizing both signals and controlling the phase shift so that the sum output is minimized. CONSTITUTION:A signal inputted from a input terminal 6 is a superimposed wave between a signal component (modulation wave subject to spread spectrum) and an opposite continuous wave noise, and the input signal is halved by a branching device 1. Then the one is subject to phase change by a variable phase shifter 2 and the other is outputted directly or via a circuit having a loss equivalent to the phase shifter 2 and both signals are synthesized by a multiplexer 3. An output of the multiplexer 3 is detected by a detector 4 and fed back to the variable phase shifter 2 through a controller 5 to control the phase shift of the variable phase shifter 2 so that the detection output is minimized. The signal synthesized by the multiplexer 3 is sent to an inverse spread spectrum section via a terminal 7. Thus, a constant wave(CW) interference and other interference waves are eliminated and the signal is received with excellent C/N.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a demodulator used for direct spread spectrum communication, in which interference is eliminated.

[0002]

2. Description of the Related Art Recently, communication using spread spectrum is
Various applications have been considered due to its various characteristics, that is, multipath characteristics, continuous wave (CW) interference, random access, confidentiality, and the like.

In the direct spread spectrum communication system,
By using a spreading code having a specific cycle (the cycle is represented by the number of chips), the modulated wave and the spreading code are multiplied to spread (spread spectrum) in a wide frequency band and transmit. 5A and 5B are block diagrams of a transmitter and a receiver in spread spectrum communication, respectively.

In the transmitter of FIG. 5 (a), it is assumed that the information transmission rate is B _{0 bps} , the spreading code is 127 chips, one bit of information is directly spread by 127 chips, and BPSK modulation is performed. In the transmitter, the information generation unit 12 has B
A signal of _{0 bps} is generated and sent to the spread spectrum unit 13.
Here, 1 bit of information is multiplied by 127 chips of spreading code. Specifically, processing is performed using exclusive OR or the like. The transmission signal generated at this time is 127 × B _{0 bps}
It becomes the amount of information of.

This signal is generated by the BPSK modulator 14 at the next stage by converting the carrier from the carrier generation source 18 into BPSK.
Modulate. Then, it becomes a signal occupying a band of 2 × 127 × B ₀ (Hz), and is wirelessly transmitted from the transmitting antenna 17 through the band pass filter (BPF) 15 having this band and the output amplifier 16.

In the receiver shown in FIG. 5B, the receiving antenna 1
9, the signal is amplified by the reception amplifier 20, passed through the BPF 21, and then, the spectrum despreading unit 22 performs despreading using a means such as a delay lock loop (DLL) or a matched filter, and then the BPSK demodulation unit 23. And is decoded by the information decoding unit 24 to obtain an information signal.

As described above, the band 2 × B ₀ (Hz), which is originally necessary for transmitting information, is multiplied by 127 to be 2 × 12.
By expanding to 7 × B ₀ (Hz), the signal is transmitted over a wide band, so that it has resistance to frequency-dependent multipath and continuous wave (CW) interference appearing at a specific frequency. In addition, since the receiver can despread and demodulate only to the receiver having the same spreading code, random access is possible and confidentiality is provided.

Generally, the gain obtained by this despreading is called a processing gain, and in the case of 127 chips, it is 1 in power.
A diffusion gain of 0log127 = 21 dB will be obtained.

[0009]

In such a spread spectrum communication, the band used during transmission and reception is extremely wide, for example, 127 times the normal BPSK modulation. Therefore, the CW from the outside is occasionally included in the band. Noise may be mixed in. 1 in simple calculation compared to the past
27 times more likely. Therefore, CW noise is often a problem.

The CW noise has a single spectrum and its power is W _cw, and the spread spectrum communication signal has a band B _ss and its power is W _s .

In the case of general digital modulation, the noise W _cw in the signal band becomes noise as it is, but in the spread spectrum, the energy after despreading is 2 × 127 ×.
Assuming that they are evenly distributed over B ₀ (Hz), about 1/1
27W _cw becomes noise.

Therefore, if CW noise having a power of W _cw = 10 W _s exists, the spectrum despreading unit 2
The noise after two passes is about 10/127 W _s, which means that the carrier-to-noise ratio C / N is strong against CW noise.

[0013] However, if the CW noise is strong much, for example in the case of such W _cw ≧ 127W _s are noise after the spectrum despreader 22 passes, it becomes W _s equal to or greater than, C / N is very It gets worse and falls into a situation where communication is not possible.

There are few practical means for avoiding this, and one proposed example is described in 1985 Ultrasonic Symposium Proceedings P108-113.
This will be briefly described below.

The input signal is considered to have a form in which two signals, that is, the spread spectrum communication wave and the CW signal are superimposed, and is represented by the sum of the two signals.

