JPH08331192A - Qam demodulator - Google Patents

Abstract

PURPOSE: To demodulate a modulating signal by detecting a synchronization symbol from a position of the symbol at which the result of product sum operation between the amplitude error and the phase error has a minimum value so as to correct the phase thereby conducting phase correction based on detection of accurate synchronization with a simple configuration and operation. CONSTITUTION: A regeneration circuit regenerate a symbol timing signal nearly synchronously with a symbol timing of a transmitter side. An AGC circuit 10 extracts M-sets of higher amplitude values for a prescribed period as to each amplitude of symbol signals when M-sets of synchronization symbols and pilot symbols decided to take a maximum amplitude for a prescribed period are in existence to adjust the gain. An error detection circuit obtains an error of a synchronization symbol with respect to an amplitude and an error of the synchronization symbol with respect to a phase respectively as to each of S-sets of optional consecutive symbols when S-sets of synchronization symbols are in existence for a prescribed period. A minimum value detection circuit detects a symbol in which result of product sum between the amplitude error and the phase error is a minimum value. Then the synchronization symbol is detected from a position of the symbol providing the minimum value as above to correct the phase.

Description

Detailed Description of the Invention

[0001]

This invention relates to quadrature amplitude modulation (QA
The present invention relates to a QAM demodulator that demodulates a modulated wave signal according to M), and in particular, a QAM that can detect a synchronization symbol or a pilot symbol with a simple configuration and arithmetic processing to perform phase correction.
The present invention relates to a demodulator.

[0002]

2. Description of the Related Art An MCA (Multi Channel Access) system is a system in which different users jointly use a plurality of channels and is widely used for business purposes. Further, in this MCA system, in addition to voice wireless communication, it is also used as various data communication.

The number of users in this MCA system has been increasing rapidly in recent years, and there is also a great demand for a system suitable for data transfer. Therefore, digital MC
The specifications of the A system were decided and put to practical use.

In such a system, multilevel QAM (Qu
adrature Amplitude Modulation) is used, for example, multi-valued quadrature amplitude modulation in which 4-valued amplitude modulation is performed by carrier signals having 90-degree different phases to have a 16-valued state.

Further, as a modulation method of a digital MCA system, RCR (Research & Development Center fo
The STD-32 digital MCA system defined as the standard of the Radio Systems: Development Center (Radio Systems Foundation) uses the M16QAM system and uses four subcarriers.

By the way, in wireless communication, the signal strength on the receiving side may change greatly due to changes in the propagation state, and automatic gain adjustment (automatic gain c) may occur in the processing of the received signal.
ontrol: AGC) is indispensable.

For example, the AGC circuit 1 for realizing AGC of a signal by frequency modulation or phase modulation is realized by a circuit as shown in FIG. That is, the integrated result of the integrator 3 is processed by the low-pass filter 4a so that the integrated value of the output of the variable gain amplifier 2 becomes constant, and then input to the gain control terminal of the variable gain amplifier 2 to adjust the gain. Was there. In this case, since the amplitude of the signal by frequency modulation or phase modulation is always constant, such processing can be realized.

By the way, since the QAM signal also has information on the amplitude value, the integrated value of the amplitude value is not always a stable value. Therefore, AGC cannot be realized by using the value obtained by integrating the amplitude value of the signal as shown in FIG.

Here, the multilevel QAM modulated signal can be expressed by the following equation. I (t) = A (t) cosφ (t) Q (t) = A (t) sinφ (t) That is, this multi-level QAM modulated signal is modulated so as to have information on the phase and the amplitude.

As described above, the AGC and the phase correction for the multi-level QAM modulated signal having information on the amplitude component are as shown in FIG.
It was general to carry out as follows using a circuit as shown in FIG.

First, the quadrature demodulator 4 performs quadrature demodulation to separate the in-phase component I (T) and the quadrature component Q (T). Then, the synchronization detector 5 detects a synchronization symbol from I (T) and Q (T), and the timing extractor 6 obtains the timing of the synchronization symbol from this synchronization symbol. Then, the phase / amplitude corrector 7 performs the amplitude correction (AGC) and the phase correction of I (T) and Q (T) by using the amplitude value and the phase value of the synchronization symbol which are known in advance.

In this case, the synchronization detector 5 has to perform the synchronization detection in a state where the amplitude value before the AGC is indefinite, which is a very difficult process. That is, the synchronization symbol is detected by utilizing the fact that it is defined by both the amplitude value and the phase angle, but the detection in the indefinite state depends on the hardware and processing of the processing circuit. There was a problem that complicated things were required in terms of algorithms.

