JP3173881B2 - Automatic gain control method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a time division multiple access (T)
ime Division Multiple Access) Communication system (hereinafter TD
This is referred to as an MA communication system. ) Relates to an automatic gain control method for demodulating a received signal. In satellite communication using an artificial satellite such as a geostationary satellite as a repeater, a communicable service area is very wide, and a large number of stations within this service area can be easily connected in a mesh manner. As a communication method for connecting a large number of stations in a mesh shape, the TDMA communication method is excellent. In the TDMA communication system, transmission timing is controlled so that signals are transmitted from each transmitting station in a time-division manner, so that signals from different stations on a satellite do not overlap. On the other hand, on the receiving side, signals transmitted from each station are received in a time series in bursts.

[0002] Since the levels of these transmission signals are not always the same at all transmitting stations, the levels of burst-like reception signals differ from burst to burst. Also, even if the level of the transmission signal from each transmitting station is the same, if the propagation loss due to rainfall and the like differs depending on the path between each transmitting station and the receiving station, the level of the received signal will also depend on the receiving path for each burst. Will be different.

As described above, stable demodulation is desired even for a received signal having a different level for each burst. In a device in which code error correction is introduced, performance degradation due to level fluctuation occurs, and a problem is particularly large in a soft decision decoding method. As a method of determining received data, a so-called hard decision and a soft decision are known. In the hard decision, logic “1” and “0” are determined based on whether a demodulated baseband signal exceeds a certain threshold value. I do.

On the other hand, in recent years, a soft-decision Viterbi decoding method has been introduced for the purpose of dramatically improving the bit error rate of demodulated data (JP-A-60-145713 and JP-A-60-1987).
-72227). In this soft-decision Viterbi decoding method, the likelihood of the logic determined to be "1" or "0" is used as code error correction information. The accuracy of the determination of “probability” varies according to the variation of the level of the received signal. Therefore, automatic gain control in which the level of the demodulated baseband signal does not change is required.

[0005]

2. Description of the Related Art In a TDMA communication system, received signals are received in bursts, and there is no signal interval between the burst signals. Therefore, in the automatic gain control method (AGC) based on the normal feedback control, the gain at the head of the burst immediately after the no-signal section becomes very large, and waveform distortion occurs in the demodulated signal. That is, the demodulation performance deteriorates due to the transient response of the AGC pull-in. In order to solve this problem, a peak holding type AGC is conventionally used. This conventional example is shown in FIG.

Referring to FIG. 5, a conventional peak holding type automatic gain control device includes a voltage variable gain amplifier 51 for amplifying an input intermediate frequency signal, a capacitor C for holding a peak voltage of an output intermediate frequency signal, and an output intermediate gain signal. includes a diode D1, D2 for passing only high voltage signal from the voltage held in the frequency signal sac Chi capacitor C, a high resistor R, an operational amplifier 52 taking the difference between the holding voltage and the reference voltage .

FIG. 6 is a time chart for explaining the operation of the apparatus shown in FIG. As shown in FIG. 1A, input intermediate frequency signals are burst signals A, B, C,
A, ... In the illustrated example, the level of the burst signal A is the highest, and this level is held in the capacitor C as a peak detection voltage (e) as shown in FIG. 6 (2), and the burst signals B and C lower than the held voltage are held. Received, the holding voltage is not updated. However, when the bursts B and C are received, the holding voltage is discharged via the high resistance R, so that the peak holding voltage slightly decreases until the next burst A is received as shown in the figure. Since the decrease in the holding voltage due to this discharge can be ignored, the automatic gain control device of FIG. 5 controls the maximum amplitude among the signal amplitudes arriving in a burst form to be always constant. According to this conventional method, since the AGC is not pulled in every burst and operates so as to keep the peak burst level constant, waveform distortion due to a transient response when changing from a no-signal section to a burst-like signal, As a result, deterioration of the demodulation performance can be prevented.

