JPH02305135A - Space diversity reception system for multi-value quadrature amplitude modulation - Google Patents

Abstract

PURPOSE:To attain the transmission with high quality and simple hardware constitution in a line having a rapid fading fluctuation by integrating fading distortion compensation and maximum ratio synthesis space diversity synthesis in the multi-value quadrature amplitude modulation system. CONSTITUTION:A transmission line distortion estimate section 29 estimates a transmission line distortion of each branch from a reception base band signal in a frame timing. A phase compensation section 30 based on an obtained phase fluctuation estimate value compensates the phase distortion due to fading of a reception signal of each antenna to make phases of signals of each antenna in-phase. A synthesis section 31 applies weighting proportional to the amplitude fluctuation of each antenna due to fading obtained in this way and applies synthesis. A decoding section 32 calculates the discrimination threshold level after synthesis to decode a data. Thus, the transmission with high quality and simple hardware constitution is attained.

Description

DETAILED DESCRIPTION OF THE INVENTION (1) Field of Industrial Application The present invention provides fading distortion compensation and spatial diversity reception functions when applying a multilevel orthogonal amplitude modulation method to a fading line where the transmission path fluctuates rapidly. The present invention relates to a reception method that has the following features.

(2) Conventional technology In digital wireless lines, especially land mobile communication lines, the amplitude and phase of received waves vary due to the influence of fading.
fluctuate very quickly.

Conventionally, in such lines, a frequency modulation or phase modulation method has been adopted in which the envelope does not contain information, taking into account that the envelope fluctuates by 20 dB or more.

However, in order to further improve the efficiency of frequency use, it is necessary to apply a multilevel orthogonal amplitude modulation method that also includes information in the amplitude. However, in the multilevel orthogonal amplitude modulation method, since data is included in the amplitude and phase of the modulated signal, highly accurate fading distortion compensation is required.

In addition, in order to further compensate for the significant deterioration in transmission quality due to fading, two or more spatially separated antennas are used to combine or select the signals of each antenna to reduce the probability of a drop in the level of the received wave. It is known that a spatial diversity reception method is effective.

Therefore, in land mobile communications, in order to effectively utilize frequencies and achieve high transmission quality, it is effective to use a multi-level orthogonal modulation method that applies a highly accurate fading distortion compensation method in combination with a spatial diversity reception method. It is.

The signal combining method most often used in spatial diversity is the selective combining method.

As shown in Figure 1, the selective combining method uses the signals of each branch (each antenna during divergent reception) to
This method selects the one with the highest received bell. Using this method, a gain of about 6 dB can be obtained compared to the case without diversity.

On the other hand, rather than selecting the received signal of each antenna,
As shown in the figure, a method is used in which the signals of each branch are made in phase and then weighted and combined so that the S/N after combination is maximized (maximum ratio combining method). Then, a gain of about 1.5 dB can be obtained.

There is also an equal gain combining method in which the weights of the signals of each part lunch in the maximum ratio combining method are made equal to simplify the circuit. The characteristics of the equal gain combining method are better than the selective combining method, but worse than the maximum ratio combining method.

Therefore, when diversity reception is performed, maximum ratio combining has the best characteristics.

However, the equal gain combining method and the maximum ratio combining method require phase control and signal weighting in the intermediate frequency band (IF band) to bring the signals of each branch into the same phase, and compared to the selective combining method, The hardware scale becomes considerably large. On the other hand, since it can only provide a maximum gain of 1.5 dB compared to the selective synthesis method, it has hardly been put into practical use so far.

(3) Problems to be solved by the invention Applying the multilevel orthogonal amplitude modulation method to wireless communication lines with severe fading fluctuations, such as land mobile communication,
In addition, in order to ensure high transmission quality, it is necessary to compensate for amplitude and phase distortion due to fading and to realize diversity combining using maximum ratio combining with high accuracy. Furthermore, in order to make such hardware practical, the hardware needs to be simple.

(4) Means for solving the problem In order to realize maximum ratio combining diversity while compensating for transmission line distortion, it is necessary to follow the steps 1) to 3) below.

■) Estimate amplitude and phase fluctuations due to fading in the antenna.

2) Based on the estimated value of the phase fluctuation obtained in 1), compensate for the phase distortion due to fading of the received signal of each antenna, and make the phases of the signals of each antenna the same.

