JPH11196141A - Soft judgement method and receiver for psk - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a soft judgement method and a receiver of PSK capable of reflecting both of the degrees of uncertainy in the phase direction and of uncertainty in data due to small amplitudes on a soft judgement value and demonstrating the characteristics of an error correction code to a maximum. SOLUTION: When an input of a delay detection circuit 21 is represented by IQ, IQ → phase amplitude conversion is performed in a conversion circuit 211, and a phase difference θ between a symbol at the present point in time and a delayed symbol is obtained by a delay circuit 212 and a subtractor 213, so that a phase soft judgement value is obtained. Also, from amplitude information amp0 of the symbol at the present point in time and amplitude information amp1 of a symbol delayed in a delay circuit 217, an amp-factor as a weighting factor is generated in a weighting factor generation circuit 218. Then, weighting is performed to the phase soft judgement value by the amp-factor and the soft judgement values I, 'Q' are obtained. By performing Viterbi decoding through the use of the soft judgement values, correction ability is improved more than the case of soft judgement only from phase information or amplitude information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention finds a phase soft decision value of each bit from a phase component when receiving a PSK (Phase Shift Keying) signal transmitting information by a phase component. The present invention relates to a PSK soft decision method and a soft decision circuit.

[0002]

2. Description of the Related Art In recent years, digital broadcasting has been actively researched. Research on terrestrial digital broadcasting is also being actively conducted, and its transmission method is OFDM (Orthogonal Frequency).
Division multiplex (orthogonal frequency division multiplexing) modulation is considered promising. As the error correction code, a concatenated code in which the outer code is a Reed-Solomon code and the inner code is a convolutional code is effective. For information transmission, a PSK method for transmitting information by modulating a phase component has been considered.

[0003] In a conventional transmission method of a phase modulation signal, a demapping process of a received signal is performed from phase information. For example, DQPSK (Differential Qua
drature Phase Sift Keying: Differential 4-phase phase shift keying
Then, when the two symbols S1 and S2 in FIG. 18A are received, a hard decision is made from the phase difference θ at the boundary of θ = nπ / 2 + π / 4. In this example, as shown in FIG. 18B, the hard decision result [y1, y0] = [10]. When convolutional coding-Viterbi decoding is used as an error correction code,
It is known that the correction ability is improved by softly determining [y1, y0].

On the other hand, as shown in FIG. 19, a QPS that makes a soft decision based on the phase difference between two differentially encoded symbols
The K soft decision method is described in Japanese Patent Application Laid-Open No. Sho 62-60339 (Japanese Patent Application No. 60-200017, hereinafter referred to as the first cited document). The receiving apparatus according to the soft decision method described in the first cited reference includes a delay detection circuit 11
When the input of is represented by IQ, the conversion circuit 111 outputs IQ →
The phase conversion is performed, the symbol delay circuit (Z ⁻¹ ) 112 and the subtractor 113 determine the phase difference θ between the current symbol and the delayed symbol, and the phase soft decision circuit 114 obtains a soft decision value from the phase information. The Viterbi decoding is performed by the QPSK Viterbi decoder 12 using the soft decision value. If this soft decision method is used, the correction ability is improved as compared with the hard decision.

[0005] Fig. 20 shows an example of soft decision in the receiving apparatus. Now, soft decision value of strong “0” = 0, weekly
Soft decision value of “0” = 3, soft decision value of strong “1” = 7,
Consider an 8-value soft decision where the soft decision value of weekly “1” = 4. According to FIG. 20 (a), from the phase difference θ, y0 = "0"
And a soft decision is made for y1. In this case, y1
20 has a soft decision value of 5, and y2 has a soft decision value of 0.
It is mapped as shown in (b) (symbol S3),
Inputs I ″ and Q ″ of the QPSK Viterbi decoder 12.

However, in the above soft decision method, the uncertainty of the data due to the small amplitude is
There is a problem that it is not reflected in the soft decision value.