[0008]

According to the above-mentioned conventional method, the burst signal having the voltage equal to or lower than the held peak voltage is output as it is, so that the level difference between the burst signals due to the difference in the transmitting station still remains. ing. With satellite communications, the service area can be very large,
As a result, the weather conditions of each earth station have little correlation,
The rainfall area is random, and the incoming power to the satellite varies from station to station. Therefore, in the conventional method of FIG. 5 in which the automatic gain control does not track every burst, there is a problem that the capability of soft decision Viterbi error correction cannot be sufficiently obtained.

An object of the present invention is to provide a stable demodulation by making the output level of a demodulated baseband signal constant even for a received signal having a different level for each received burst in view of the above-mentioned problems in the prior art. Is to make it possible.

[0010]

FIG. 1 is a block diagram showing the principle of the present invention. In the figure, a demodulation device of a time division multiple access communication system is shown, 4 is a variable gain amplifying means for amplifying a burst-like received signal, and 5 is a synchronous detection of an output of the variable gain amplifying means 4 with a demodulation reference carrier. 1 is an analog / digital converting means for converting a demodulated baseband signal obtained at the output of the synchronous detecting means 5 into a digital signal, and 2 is a digital signal which accumulates and holds a demodulation synchronization pre-signal portion. The accumulation holding means 3 includes an output signal of the accumulation holding means 2 and the analog / digital conversion means 1.
A product operation means for outputting a product operation with the output signal of the variable gain amplification means 4 for controlling the gain of the variable gain amplification means 4 in accordance with the difference between the amplitude information obtained at the output of the accumulation holding means 2 and the reference amplitude This is gain control means.

The analog / digital conversion means 1, the accumulation holding means 2, and the product calculation means 3 constitute a feed-forward type automatic gain control means. Corrected by 6,
Individual burst fluctuations are corrected by feedforward type automatic gain control means.

[0012]

The demodulated baseband signal received in a burst is converted into a digital signal, and the demodulation synchronization pre-signal portion of the digital signal is accumulated and held, and the output signal of the accumulation holding means 2 and the analog / digital conversion. By outputting a product operation with the output signal of the means 1, a feed-forward type automatic gain control system is configured. By this product operation, a constant output amplitude is obtained at the output of the product operation means 3. Further, a feedback type automatic gain control means 6 is added so that a commonly fluctuating component of the reception fluctuation is corrected by the feedback type automatic gain control means 6, and an individual burst fluctuation is corrected by the feedforward type automatic gain control means. did.

[0013]

FIG. 2 is a block diagram showing in detail the automatic gain control of feedforward included in the present invention.
In the figure, the modulation method of the received signal is, for example,
A phase modulation method such as K or QPSK is used. As an effective means for demodulating such a phase-modulated wave, there is synchronous detection using a reference carrier synchronized with a received signal as a reference signal. Reference numeral 21 denotes this synchronous detection. The received signal is demodulated by the synchronous detector 21, and unnecessary wave components are removed by the low-pass filter 22. The output signal of the low-pass filter 22 is converted by an analog / digital converter (A / D converter) 23 into, for example, an 8-bit digital signal. The most significant bit MSB of this digital signal becomes ordinary reproduced digital data. In a code error correction system using a soft decision signal, bits below the MSB are used. That is, the product operation of the amplitude information obtained at the output of the accumulator 24 and the output of the A / D converter 23 is performed so that the amplitude of the output bit does not change even if the amplitude of the demodulated baseband signal changes. The operation is performed by the arithmetic unit 25, whereby the digital code indicating the output amplitude, which is the output of the A / D converter 23, is converted in accordance with the amplitude information obtained in the output of the accumulator 24, and is converted to the output of the product arithmetic unit 25. An output for 3-bit constant amplitude soft decision is obtained.

As the product computing unit 25, an ordinary digital product computing unit, a reference method using a read only memory (ROM), or the like can be used. In this embodiment, the accumulator 24 for obtaining the amplitude information is composed of a full adder (full adder) 241 and a latch circuit 242, and performs accumulation by gating only the signal of the time to be accumulated.