That is, the synchronization detector 5 detects a synchronization symbol having a predetermined phase as follows. Here, assuming that the amplitude is A (T) and the phase is φ (T), tanφ (T) = (A (T) sinφ (t) / A (T) cosφ (t)) = (Q (T) / I (T)).

Therefore, φ (T) = Arctan (Q (T) / I (T)).

Thus, regardless of the amplitude A (T), I
Φ (T) can be obtained from (T) and Q (T), and the process of detecting this phase φ (T) is performed by the phase detection unit 5a in FIG. Further, the phase φ (T) thus detected is compared with the phase of the sync symbol read from the phase table 5c by the phase comparator 5b, and the position of the sync symbol is detected.

[0016]

The synchronization detection process as described above has a problem that division is required to remove the amplitude value A (T). The circuit that realizes this division is generally more complicated than the processing such as integration. Furthermore, the calculation of Arctan is required to obtain the phase φ (T), which complicates the processing. Also, a lot of processing time is required for processing such as table reference.

Further, in addition to the above-described phase synchronization detection,
Detection from the amplitude may also be performed incidentally. In this case, the sync detector 5 detects the sync symbol having the maximum amplitude value.
It is necessary to detect the amplitude value information for at least one frame and search the stored contents. Therefore, the synchronous detection from the amplitude also has a problem that the circuit for storing the amplitude value information becomes large in scale.

Therefore, even if it is realized by a DSP, the arithmetic processing becomes troublesome and there is a problem that high-speed arithmetic is required.

The present invention has been made in view of the above points, and an object thereof is to demodulate a modulated wave signal by performing phase correction based on accurate synchronization detection with a simple configuration and calculation. It is to realize a QAM demodulator.

[0020]

The inventor of the present application has conducted earnest research to improve the drawback of the complexity of the arithmetic and processing circuits in the synchronization detection in the conventional QAM demodulator, and as a result, during the predetermined period. The present invention has been completed by finding a new method for performing accurate synchronization detection by paying attention to the continuous number of synchronization symbols and the errors in their phase and amplitude.

Therefore, the present invention, which is a means for solving the above-mentioned problems, is configured as described below.

That is, according to the present invention, which is a means for solving the problem, a frame constituted by an arbitrary number of consecutive symbols at a predetermined phase angle on a circle passing through a symbol of maximum amplitude in signal space coordinates. In a QAM demodulation device for demodulating a QAM signal including a synchronization symbol of 1 and a pilot symbol inserted in the middle of a frame for facilitating phase correction, a symbol timing reproduction signal substantially synchronized with the symbol timing on the transmission side, A reproducing circuit for reproducing a symbol signal composed of orthogonal components I (T) and Q (T), and M synchronization symbols and pilot symbols determined to have the maximum amplitude within a predetermined period exist. Then, the upper M values of the amplitude value of the symbol signal within a predetermined period are extracted, and the extracted M values are extracted.
AGC circuit that performs gain adjustment using the average value or integrated value of the amplitude values of the individual symbol signals, and if there are S synchronization symbols within a predetermined period, for any consecutive S symbols, An error detection circuit that obtains an error (amplitude error) with respect to the amplitude of the sync symbol and an error (phase error) with respect to the phase of the sync symbol, and a sum of products for performing a sum-of-products calculation of the amplitude error and the phase error Computation means and a minimum value detection circuit for detecting a symbol for which the product-sum operation result from the product-sum operation means is the minimum value are provided, and a synchronization symbol is detected from the symbol position where the product value is the minimum value to perform phase correction. It is characterized by that.

[0023]

In the QAM demodulator which is a means for solving the problem, when receiving a QAM signal having a predetermined M number of synchronization symbols and pilot symbols which are determined to have the maximum amplitude within a predetermined period, Focusing on the upper M amplitude values of the amplitude value per frame, storing this,
The variable gain is adjusted so that the error between the average value or integrated value of the stored amplitude values and the value determined as the maximum amplitude on the demodulation side approaches zero. Then, when there are S synchronization symbols in a predetermined period, the error with respect to the amplitude of the synchronization symbol (amplitude error) and the error with respect to the phase of the synchronization symbol (phase error) are determined for any consecutive S symbols. Each of them is obtained, and the amplitude error and the phase error are subjected to sum-of-products operation to obtain the sum-of-products operation result, and the synchronization symbol is detected from the symbol position where the sum-of-products operation result becomes the minimum value.
Then, the phase correction is performed based on the synchronization symbol thus detected.