3) Weighting is performed in proportion to the amplitude fluctuations due to fading of each antenna obtained in ①), and the signals are combined.

4) Calculate the judgment threshold after combination and decode the data.

Therefore, in the present invention, the transmitting/receiving section has the following configuration.

1) The transmitter transmits a known symbol once every N-1 information symbols (N is a natural number) for measuring channel distortion. Therefore, when transmitting the same amount of information, the transmission band is
Compared to the conventional method, the number is N/(N-1) times. FIG. 3 shows the frame structure when frame symbols are inserted.

2) The receiving section first detects symbol timing and frame timing.

3) Since the frame symbol is a known signal, it is used to measure the amplitude and phase distortion of each branch at the frame timing.

4) In each branch, the amplitude and phase distortion in symbols other than frame symbols (information symbols: symbols in which information is transmitted) are estimated by interpolating the amplitude and phase distortion at frame timing.

5) Based on the phase distortion estimated in 4), compensate the phase distortion of the received signal of each branch and bring it into phase.

6) The signals of each branch are weighted in proportion to the amplitude value estimated in 4) and then combined.

7) Based on the information in 4), calculate a threshold value for data judgment.

8)6),? ), the transmitted symbol is estimated, and the data is recovered by decoding the signal.

(5) Effect Figure 4 shows 1 as a typical example of the multilevel orthogonal amplitude modulation method.
A 6QAM signal space diagram (indicating signal points of a complex baseband signal on a complex plane) is shown.

16QAM is a method of arranging signals at equal intervals on a complex plane, as shown in FIG. In addition, the amount of information contained in one symbol in M-value QAM is: IQg2(M)
It's a bit. Therefore, in the case of FIG. 4, one symbol contains 4 bits of information.

FIG. 5 shows the configuration of the transmitter. First, data is divided bit by bit in a serial/parallel converter (11), and then converted into a corresponding complex baseband signal in a baseband signal generator (12).

Next, a frame synchronization insertion unit (13) inserts a frame symbol (known) once for every N-1 information symbols for transmission path distortion measurement.

When measuring channel distortion using frame symbols, in order to improve estimation accuracy, it is necessary to
It is necessary to increase N. Therefore, as a frame symbol, the points giving the maximum amplitude (Fig. 4 A, B, C
,D) is appropriate. In the following, point A (3+J·3) will be used as a frame symbol.

Thereafter, the signal is band-limited in a transmission filter FgJ (14), modulated in an orthogonal modulation section (15), power amplified in an amplification section (16), and then transmitted from an antenna section (17).

The 16QAM transmission signal x(t) after the above operations is
It is written as the following formula.

x(t)=a+ (t)cos(c,>t)−ao(
t) sin(c, + t) (1) Here, at(t): in-phase component aa(t
): Orthogonal component of the transmitting baseband signal ω: Spilling angular frequency. Furthermore, at(t) and aQ(t) are waveforms band-limited by the transmission filter section (14).

Here, let L be the number of branches in the receiving section. Further, in FIG. 6 showing the configuration of the diversity receiving section, the case where the number of branches is two is shown to make the drawing easier to understand.

Note that if the number of branches is three or more, the number of branch units (34) in FIG. 6 may be increased.

The antenna section (21) receives the signal, and the reception filter section (
After removing out-of-band noise in 22), AGC (A
automatic Ga1n Controller)
In the section (23), the signal is amplified to an appropriate level. Also, A
FC (Auto-matic Frequency)
Controller) gPJ (24) uses quasi-synchronous detection (does not regenerate the carrier wave and uses local oscillation 1 (2) of the receiver).
6) Compensate for the difference (offset frequency) between the progressive frequency and the local oscillation frequency when using the detection method (detection method).

AFC part (
The output signal yp(t) of 24) is expressed by the following equation.

yp(t)=rp(t)sr(t)cos(ut+θ
1)(1))-rp(t)so(t)sin(c,>
t+θp(t)) (2) However, rp(t): Amplitude fluctuation θp(t) of the p-th branch
: Let it be the phase fluctuation of the p-th branch. Also, sr(t
), 5o(t) are baseband waveforms after being band-limited by the reception filter section (12).

Thereafter, a quasi-synchronous detection section (25) performs orthogonal detection to obtain a received complex baseband signal up (t).