FIG. 3 shows the feedforward A shown in FIG.
6 is a time chart illustrating an operation of a GC unit. The accumulation operation will be described with reference to FIG. In the TDMA communication system, since signals are transmitted in bursts from each station, it is necessary to track and demodulate the burst signals at high speed. In order to operate this demodulation device at high speed, a signal is added to the head of the burst as a training signal. This training signal includes the following signals as shown in FIG.

A demodulation reference carrier recovery circuit synchronization signal CR which is an unmodulated signal, a bit timing recovery circuit synchronization signal BTR which is an alternating signal of 1 and 0, and a UNIC word signal U which is a specific code indicating the start of data
W. The demodulator needs to complete the synchronization within the period of these training signals. In this embodiment, the signal level of this portion is measured and held by using the demodulation reference carrier recovery circuit synchronization signal CR, and the demodulation output level is made constant after the signal by the magnitude of the held signal.

For simplicity of explanation, the received modulated wave is B
A PSK modulated wave (a modulated wave in which the phase of a carrier is assigned to 0 ° and 180 ° according to digital 1 and 0) is assumed.
When this signal is demodulated by the synchronous detector 21 using the demodulation reference carrier, a signal shown in FIG. 3 (2) is obtained at the output of the low-pass filter 22. As shown in the figure, the CR section is an unmodulated wave,
The amplitude becomes constant. The BTR section is an alternating signal of 1 and 0, and the phase changes between 0 ° and 180 °, so that the waveform becomes a repetitive waveform as shown in the figure. These baseband signals are shown in FIG.
A / D conversion is performed according to the clock signal shown in (3). The latch circuit 242 is reset by a reset signal shown in FIG.
Next, by the full adder 241 and the latch circuit 242,
FIG. 3 shows a portion of the A / D conversion output corresponding to the CR signal.
A of the clock signal of (3) and the CR gate signal of FIG.
FIG. 3 is a diagram showing an example of accumulation according to an accumulation control signal obtained by ND.
The signal shown in (7) is obtained. The accumulation control signal is, as shown in FIG. 3 (5), a clock signal in a portion where the CR signal exists.

The accumulator 24 operates as a perfect integrator,
The signal obtained by integrating the CR section becomes baseband amplitude information proportional to the average baseband amplitude during the period of the CR signal.
The product of the baseband amplitude information and the instantaneous amplitude of the baseband signal output from the A / D converter 23 is calculated by the product calculator 25 so that the amplitude of the product output is constant. As a result, the level of the product output is kept constant even if the level differs for each burst.

The gate signal indicating the CR section in FIG. 3D is obtained from, for example, a TDMA controller (not shown).
That is, by using the unique word signal and performing frame synchronization, the time of the received signal can be predicted in advance, thereby obtaining a gate signal indicating the CR section. The demodulation reference carrier used in the synchronous detector 21 is
Obtained from the reference carrier recovery circuit.

As the clock signal in FIG. 3C, an output signal of the bit timing reproducing circuit can be used. FIG. 4 is a block diagram showing an automatic gain control device according to an embodiment of the present invention. 2, the same parts as those in FIG. 2 are denoted by the same reference numerals, and the difference from FIG. 2 is that a feedback AGC is added to the feed-forward AGC in FIG. The feedback type AGC includes a magnitude comparator 42 for comparing the output of the accumulator 24 with a reference code 41, an up / down counter 43 controlled in accordance with the output, a digital / analog conversion of the output of this counter, and an input side. Variable gain amplifier 45
And a D / A converter 44 for controlling the gain of the control signal.

The level of the received signal fluctuates on average (common burst fluctuation) due to rainfall near the receiving station, in addition to the level difference between bursts (individual burst fluctuation) caused by the difference between transmitting stations. In the embodiment of FIG. 4, the individual burst fluctuation is corrected by the feed-forward type AGC described in the embodiment of FIG. 2, and the common burst fluctuation is corrected by the feedback AGC.