[0024]

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

FIG. 1 is a configuration diagram showing a principle configuration of a main part of a QAM demodulator of the present invention, FIG. 2 is a configuration diagram showing a part of FIG. 1 in detail, and FIG. 4 is a more detailed diagram of FIG. It is explanatory drawing shown in FIG. FIG. 5 is a block diagram showing the overall configuration of the QAM demodulation device of the present invention. 1 to 5, the same components are designated by the same reference numerals. In this embodiment, the case where the number of subcarriers is 4 will be described as an example.

First, the overall configuration of the QAM demodulator, its operation and the outline of the processing procedure will be described with reference to FIGS.
After that, the relationship between the AGC and the synchronization detection, which are the characteristic parts of the present invention, and the processing content will be described with reference to FIGS.

The QAM demodulator shown in FIG. 5 is roughly divided into a quadrature demodulator 8 for separating subcarriers and quadrature demodulating.
And a symbol extraction unit 9 for extracting symbol data,
It is divided into an AGC circuit 10 that performs AGC on the amplitude value, a phase correction unit 20 that performs phase correction, and a data demodulator 30 that performs data demodulation. The reproduction circuit is composed of a quadrature demodulation unit 8 and a symbol extraction unit 9. Hereinafter, description will be given separately for each part.

[Subcarrier Separation] In FIG. 5, the reception signal received by an antenna (not shown) and converted into an electric signal is frequency-converted into a modulated wave signal having an intermediate frequency fR ′, and the orthogonal demodulation unit 8 Is supplied to. And
This quadrature modulator 8 demodulates the same number of subcarriers and outputs in-phase components (I signals: I1 (t), I2 (t), I3).
(T), I4 (t)) and the orthogonal component (Q signal: Q1
(T), Q2 (t), Q3 (t), Q4 (t)) are obtained.

FIG. 6 is an explanatory view schematically showing the arrangement of the subcarriers, in which four subcarriers (Sub1 to Sub4) are arranged around the frequency f0 of the carrier. Here, the subcarriers are arranged at intervals of 4.5 kHz, and the center frequency of each tip is f0-6.75.
kHz, f0 -2.25 kHz, f0 +2.25 kHz
z, f0 +6.75 kHz.

In order to demodulate a modulated wave signal having such subcarriers, in a conventional general apparatus, the same number of subcarriers as four orthogonal demodulators are supplied. Each of the four quadrature demodulators has a frequency f of the intermediate frequency signal.
The frequency fL of the difference between R ′ and the frequency of the desired subcarrier
Quadrature signals (1) to fL (4)
A local oscillator signal from a local oscillator that produces signals (sin and cos components) is also provided. In this way, quadrature demodulation and subcarrier separation are performed.

However, in such a conventional general demodulation (subcarrier separation) system, the same number of local oscillators as the subcarriers are required to separate the subcarriers,
There is also a problem that the circuit configuration becomes complicated.

For this reason, the inventor of the present application has already proposed a method of separating subcarriers by a local oscillator smaller than the number of subcarriers as of April 21, 1995, and uses this technique. It is possible to

That is, in a QAM demodulator for quadrature demodulating a modulated wave signal using a plurality of subcarriers, first quadrature demodulation is performed on the modulated wave signal before separating the subcarriers to obtain an in-phase component and a quadrature component. For the in-phase component and the quadrature component for extracting the in-phase component, the second quadrature demodulation by the frequency of the difference between the center frequency of the modulated wave signal and the sub-carrier is performed to obtain the signal from the local oscillator whose number of sub-carriers is smaller than A quadrature demodulator using
An arithmetic processing circuit for arithmetically processing the result obtained by the second quadrature demodulation and an LPF for filtering the arithmetic processing result by the bandwidth of the subcarrier are provided, and the subcarrier separation and the target in-phase component are performed. The demodulation with the orthogonal component is performed.

By doing so, it becomes possible to separate the subcarriers with respect to the modulated wave signal using a plurality of subcarriers by the number of oscillators smaller than the number of subcarriers. From the above, the number of oscillators can be reduced, and the circuit configuration and scale can be reduced. It is also advantageous when processing with a DSP. Moreover, each frequency of the local oscillator used to separate the subcarriers is a simple integer ratio. This is because the frequency intervals of normal subcarriers are even. As a result, the original frequency used for frequency division and frequency multiplication can be shared. This point also works advantageously when processed by the DSP.