In Japanese Patent Application Laid-Open No. 4-352541 (Japanese Patent Application No. 3-127660, hereinafter referred to as a second cited document), a soft decision is made by weighting hard-decided decoded data with amplitude information. A method is described. However, this method has a problem that the degree of uncertainty in the phase direction is not reflected in the soft decision value.

In terrestrial digital broadcasting, as shown in FIG. 21 (a), the reception level fluctuates due to multipath (ghost) or fading (in FIG. 21A, the reception level is large at arrow A, and the reception level is large at arrow B). The reception level is low.) For example, in the OFDM modulation, when the mapping is DQPSK, as shown in FIG. 21B, in the delay detection between carriers having a large reception level in the range indicated by the arrow A, the reliability of the phase information is high as shown in FIG. In delay detection between carriers having a low reception level in the range indicated by B, as shown in FIG. 21C, the reliability of the phase information is low. In particular, when there is noise, the latter is easier to make an error.

[0009] The PSK modulated signal receiving apparatus as described above may be configured using a delay detection circuit 13 for performing delay detection by conjugate complex multiplication, as shown in FIG. The delay detection circuit 13 includes a delay circuit 131 for delaying an input IQ signal, and an input IQ signal and a delay circuit 1.
And a complex multiplication circuit 132 for performing conjugate complex multiplication of the output of the output 31. By shifting the output I′Q ′ of the complex multiplication circuit 132 by π / 4 phase by the phase shift circuit 14, the input soft decision value I ″ Q ″ of the Viterbi decoder 15 is obtained.

[0010]

As described above, in the conventional soft decision method of PSK, in the method of the first cited document, the uncertainty of data due to small amplitude is not reflected in the soft decision value. Further, in the method of the second cited document, the degree of uncertainty in the phase direction is not reflected in the soft decision value.

In order to solve the above problem, the present invention can reflect both the degree of uncertainty in the phase direction and the degree of uncertainty of data due to small amplitude in the soft decision value. An object of the present invention is to provide a PSK soft-decision method and a receiving apparatus capable of maximizing the characteristics of a correction code.

[0012]

Means for Solving the Problems In order to achieve the above object, a soft decision method of PSK according to the present invention uses the following means.

(1) PS for transmitting information by phase component
When a K-system signal is received, a soft decision value of each bit is weighted by weighting a phase soft decision value of each bit obtained from a phase component with a weighting coefficient obtained from amplitude information.

(2) In the method of (1), when the received signal is modulated by any one of a multi-level PSK, a multi-level DPSK, a multi-level APSK, and a multi-level DAPSK,
The weighting coefficient is an amplitude value obtained by clipping the normalized amplitude or more.

(3) In the method of (1), when the received signal is modulated by a multi-level DPSK or a multi-level DAPSK system, the weighting coefficient is an arithmetic mean of amplitudes of two symbols, and a multiplication of two symbols. The average is the product of the amplitudes of the two symbols, the smaller of the two symbols, or the current amplitude of the two symbols.

(4) In the method of (3), the amplitude used as the weighting coefficient is an amplitude value obtained by clipping the normalized amplitude or more.

The receiving apparatus using the PSK soft decision circuit according to the present invention has the following configuration.

(5) PS for transmitting information by phase component
Phase-amplitude conversion means for extracting phase information and amplitude information from a signal according to the K method, phase soft-decision means for obtaining a phase soft-decision value for each bit from the phase information obtained by this means, Weighting means for weighting and outputting the soft decision value with a predetermined weighting coefficient.

(6) PS for transmitting information by phase component
Phase / amplitude conversion means for extracting phase information and amplitude information from a signal according to the K method, phase soft decision means for obtaining a phase soft decision value of each bit from the phase information obtained by this means, and phase / amplitude conversion means Weighting coefficient generation means for generating a weighting coefficient based on the obtained amplitude information; and a soft decision value obtained by weighting the phase soft decision value obtained by the phase soft decision means with the weighting coefficient obtained by the weighting coefficient generation means. And weighting means for outputting

(7) In the configuration of (6), when the signal to be received is modulated by a multi-level DPSK or a multi-level DAPSK system, the weighting coefficient generating means operates on the arithmetic mean of the amplitudes of two symbols. The geometric mean of the symbols, the product of the amplitudes of the two symbols, the amplitude of the smaller of the two symbols,
Let it be one of the current amplitudes of the two symbols.