In the above-mentioned proposed example, this is subjected to a fast Fourier transform to transform the input signal from the time axis to the frequency axis, and then the frequency component causing the CW interference is steep notch filter (BRF: band). It is a method to prevent CW interference by removing it with a rejection filter).

In this case, the frequency of the CW noise can be surely grasped by performing the Fourier transform, so that
Even if W _cw << W _s , the CW noise can be detected and removed, so it is considered to be an ideal CW interference removal method.

However, in reality, it is difficult to perform the fast Fourier transform, and the applicable transmission rate is limited to an extremely low transmission rate. Also in that case, the hardware and software required for the Fourier transform become large.

[0019]

In the demodulator of the present invention, the received direct spread spectrum signal is demultiplexed into a channel passing through a variable phase shifter on the one hand and a channel whose phase is not changed on the other hand, and both channels are demultiplexed. The signals of (1) and (2) are combined and the amount of phase shift of the variable phase shifter is controlled so that the total output is minimized.

[0020]

When the received signal is divided into two and one is subjected to phase conversion, the other is left as it is or after the necessary processing is performed, the two are combined and the phase shift of the variable phase shifter is performed. 180 ° phase at specific frequency depending on quantity
It can be shifted so that the combined output becomes zero. It also reduces its combined output before and after a certain frequency. When the CW noise has a great influence on the C / N, the CW signal is much larger in terms of electric power, so minimizing the total electric power makes the phase 1 for the CW noise.
It means that they are shifted by 80 °. As a result, CW noise can be reduced and the overall C / N can be greatly improved. Other interference waves can be reduced.

[0021]

1 is a block diagram of an embodiment of the present invention provided in a spread spectrum demodulator.

In FIG. 1, the signal input from the input terminal 6 includes a signal component (spread spectrum modulated wave) and
The CW noise (single spectrum) is a superposed wave of two waves, and other noises (white noise etc.) are not considered.

This input signal is split into two in the demultiplexer 1, one of which has its phase changed in the variable phase shifter 2 and sent to the multiplexer 3. The other is sent to the multiplexer 3 as it is or after passing through a circuit corresponding to the loss of the phase shifter 2.
These two signals are combined in the multiplexer 3, and the output is detected by the detector 4 and fed back to the variable phase shifter 2 via the controller 5, and the variable phase shifter 2 is set so that the detection output becomes the minimum. To control. The signal multiplexed by the multiplexer 3 is sent to the spectrum despreading unit 22 via the terminal 7.

Now, the characteristics of the variable phase shifter 2 will be examined. A variable phase shifter is one that changes the electrical length, and here we consider a variable phase shifter that does not have frequency dependence. Therefore, the relationship between the frequency and the phase delay when the electric length is constant is monotonically increasing as shown in FIG. When a signal of S ₀ = sin ωt is input to this variable phase shifter, this output signal is represented by sin ω (t−t ′) (t ′: delay due to electrical length).

As described above, when the signal is divided into two and one is output to the variable phase shifter and the other is output as it is, and multiplexed again,
The output is summed as follows:

[0026]

[Equation 1]

Consider this situation from the output power. When the load is R, the output power W _out is expressed as follows.

[0028]

[Equation 2]

This state is shown in FIG. As ω changes,
The output voltage changes sinusoidally, and the value shown by the broken line is the maximum value, that is, the output voltage when there is no variable phase shifter.

Further, although there is a point where the power becomes zero for a certain single frequency, in the case of a wideband signal, when viewed in a long period, this sinusoidal average power is obtained, and the power is about 63. It will be attenuated to 7%.

The reason why the CW interference can be eliminated by using the variable phase shifter in the present invention will be described below.

It is assumed that the signal input from the input terminal 6 is a spread spectrum signal having a bandwidth of B _ss and spread over 127 chips, and the input power is W _s . Also, originally BPSK
The modulated wave and the like do not have a uniform power spectrum distribution in the band, but here it is assumed that it spreads uniformly in the B _ss band for simplicity.

On the other hand, the CW interference wave is B_ssIn-band, single
With the frequency of_cwAnd At this time,
The signal power after despreading is W_s, CW interference power is W_cw/ 12
7, which can be represented as C / N. Shi
Therefore, C / N = 127W _s/ W_cwBecomes

[0034] Thus, when W _s ≒ W _cw, but CW jammer is not a problem, when the W _cw »W _s, the noise component is increased and problems.

By the way, the input power at this time is represented by W _s + W _cw because W _s and W _cw are independent.

In the present invention, when the phase shift amount of the variable phase shifter is arbitrarily set, since the W _s wave is a wide band signal, the output of the composite wave is W _{s out} = regardless of the phase shift amount.
It becomes 0.637 W _s . On the other hand, since the W _cw wave is assumed to be a single frequency signal as described above, the output changes according to the phase shift amount, and changes between 0 and W _cw . Therefore, the output of the composite wave can be expressed as W _cwout = f (t ').