[Symbol Extraction] On the transmitting side, FIG.
As shown in, both the in-phase component I (T) and the quadrature component Q (T) of the orthogonal code are generated as signals having a plurality of stepwise levels. Here, four levels are shown. When this is demodulated by the quadrature demodulator 8, FIG.
As shown in (b), the analog signal has a gentle slope. Therefore, the symbol extraction unit 9 performs symbol timing reproduction to reproduce the symbol clock (FIG. 7 (c)). Then, using the regenerated symbol clock, symbols are extracted from the orthogonal code of the analog signal (FIG. 7 (d)). In this way, the symbols are extracted and the orthogonal codes (I1 (T), I2 (T),
I3 (T), I4 (T), Q1 (T), Q2 (T), Q
3 (T) and Q4 (T)) are obtained.

[AGC] As shown in FIG. 1, in the AGC circuit 10, a variable gain amplifier 1 for amplifying with a variable gain is provided.
1 is arranged, and the gain is adjusted by a control signal described later supplied to the gain control terminal to realize the AGC. Although only one of the signal processing systems for four subcarriers is shown in FIG. 1 for the sake of explanation, in the overall circuit configuration corresponding to FIG. Is to do.

First, the amplitude value extractor 12 receives the output signal of the variable gain amplifier 11 and extracts the amplitude value. In this case, the amplitude value is extracted by I ² (T) + Q ² (T) or (I
² (T) + Q ² (T)) ^1/2 .

Then, when receiving a QAM signal having M synchronization symbols and pilot symbols which are determined to have the maximum amplitude within a predetermined period, the error extraction unit 1
In 3, the amplitude information from the amplitude value extraction unit 12 is received and the upper M amplitude values of the amplitude value per frame are stored. Then, the average value or integrated value of the stored amplitude values is taken. A control signal for adjusting the gain of the variable gain amplifier 11 is generated so that the error between the average value or integrated value and the value determined as the maximum amplitude on the demodulation side approaches zero.

Further, the integrator 14 integrates the control signal generated by the error extractor 13 to multiply the time constant by which the control system of this AGC circuit operates stably.

Therefore, the variable gain amplifier 11 receives the control signal thus adjusted and adjusts the gain to generate the orthogonal code with the amplitude value corrected. That is, such processing is performed for each subcarrier to obtain an orthogonal code (I
1 (T), I2 (T), I3 (T), I4 (T), Q1
(T), Q2 (T), Q3 (T), Q4 (T)) are obtained.

According to such AGC processing, the processing can be performed without going through the synchronization detection procedure which has been conventionally required, and moreover, the optimum amplitude correction can be performed on the QAM signal. Further, since the accurate amplitude correction is performed here, the subsequent synchronization detection and phase correction processing becomes easy.

[Phase Correction] The phase correction unit 20 corrects the phase of the orthogonal code whose amplitude has been corrected by the AGC circuit 10.

Here, the carrier angular frequency on the transmitter side is ω
c is the difference from the demodulation angular frequency on the receiver side, θ (t)
In this case, if the frequencies on the transmitter side and the receiver side are both stable, the above-mentioned θ (t) is proportional to the time t.

Conventionally, the correction of θ (t) is performed when the value of the nominal vector is known in advance, that is, when the pilot symbols Ip and Qp are transmitted. In this case, in the conventional general phase correction unit, the phase difference between the data actually received in the pilot symbol period and the pilot symbol is obtained, and the correction (phase estimation) is performed based on the phase difference data. Was there.

However, in order to make such a correction, the trigonometric function must always be calculated. For this, it is necessary to always perform series expansion or table reference.
There was a problem that the processing became complicated. Therefore, even when it is realized by a DSP, there is a problem that the arithmetic processing becomes troublesome and high-speed arithmetic is required.

For this reason, the inventor of the present application has already obtained a fixed coefficient as of April 12, 1995,
We have proposed a method to perform phase correction according to this coefficient,
It is possible to use this technique.