(8) In the configuration of (6), the weighting coefficient generation means outputs the amplitude used as the weighting coefficient as
A value equal to or greater than the normalized amplitude is defined as the clipped amplitude value.

The delay detection circuit according to the present invention has the following configuration.

(9) Clipping means for clipping the input signal, delay means for delaying the output of the clipping means, and complex multiplication means for performing conjugate complex multiplication of the output of the clipping means and the output of the delay means. Is provided.

(10) Amplitude detecting means for obtaining an amplitude from an input signal, clipping means for performing clipping on the input signal using the output of the amplitude detecting means, delay means for delaying the output of the clipping means, and clipping Complex multiplying means for performing conjugate complex multiplication of the output of the means and the output of the delay means.

[0025]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The embodiments of the present invention will be described below in detail with reference to FIGS.

[First Embodiment] FIG. 1 shows the configuration of a DQPSK receiving apparatus employing a soft decision method of a modulated signal according to the DQPSK method as a first embodiment according to the present invention. In the delay detection circuit 21 used in this receiving apparatus, the DQPSK modulation inputs I and Q are supplied to an IQ → phase / amplitude conversion circuit 211 and converted into phase information and amplitude information for each symbol. The phase information is supplied to a phase soft decision circuit 214 as phase difference information θ between the current symbol and the delayed symbol by a symbol delay circuit (Z ⁻¹ ) 212 and a subtractor 213, where the phase difference information is softened. The judgment value is obtained. This soft phase decision value is calculated by the multiplier 21
The data is output to the Viterbi decoder 22 for DQPSK via the 5, 5 and 216, and is converted into serial data by Viterbi decoding.

On the other hand, in the delay detection circuit 21, the amplitude information obtained by the conversion circuit 211 is delayed by one symbol period by a symbol delay circuit (Z ^-1 ) 217, and the delay output amp1 is the amplitude information of the current symbol. The signal is supplied to the weighting coefficient generation circuit 218 together with amp0. The weighting coefficient generation circuit 218 outputs the current symbol amplitude information amp0.
Weighting coefficient am from amplitude information amp1 of the delayed symbol
Generates a p-factor, and this weighting factor amp-fact
or is supplied to multipliers 215 and 216 and becomes a multiplication coefficient of the phase soft decision value.

In the above configuration, the first to sixth embodiments
The soft decision method of DQPSK modulation will be described with reference to the embodiment.

In FIG. 1, when the input of the delay detection circuit 21 is represented by IQ, IQ → phase and amplitude conversion is performed by the conversion circuit 21, and the current symbol and the delayed symbol are converted by the delay circuit 212 and the subtractor 213. The phase difference θ is obtained, and the phase soft decision value is obtained as in the case shown in FIG. Also,
A weighting coefficient generation circuit 218 generates a weighting coefficient amp-factor from the current symbol amplitude information amp0 and the symbol amplitude information amp1 delayed by the delay circuit 217. Then, the phase soft decision value is weighted by a weighting coefficient amp-factor, and a soft decision value I "Q" is obtained as shown in FIG.
When Viterbi decoding is performed using this soft decision value, both the degree of uncertainty in the phase direction and the degree of uncertainty of the data due to the small amplitude can be reflected in the soft decision value, and the characteristics of the error correction code Can be maximized, so that the correction ability can be improved as compared with the soft decision based on only the phase information and the soft decision based on only the amplitude information.

(First Embodiment) Weighting coefficient amp-factor
Is the arithmetic mean of the amplitudes of the two differentially encoded symbols, that is, amp-factor = (amp1 + amp0) / 2. For example,
When amp1 = 1.0 and amp0 = 0.5, amp-factor = 0.75 is obtained. When amp1 = 0.5 and amp0 = 1.5, amp-factor =
1.0 is obtained.