The received complex baseband signal up(t
) is up(t):up+(t)+,j-upo(t):j”
, (t)exp(j ・Cp(t)Xs+(t)+j
−so(t))=cp(tXs+(t)+j−so(t
)) (3) becomes. However, Cp (
t) is the fading distortion of the p-th branch, cp(t)=cp+(t)''J-cpo(t)=rp(
t) exp(j・θ0(t)) (4
). Therefore, uO(t) includes the transmission symbol and co-fading distortion.

Next, using up(t) (1≦p≦L), the clock is recovered in the clock recovery section (27) and the frame timing is recovered in the frame detection section (2nd).

Clock timing can be obtained not only from the complex baseband signal but also from the envelope of the received wave.

Since the frame symbol is the signal point with the maximum amplitude, u
p(t) periodically includes the maximum amplitude. Therefore, by detecting the timing, the timing of the frame symbol can be detected.

Further, the clock and frame timing can be reproduced by using all the received complex baseband signals up(t) of each branch, or by selecting an arbitrary branch.

The transmission path distortion estimator (29) estimates the transmission path distortion of each branch from the received baseband signal at the frame timing. The method is as follows.

First, the reception timing of the q-th frame symbol is
Let tq=qNTs. Here, T is one symbol length. At that time, the received complex baseband signal u, (tq) of the p-th branch is expressed as Lip(tq
)"ljp + (tq')+J-upo(tq)
= (3+j・3)cp(tJ (
5). Therefore 1=1. The estimated value of co(t) in (1) is
(6) is obtained.

On the other hand, εp(tq) is cp(t) in NT5 (sec)
This corresponds to sampling at intervals. Therefore, cp
(t) is set to one interval or less and an interpolation method is used to obtain the estimated value of cp(t) in the information symbol.

Interpolation methods include Newton's formula and Gauss's formula.
There are several formulas etc. Here, as an example, the quadratic G
A method using the auss interpolation formula will be described.

Let the transmission path distortions of the p-th branch be c, 1, co2, c, and 3, respectively, when t=NT, 2NT, 3NT9 is the reception timing of the frame symbol. Also,
It is assumed that the frame symbol insertion interval is sufficiently smaller than the Nyquist interval determined by the band of Cp(t).

In that case, cp(t) at 2NT, ≦L≦3NT
can be interpolated with a quadratic function as follows.

Cp3-Cpl “　　　　　　(j-2NTs)+Cp2(7)2NT.

=F, (t)exp(υ, (t))
(8) When the transmission path fluctuation is very slow relative to the symbol rate, first smooth the frame symbols and S/N
It is also possible to worship inside the shrine after improving one's abilities.

After that, first, up(t) is multiplied by exp(-peak(t)), and the phase distortion is compensated for the signal vp(t):up(t)exp(-8p(t))=rp
(tXs+ (t)+j-so(t))
We obtain (9). Moreover, this makes the signals of each branch in phase.

Next, in the synthesis section, according to the following equation, vp(t) is added to Fp
(t), a diversity-combined signal 2(t) is obtained.

Further, the synthesis section judges the synthesis output and calculates a judgment threshold H(t) for decoding.

There are two possible data determination methods.

[After calculating method 1 and completely compensating the amplitude distortion, the threshold value is set to H(t)=0, ±2, ±2・j
(13) and the method of determination.

[Method 2] A method in which the amplitude distortion of 2(t) is not compensated for using the % formula %(14), and the determination threshold is set as follows.

Method 1 and method 2 are completely equivalent methods, so they can be selected as appropriate depending on the system. .

The above synthesis and judgment threshold calculation are performed by the synthesis section (31).
It will be carried out at

Thereafter, zl(t) and the decision threshold H(t) are transferred to the decoding section (32), the transmitted symbol is regenerated, and the information of the bits included in the symbol is regenerated.

The parallel/serial converter (33) converts the reproduced bit information into serial information and outputs it as reproduced data.

In the above receiver configuration, there is a transmission path distortion estimator (29) that performs fading distortion compensation and diversity combining, a phase compensator (30), a frame detector (2B), and a combiner (31).
is DSP (Digital Signal Pro
cessor), in which case the hardware becomes very simple. Further, this method can be applied as is to other multilevel QAMs.