That is, in the case of demodulating a multi-level QAM modulated signal in which pilot symbols Ip, Qp having predetermined predetermined phases and amplitudes are inserted, when the pilot symbols Ip, Qp of the multi-level QAM modulated signal are received. Sin β and cos β corresponding to the phase difference β between the pilot symbols Ip and Qp and the true values Io and Qo are obtained as phase difference coefficients having fixed values, and the phase difference coefficients sin β and c
It is possible to obtain true values Io and Qo by performing phase correction by integration and addition / subtraction using osβ and the in-phase component Ir and quadrature component Qr in each symbol of the received data. As a result, a simple calculation of addition and addition
It becomes possible to perform highly accurate phase correction.

In this case, it is necessary to accurately detect the synchronization by the synchronization detector 21. Therefore, the synchronization detection circuit having the configuration shown in FIG. 2 is used to perform the synchronization detection in the method described below.

According to this method, since the amplitude values of the I (T) and Q (T) signals entering the sync detector 21 have already been corrected by the above-mentioned AGC, it is assumed that the number of sync symbols is S. , The amplitude error and the phase error with respect to the synchronization symbol of any continuous S symbols are independently calculated by the phase angle error detector 21a and the amplitude error detector 21c, and the symbol position where the sum is the minimum is detected. The synchronization symbol can be detected at this timing by detecting with the device 21f.

The method will be described first with respect to the phase angle error.

The phase angle of the sync symbol arranged on the circumference passing through the maximum amplitude on the signal space coordinate is Arcc from the I (T) or Q (T) signal whose amplitude is corrected.
It can be easily obtained from the calculated value of os or Arcsin and the sign of I (T) and Q (T). In Fig. 3, I (T) and Q in each quadrant, which is one of the factors for calculating the phase angle,
The symbol (T) is shown.

The process for obtaining such a phase angle is shown in FIG.
After the phase angle detector 21g in the phase angle error detector 21a as shown in (a) has performed, the phase angle comparison unit 21i that refers to the phase angle table 21h performs it.

When the difference between the detected phase angle deviation and the predetermined synchronization symbol phase angle deviation is calculated, the error at the sync symbol position is minimized.

Thus, also in the detection of the phase angle error,
It does not require the complicated divisions mentioned in the general method.

Next, the amplitude error will be described.

The synchronization symbols are arranged on the circumference passing through the maximum amplitude on the signal space coordinates. Accordingly, the amplitude information value I of I (T) and Q (T) that has already been amplitude-corrected
² (T) + Q ² (T) or (I ² (T) + Q
² (T) ^1/2 is supposed to continuously reach a predetermined maximum value at the sync symbol position, and the error from this maximum value becomes the minimum at the sync symbol position.

In this method, it suffices to pay attention only to S consecutive symbols, and there is a new section (normally 1
It is not necessary to search for the maximum value of (frame).

That is, it suffices to perform processing on S consecutive symbols by the amplitude detector 21j and the comparator 21k in the amplitude error detector 21c shown in FIG. 4B.

Therefore, the memory capacity can be saved as compared with the conventional device, and the device can be downsized. Also,
The simplification of the search processing can be realized, the calculation speed becomes faster if the processing capacity is the same as the conventional one, and a low-speed processing device can be used when the processing time is the same as the conventional one.

To determine the position of the sync symbol, if the detection of only one of the amplitude error detection and the phase angle error detection is used, a symbol string similar in amplitude to the sync symbol, or Since the determination is erroneous when a symbol sequence similar to the synchronization symbol arrives on the phase angle, both are multiplied by a coefficient and the determination is performed from the minimum value of the addition results. The coefficient ratio at this time is not limited to 1: 1 and may be set according to the characteristics of each system.

As described above, the amplitude error and the phase angle error are detected for each of the S consecutive symbols, both of them are multiplied by the coefficient, and the synchronization symbol is detected from the minimum value of the addition results. , The arithmetic processing can be simplified, and more accurate synchronization detection suitable for each system can be performed.

Further, by performing the synchronization detection described here, it is possible to easily realize a system in which the applicant has already proposed a fixed coefficient and performs phase correction according to this coefficient. Like

[Data Demodulation] Then, the orthogonal code whose amplitude and phase are corrected as described above is demodulated by the data demodulator 30 according to a predetermined algorithm to obtain digital data of the bit stream output.

[Effect: Comparison with the conventional example] As described above, a QAM signal having a predetermined number M of synchronization symbols and pilot symbols determined to have the maximum amplitude within a predetermined period is received. In this case, paying attention to the top M amplitude values of the amplitude value in one frame, storing the values, the average value or integrated value of the stored amplitude values, and the value determined as the maximum amplitude on the demodulation side. The variable gain is adjusted so that the error between and becomes close to 0, and when there are S synchronization symbols within a predetermined period, for any consecutive S symbols, the error with respect to the amplitude of the synchronization symbol ( Amplitude error) and error with respect to the phase of the synchronization symbol (phase error)
Respectively, the sum of the amplitude error and the phase error is calculated to obtain the sum of products calculation result, and the synchronization symbol is detected from the symbol position where the sum of products calculation result is the minimum value.