(Second Embodiment) Weighting coefficient amp-factor
Is the geometric mean of the amplitudes of the two differentially encoded symbols, that is, amp-factor = √ (amp1 × amp0). For example, am
When p1 = 1.0 and amp0 = 0.5, amp-factor = 0.707 is obtained. When amp1 = 0.5 and amp0 = 1.5, amp-factor =
0.866.

(Third Embodiment) Weighting coefficient amp-factor
Is the product of the amplitudes of the two differentially encoded symbols, ie, am
Let p-factor = amp1 × amp0. For example, amp1 = 1.0, am
When p0 = 0.5, amp-factor = 0.5. amp1 = 0.
5, when amp0 = 1.5, amp-factor = 0.75.

(Fourth Embodiment) Weighting factor amp-factor
Is the smaller amplitude of two differentially encoded symbols, that is, amp-factor = min (amp1, amp0). For example,
When amp1 = 1.0 and amp0 = 0.5, amp-factor = 0.5 is obtained. When amp1 = 0.5 and amp0 = 1.5, amp-factor =
0.5 is obtained.

(Fifth Embodiment) Weighting coefficient amp-factor
Is the current amplitude of the amplitudes of the two differentially encoded symbols, that is, amp-factor = amp0. For example, amp1
= 1.0, amp0 = 0.5, amp-factor = 0.5. When amp1 = 0.5 and amp0 = 1.5, amp-factor = 1.
5 is obtained.

(Sixth embodiment) Amp1 and amp0 are clipped with the normalized amplitude value 1. As an example, when the weighting factor amp-factor is formed by the arithmetic mean of the amplitudes of two differentially encoded symbols, and if amp-factor = (amp1 + amp0) / 2, when amp1 = 0.5 and amp0 = 1.5, When clipping is not performed, amp-factor = 1.0 is obtained. However, when clipping is performed to amp0 = 1.0, amp-factor = 0.75. By performing the clipping process in this manner, the low reliability of the phase information due to the small amplitude of amp1 = 0.5 can be obtained by using the amp-fact.
The correction capability of Viterbi decoding can be improved as compared with the case where clipping is not performed.

The level of the normalized value 1 of the amplitude information amp1 and amp0 may be on the unit circle in FIG. 18, inside the unit circle, or outside the unit circle. In addition, the interval may be linear or non-linear.

[Second Embodiment] As a second embodiment of the present invention, a method of soft-decision of a modulation signal by the 8PSK method will be described with reference to FIG.

When there is a received symbol S4 after synchronous detection at the position of the white circle in FIG. 3, a phase soft decision value is obtained from the phase component θ. θ is [y2, y1, y0] = “011” and [y2
, Y1, y0] = “001”, so that y2 =
Hard decision is made as 0, y0 = 1, and soft decision is made about y1. Thus, the soft phase decision value of y2 = 0, the soft phase decision value of y1 = 2, and the soft phase decision value of y0 = 7 are obtained. Next, the amplitude component
The phase soft decision value of each bit is weighted by amp0. As shown in FIG. 4A, the soft decision value of y2 = 1, and FIG.
A soft decision value of y1 = 3 and a soft decision value of y2 = 6 are obtained in the manner shown in FIG.

When amp0 is greater than 1, weighting may be performed in a direction in which the phase information is likely, that is, in a direction in which the soft decision value is close to 0 or 7. When amp0 is larger than 1, amp0 is clipped with the normalized amplitude value 1 and used.
That is, the soft phase decision value may be used as the soft decision value.

When Viterbi decoding is performed using this soft decision value, the degree of uncertainty in the phase direction and the high reliability of the phase information due to the large amplitude are reflected in the amp-factor, and are compared with those without clipping. As a result, the correction capability of Viterbi decoding can be improved, and the correction capability can be improved as compared with the case of soft decision using only phase information and the case of soft decision using only amplitude information.