Note that if Equation (10) is expressed as follows, the selective synthesis method can be realized. In addition, if (■) is set, equal gain synthesis can be realized. In this case as well, the same two types of ideas as in the case of the maximum ratio combination method can be applied as the data judgment method.

(6) Effects of the Invention By using the present invention, it is possible to construct a maximum ratio combining type diversity receiver using a multilevel orthogonal amplitude modulation method with a simple hardware configuration.

FIG. 7 shows the error rate characteristics of 160AM when the present invention is applied. However, the maximum Doppler frequency (fd) is 8
It is 012. In addition, the maximum ratio combining method (Max, Rati
o) In addition to selection synthesis method (Selection), equal gain synthesis? 1 (Eq, Ga1n), and the characteristics without using duversity (Without Div,) are also shown.

Error rate (BER: Bit Error Rate) is 1
When compared at the point of 0-2, the characteristics of the selective combining method, equal gain combining method, and maximum ratio combining method are 6 dB, 6.9 dB, ?, respectively, compared to the case without diversity. . 5d
It can be seen that there is a gain of B. These gains almost match the theoretical values. Further, although FIG. 7 shows the characteristics in the case of fd: 80H2, the characteristics are almost the same in the case of other values.

[Brief explanation of drawings]

Figure 1 is a configuration diagram of a selective combining diversity receiver, Figure 2 is a configuration diagram of a maximum ratio combining diversity receiver, Figure 3 is a frame configuration diagram during transmission, and Figure 4 is a 16QA
FIG. 5 is a block diagram of the transmitting section, FIG. 6 is a block diagram of the receiving section, and FIG. 7 is the error rate characteristic of 16QAM to which the present invention is applied. DESCRIPTION OF SYMBOLS 1... Antenna part, 2... Receiver, 3... Reception level comparison part, 4... Switching part, 5... Reception level detection part, 6... Phase detection part, 7...・Phase adjustment unit, daytime・Gain adjustment unit, 9◆・・Composition unit, 11・・Serial/parallel conversion unit, 12・・◆Baseband signal generation unit, 13・・Frame symbol insertion unit, 14 ◆◆・
Transmission filter section, 15... Quadrature modulation section, 16... Amplification section, 17... Antenna section, 21... Antenna section,
22... Reception filter section, 23... AGC section, 24
...AFC section, 25... Quasi-synchronous detection section, 26...
Local oscillation unit, 27...Clock regeneration unit, 2nd...Frame detection unit, 29...Transmission path distortion estimation unit, 30...
Phase compensation unit, 31... Combining unit, 32... Decoding unit, 3
3...Parallel/serial converter, 34...◆Branch unit. Patent Applicant: Director, Communications Research Institute, Ministry of Posts and Telecommunications 331 Figure 1 Symbol 1 Symbol l Shin A
< At Figure 3 Q-ch Figure 4 Figure 7

Claims (1)

[Claims] In the multilevel orthogonal amplitude modulation method, (1) a conventional transmitting section having a serial/parallel converting section, a baseband signal generating section, a transmission filter section, an orthogonal modulating section, an amplifying section, and an antenna section; a transmitting section with a frame symbol insertion section that inserts a known frame symbol for measuring time-varying fading distortion on the receiving side every N-1 information symbols; (2) antenna section, reception filter section; , AGC Department, AFC
The quasi-synchronous detection section detects the received baseband signal using the conventional quasi-synchronous detection waveform demodulation section. A fading distortion estimator estimates fading distortion by interpolation, and a phase distortion component obtained in the fading distortion estimator is used to compensate for phase fluctuations due to fading and to adjust the signal of each branch. (3) A local oscillation unit for quasi-synchronous detection; (4) A clock recovery unit and frame unit that recovers a timing signal and frame timing from a received baseband signal. (5) Based on the amplitude information obtained in the fading distortion estimator of each branch, the signals whose phase distortion has been compensated for in each branch are weighted and synthesized, and for data judgment. (6) a decoding unit that determines data based on the output of the synthesis unit; and (7) a parallel/serial conversion unit that converts the decoded parallel data into serial data. By integrating fading distortion compensation and maximum ratio combining type spatial diversity combining in the multilevel quadrature amplitude modulation method, it is possible to use a line with severe fading fluctuations with a simple hardware configuration. A spatial diversity reception method for multilevel orthogonal amplitude modulation, which is characterized by the ability to achieve high-quality transmission.