The operation of the feedback type AGC is as follows.
The magnitude comparator 42 compares the accumulated output from the accumulator 24 with the level given as the reference code, and controls the up / down counter 43 with the output of the magnitude comparator 42. That is, if the accumulated output is larger than the reference code, the up / down counter 43 is operated as a down counter, and if the accumulated output is smaller than the reference code, the up / down counter 43 is operated as an up counter. In addition, the operation of the up / down counter is prohibited for other than the reference station burst by the reference station signal gate indicating that the signal is a burst signal from the reference station. As a result, the feedback type AGC operates so as to set the reference station burst signal to a constant level. The output of this counter is D / A converted, and the gain of the variable gain amplifier 45 on the input side is controlled by the obtained analog signal. Thereby, a feedback type AGC is realized.

The feedback type AGC shown in FIG. 4 is based on digital signal processing, but can also be realized by analog processing. Further, in the example of FIG.
Is performed on the transmission burst signal of the reference station, but it is also possible to use another station, for example, a return signal transmitted by the own station. Further, the operation may be performed so as to keep the peak burst level constant without identifying the station.

[0024]

As is apparent from the above description, according to the present invention, the signal having a level difference between bursts
Since the C operation can be performed stably, it is possible to introduce a code error corrector using soft decision information. Further, although the level can be made constant for each burst, waveform distortion due to a transient response at the time of AGC pull-in does not occur, so that correct data demodulation can be performed from the head data portion of the burst signal.

Further, by combining the feedback type AGC and the feed forward type AGC, the above effects can be obtained over a very wide range of input level fluctuation. Further, since the operation is realized by digital signal processing, stable operation can be performed without adjustment, and LSI can be easily realized.

[Brief description of the drawings]

FIG. 1 is a principle block diagram of the present invention.

FIG. 2 is a block diagram of a feedforward AGC of the present invention.

FIG. 3 is a time chart for explaining the operation of the apparatus shown in FIG. 2;

FIG. 4 is a block diagram showing an automatic gain control device according to another embodiment of the present invention.

FIG. 5 is a block diagram showing a conventional automatic gain control device.

FIG. 6 is a time chart for explaining the operation of the conventional example shown in FIG. 5;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Analog / digital conversion means 2 ... Accumulation holding means 3 ... Product calculation means 4 ... Variable gain amplifying means 5 ... Synchronous detection means 6 ... Feedback type automatic gain control means

Continuation of the front page (56) References JP-A-4-167629 (JP, A) JP-A-4-196929 (JP, A) JP-A-4-14335 (JP, A) JP-A-3-218136 (JP) JP-A-62-272227 (JP, A) JP-A-60-145713 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H04B ^7/ 14-7/22 H04B 1/06 H04B 1/16 H04J 3/00

Claims (1)

(57) [Claims] 1. A demodulator for a time division multiple access communication system, comprising: a variable gain amplifying means for amplifying a burst reception signal; and synchronously detecting an output of the variable gain amplifying means with a demodulation reference carrier. Synchronous detection means (5) for performing demodulation, analog / digital conversion means (1) for converting a demodulated baseband signal obtained at the output of the synchronous detection means (5) into a digital signal, Accumulative holding means (2) for accumulating and holding a prefix signal portion; and product calculating means (for outputting a product operation of an output signal of the accumulative holding means (2) and an output signal of the analog / digital conversion means (1)) 3) a feedback type automatic gain control means (6) for controlling the gain of the variable gain amplifying means (4) according to the difference between the amplitude information obtained at the output of the accumulation holding means (2) and the reference amplitude; With The analog / digital conversion means (1), the accumulation holding means (2), and the product calculation means (3) constitute a feed-forward type automatic gain control means. An automatic gain control method wherein the feedback type automatic gain control means (6) corrects the individual burst fluctuations by the feedforward type automatic gain control means.