Note that the level of the normalized value 1 of the amplitude information amp0 may be on the unit circle in FIG. 3, inside the unit circle, or outside the unit circle. Also, the interval may be linear or non-linear.

[Third Embodiment] As a third embodiment of the present invention, a method of soft-decision of a modulated signal according to the 16-DAPSK method will be described with reference to FIGS. Here, as shown in FIG. 5, a soft decision method in delayed detection of a white circle delayed symbol S5 having an amplitude amp1 and a white circle current symbol S6 having an amplitude amp0 is considered. In this case, the phase soft decision value is obtained from the phase difference θ between the two symbols in the manner shown in FIG.

Here, the symbol of 16 DAPSK is [y
[3, y2, y1, y0]. The information of y3 is transmitted on the amplitude ratio, and the information of y2, y1, y0 is transmitted on the phase difference. There are two amplitudes depending on y3, and y
There are eight different phases depending on 2, y1, and y0.

Since θ is between [y2, y1, y0] = “011” and [y2, y1, y0] = “010”, y2
= 0 and y1 = 1, and a soft decision is made for y0.
Thus, the soft phase decision value of y2 = 0, the soft phase decision value of y1 = 7, and the soft phase decision value of y0 = 3 are obtained.

Next, the soft phase decision value of each bit is weighted using the weighting coefficient amp-factor obtained from the amplitude components amp0 and amp1. The soft decision value of y2 = 2 in the manner shown in FIG. 7A and the soft decision value of y1 in the manner shown in FIG.
5. A soft decision value of y2 = 3 is obtained in the manner shown in FIG. When Viterbi decoding is performed using this soft decision value, as in the above-described embodiment, the correction capability can be improved compared to the case of soft decision using only phase information and the case of soft decision using only amplitude information.

The level of the normalized value 1 of the amplitude information amp1 and amp0 may be on any one of the inner and outer circles in FIG.
It may be inside, outside or in the middle of the circle. Also, the interval may be linear or non-linear.

Various methods are conceivable as the position of the normalized amplitude value and the method of engraving between the positions. For example, as shown in FIG. 8A, the normalized value is set on a circle on which a small symbol is placed,
The interval may be cut at equal intervals. Also, as shown in FIG. 8B, the normalized value is set on a circle on which a large symbol is placed,
The interval may be cut at equal intervals. Further, as shown in FIG. 8C, the normalized value is set as the center of gravity of the constellation,
The interval may be cut at equal intervals. Also, as shown in FIG. 8D, the normalized value is set on a circle on which a large symbol is placed,
The space between the two circles is coarse, and the inside of the small circle may be finely cut. Alternatively, as shown in FIG. 8E, the normalized value may be set on a circle on which a large symbol is placed, and the interval may be non-linearly cut.

[Fourth Embodiment] As a fourth embodiment of the present invention, a receiving apparatus using a soft decision circuit for a modulated signal based on the DQPSK method will be described with reference to FIG. Again, consider the case where a symbol represented by an IQ signal is received.

The soft decision circuit 31 is a phase / amplitude conversion circuit 31 for obtaining phase information θ 'and amplitude information amp0 from the input IQ signal.
1, a phase soft decision circuit 312 for obtaining a phase soft decision value from the phase information θ ′, and a weighting coefficient amp-fact from the amplitude information amp0.
and a weighting circuit 314 that weights the phase soft decision value with a weighting coefficient amp-factor to obtain soft decision values I ″ and Q ″.

The phase soft decision circuit 312 is a phase delay circuit (Z ^-1 ) 3121 for obtaining one symbol delay of the phase information θ '.
And a subtractor 3122 for obtaining a phase difference θ between two symbols;
Phase difference soft decision circuit 31 for obtaining a phase soft decision value from phase difference θ
23. Also, the weighting coefficient generation circuit 313
Is composed of an amplitude delay circuit (Z ^-1 ) 3131 that obtains amplitude information amp1 delayed by one symbol of the amplitude information amp0, and a weighting coefficient generation operation circuit 3132 that obtains a weighting coefficient amp-factor from the amplitudes amp0 and amp1 of two symbols. . The method of generating the weighting coefficient may be any of the methods of the first to sixth examples described in the first embodiment.