As a result, first, the AGC process itself can be performed easily and accurately, and then the accurate synchronization detection process can be performed. Therefore, as compared with the conventional device, it becomes possible to use a simple configuration in terms of the hardware of the processing circuit and the processing algorithm while maintaining the accuracy.

[0066]

As described in detail above, according to the present invention,
QA in which a predetermined M number of synchronization symbols and pilot symbols are determined so as to have the maximum amplitude within a predetermined period
When receiving the M signal, pay attention to the upper M amplitude values of the amplitude value per frame, store these, and store the average value or integrated value of the stored amplitude values and the maximum amplitude on the demodulation side. When the variable gain is adjusted so that the error from the value determined to be close to 0 is S and there are S synchronization symbols within a predetermined period, the amplitude of the synchronization symbol for any consecutive S symbols Error (amplitude error) and the error relative to the phase of the synchronization symbol (phase error) are respectively obtained, and the product-sum operation result is obtained from the amplitude error and the phase error. From the symbol position where the product-sum operation result is the minimum value, Detect sync symbol.

As a result, the AGC process itself can be performed easily and appropriately, and then the phase correction based on the accurate synchronization detection process can be performed. Therefore, as compared with the conventional device, it becomes possible to use a simple configuration in terms of the hardware of the processing circuit and the processing algorithm while maintaining the accuracy.

Therefore, it is possible to realize a QAM demodulator capable of demodulating a modulated wave signal by performing phase correction based on accurate synchronization detection with a simple configuration and calculation.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a configuration of a main part of a QAM demodulation device of the present invention.

FIG. 2 is a configuration diagram showing a detailed configuration of a main part of the QAM demodulation device of the present invention.

FIG. 3 is an explanatory diagram showing symbols of I (T) and Q (T) in each quadrant in the phase angle calculation of the QAM demodulation device of the present invention.

FIG. 4 is a configuration diagram showing a detailed configuration of a main part of the QAM demodulation device of the present invention.

FIG. 5 is a configuration diagram showing an overall configuration of a QAM demodulation device of the present invention.

FIG. 6 is an explanatory diagram showing an arrangement of four subcarriers.

FIG. 7 is an explanatory diagram showing a state of symbol extraction.

FIG. 8 is a configuration diagram showing a configuration of a conventional AGC circuit.

FIG. 9 is a configuration diagram showing a configuration of a circuit that executes conventional amplitude correction and AGC.

FIG. 10 is a configuration diagram showing a conventional phase detection circuit.

[Explanation of symbols]

 10 AGC Circuit 20 Phase Corrector 21 Sync Detector 22 Timing Extractor 23 Phase Corrector

Claims (1)

[Claims] 1. A synchronization symbol of a frame composed of an arbitrary number of consecutive symbols each having a predetermined phase angle on a circumference passing through a symbol of maximum amplitude in signal space coordinates,
Q for demodulating a QAM signal including pilot symbols inserted in the middle of a frame for facilitating phase correction
In the AM demodulator, the orthogonal components I (T) and Q are generated by the symbol timing reproduction signal which is substantially synchronized with the symbol timing on the transmission side.
(T) a reproducing circuit for reproducing a symbol signal composed of (T), and M synchronization symbols and pilot symbols that are determined to have the maximum amplitude within a predetermined period are present.
An AGC circuit for extracting the upper M values of the amplitude value of the symbol signal within a predetermined period, and performing gain adjustment using the average value or integrated value of the amplitude values of the extracted M symbol signals; , If there are S synchronization symbols in a predetermined period,
An error detection circuit for obtaining an error (amplitude error) with respect to the amplitude of the synchronization symbol and an error (phase error) with respect to the phase of the synchronization symbol for any consecutive S symbols, and a product-sum operation of the amplitude error and the phase error. The sum-of-products calculation means for obtaining the sum-of-products calculation result and the minimum-value detection circuit for detecting the symbol for which the sum-of-products calculation result from the sum-of-products calculation means is minimum are provided. A QAM demodulator which detects a synchronization symbol and performs phase correction.