Therefore, the output power of the multiplexer 3 is W
_{s out} + W _{cw out} = 0.637 W _s + f (t '). Therefore, when this output power is detected by the detector 4, the detected output becomes a function of f (t '), and CW
It becomes dependent only on the interference wave.

Therefore, the controller 5 controls the feedback from the wave detector 4 to the variable phase shifter 2 so as to minimize the output of the wave detector 4. Specifically, the minimum power is obtained when the phase shift amount has a relationship of (2n + 1) π with respect to the CW interference wave, and in this case, ideally, the CW noise can be completely removed.

In this case, C / N has no N, so
Ideally it will be infinite. As described above, according to the present invention, by controlling the output of the detector 4 to be the minimum, 100% of the CW noise in the narrow band is removed, and the spread spectrum wave in the wide band is uniformly reduced to 63.7%. As a result, the C / N ratio is much improved as compared with the conventional example.

Further, as compared with the case where the calculation is performed by the conventional Fourier transform or the like, the output of the detector is simply controlled so as to be minimized, so that it is possible to cope with a high-speed signal,
At the same time, the hardware and software are much easier.

In this embodiment, the CW interference wave has a single frequency, but the effect of the variable phase shifter can be expected even if this is a narrow band signal like a general communication wave. The present invention does not lose its generality in such applications.

FIG. 4 is a block diagram of another embodiment of the present invention. The difference from FIG. 1 is that switches 8 and 9 are provided before and after the variable phase shifter 2, and the variable phase shifter 2 is disconnected from the circuit as necessary.

When the phase of the variable phase shifter 2 is changed and all expected phases are not lower than a certain arbitrary value, the switches 8 and 9 can be switched so as not to pass the variable phase shifter 2. ..

This embodiment will be described in detail below.
As described above, the output power is W _{s out} + W _{cw out} = 0.
673 W _s + f (t '). If W _{s out} ≤W
_{When cw out} , by changing the variable phase shifter 2,
The electric power changes greatly and the output can be reduced. On the other hand, if the output power of the variable phase shifter 2 does not change so much even if it is controlled, it is considered that f (t ′) has hardly contributed, and it is considered that W _{s out} >> W _{cw out} .

Therefore, in that case, the CW interference wave can be estimated to be at a level at which there is no problem, and since it is not necessary to pass the variable phase shifter, the changeover switches 8 and 9 stop the operation of the variable phase shifter 2. But no problem. In actual communication, the thermal power other than the CW interference wave and the spectrum power density of the BPSK wave are not uniform as assumed,
Since the spectrum of the signal wave is not evenly distributed due to multipath, etc., when there is no CW interference wave, the C / N may be better without the variable phase shifter 2. Therefore, by determining this arbitrary value by calculation, simulation, experiment, or the like by the line design, the variable phase shifter 2 is used to remove it when the CW interference is strong, and when the CW interference is weak or not. Since the variable phase shifter 2 is not used, a good C / N can be secured.

By using the present invention in this way, good C /
It becomes possible to communicate in the N state.

In the above-mentioned other embodiment, the changeover switch is used to prevent the variable phase shifter 2 from passing through. However, in addition, the phase shift amount of the variable phase shifter is set to zero, or It suffices if the method of equivalently eliminating the function of the variable phase shifter, such as using a phase shift amount that does not affect the band.

[0048]

As described above, according to the present invention, it is possible to remove the CW interference wave and other interference waves, which often become a problem in spread spectrum communication, and as a result, the C / N ratio is improved. You will always be able to receive.

Further, it can be applied to high-speed data communication, and its hardware and software are much smaller than the conventional method of performing Fourier transform to remove CW interference. realizable.

[Brief description of drawings]

FIG. 1 is a block diagram of an embodiment of the present invention.

FIG. 2 is a graph showing the relationship between frequency and phase delay of a variable phase shifter.

FIG. 3 is a graph showing the relationship between ω and output power.

FIG. 4 is a block diagram of another embodiment of the present invention.

5A is a block diagram of a transmitter, and FIG. 5B is a block diagram of a receiver.

[Explanation of symbols]

 1 Demultiplexer 2 Variable phase shifter 3 Combiner 4 Detector 5 Controller

Claims (2)

[Claims] 1. A means for demultiplexing a received direct spread spectrum signal into a channel passing through a variable phase shifter on the one hand and a channel which does not change the phase on the other hand, and the signals of both channels are combined and their total output is And a means for controlling the phase shift amount of the variable phase shifter so as to minimize the spread spectrum demodulator. 2. A means for activating the variable phase shifter when the combined output of both channels is detected, and when there is a phase shift amount where the output value is below a predetermined value. Spread spectrum demodulator.