In the receiving apparatus having the above configuration, the soft-decision circuit 31 weights the soft-decision phase difference value obtained by the soft-decision phase difference circuit 3123 with the weighting coefficient obtained by the weighting coefficient generation operation circuit 3132. Therefore, the soft decision values I ″ and Q ″ output from the soft decision circuit 31 reflect the degree of uncertainty in the phase direction and the degree of uncertainty of the data due to the magnitude of the amplitude in the soft decision value, and Characteristics can be maximized. For this reason, by inputting the weighted phase soft decision values I ″ and Q ″ to the Viterbi decoder 32, the weighted phase soft decision values I ″ and Q ″ can be reduced more than when using soft decision values obtained only from phase information or when using soft decision values obtained only from amplitude information. , The correction capability of Viterbi decoding can be improved.

It is needless to say that the phase delay circuit 3121 and the amplitude delay circuit 3131 can be constituted by one RAM when receiving a multi-carrier transmission system such as OFDM modulation.

[Fifth Embodiment] As a fifth embodiment of the present invention, a receiving apparatus using a soft decision circuit for a modulated signal based on the DQPSK method will be described with reference to FIG. Here, a case is considered in which the differentially detected symbols represented by the phase difference information θ of two differentially encoded symbols and the amplitudes amp0 and amp1 of the two symbols are input.

The soft decision circuit 41 comprises a phase soft decision circuit 411 for obtaining a phase soft decision value from the phase difference information θ and an amplitude information amp0.
And a weight generation circuit 412 that obtains a weighting factor amp-factor from amp1 and a weighting factor amp-fact
and a weighting circuit 413 for obtaining soft decision values I ″ and Q ″ by weighting with or.

The method of generating the weighting coefficients may be any of the methods of the first to sixth embodiments described in the description of the first embodiment.

The soft decision values I ″ and Q ″, which are outputs of the soft decision circuit 41 having the above configuration, reflect the degree of uncertainty in the phase direction and the degree of uncertainty in the data due to the magnitude of the amplitude. Therefore, by inputting the soft decision values I ″ and Q ″ to the Viterbi decoder 42, the Viterbi decoder 42 uses a soft decision value obtained only from the phase information or a soft decision value obtained only from the amplitude information. The ability to correct the decoding can be improved.

[Sixth Embodiment] As a sixth embodiment according to the present invention, a receiving apparatus using a soft decision circuit for a modulated signal based on the DQPSK method will be described with reference to FIG. Here, a case is considered in which a symbol after differential detection represented by the phase difference information θ of two differentially encoded symbols and the current amplitude amp0 is input. In this embodiment,
As in the fifth example described in the description of the first embodiment, soft decision in the case where the weighting coefficient amp-factor = amp0 is the current amplitude of the amplitudes of two differentially encoded symbols. It is an example of a circuit.

The soft decision circuit 51 comprises a phase soft decision circuit 511 for obtaining a phase soft decision value from the phase difference information θ, and soft decision values I ″ and Q And a weighting circuit 512 for obtaining "".

The soft decision values I ″ and Q ″, which are outputs of the soft decision circuit 51 having the above configuration, reflect the degree of uncertainty in the phase direction and the degree of uncertainty in the data due to the magnitude of the amplitude. For this reason, by inputting the soft decision values I ″ and Q ″ to the Viterbi decoder 52, the Viterbi decoder 52 uses a soft decision value obtained only from phase information or a soft decision value obtained only from amplitude information. The ability to correct the decoding can be improved.

[Seventh Embodiment] As a seventh embodiment according to the present invention, a receiving apparatus using a soft decision circuit for a modulated signal based on the 8PSK method will be described with reference to FIG. Here, it is assumed that a symbol represented by an IQ signal is received.

The soft decision circuit 61 is a phase / amplitude conversion circuit 611 for obtaining phase information θ and amplitude information amp0 from the input IQ signal.
And a phase soft decision circuit 612 for obtaining a phase soft decision value from phase information θ, and a weighting circuit 613 for weighting the phase soft decision value with a weighting factor amp-factor, ie, amp0, to obtain soft decision values I ″ and Q ″. Become.

The output of the soft decision circuit 61 having the above configuration is speed-converted by the speed conversion circuit 62, and the soft decision values I ″ and Q ″ are obtained.
Is input to the Viterbi decoder 63. Here, the output of the speed conversion circuit 62 is three in 8PSK1.
The symbol contains 3-bit information, but the Viterbi decoder inputs a 2-bit soft decision value.

According to this configuration, the correction capability of Viterbi decoding is improved as compared with the case of using a soft decision value obtained only from phase information or the case of using a soft decision value obtained only from amplitude information.

When the phase information θ and the amplitude information amp0 are directly input to the soft decision circuit 61, the first phase / amplitude conversion circuit 611 in FIG. 12 becomes unnecessary.

[Eighth Embodiment] As an eighth embodiment of the present invention, a receiving apparatus using a soft decision circuit for a modulated signal based on the DQPSK method will be described with reference to FIGS.

In FIG. 13, a delay detection circuit 71 includes an amplitude detection circuit 711 for obtaining an amplitude from an input IQ signal, a clipping circuit 712 for performing clipping in the amplitude direction on the input IQ signal using an output of the amplitude detection circuit 711, It comprises a symbol delay circuit 713 that delays the output of the clipping circuit 712 by one symbol, and a complex multiplication circuit 714 that performs conjugate complex multiplication of each output of the clipping circuit 712 and the symbol delay circuit 713. By shifting the output I′Q ′ signal of the complex multiplication circuit 74 by π / 4 phase by the phase shift circuit 72, the input soft decision value of the Viterbi decoder 73 is obtained.

Conjugate complex multiplication is equivalent to obtaining a phase difference and weighting it with a product of amplitudes.

The clipping in the amplitude direction is shown in FIG.
As shown in (1), when a symbol S7 having an amplitude value equal to or larger than the normalized amplitude is received, clipping is performed on the unit circle in the amplitude direction. That is, each IQ is divided by the amplitude value. Delay detection is performed using the IQ signal after clipping in the same manner as in the related art. 14 (b) is obtained by conjugate complex multiplication of the IQ signal after clipping.

When the I'Q 'signal thus obtained is phase-shifted by π / 4 by the phase shift circuit 72, an I "Q" signal is obtained, and a soft decision value is obtained from the I "Q" signal.

For example, when amp0 = 0.5 and amp1 = 1.5, if clipping is not performed, it is determined that amp-factor = amp1 × amp0 = 0.75. On the other hand, if clipping is performed, amp-factor = 0.
5 (<0.75). The unreliability of the phase information due to amp0 = 0.5 is reflected in the amp-factor after clipping.

By using the delay detection circuit 71, the correction capability of Viterbi decoding is improved as compared with the case where the delay detection circuit of FIG. 22 is used.

[Ninth Embodiment] As a ninth embodiment of the present invention, a receiving apparatus using a soft decision circuit for a modulated signal based on the DQPSK method will be described with reference to FIG.

In FIG. 15, a delay detection circuit 81 is a clipping circuit 81 for clipping an input IQ signal.
1, a symbol delay circuit 812 for delaying the output of the clipping circuit 811 by one symbol, and a clipping circuit 8
11 and a complex multiplication circuit 813 that performs conjugate complex multiplication of the output of the symbol delay circuit 812. Complex multiplication circuit 8
13 is output from the phase shift circuit 82 to the
By performing a / 4 phase shift, an input soft decision value of the Viterbi decoder 83 is obtained.

Conjugate complex multiplication is equivalent to obtaining a phase difference and weighting it by a product of amplitudes.

When clipping other than clipping in the amplitude direction is performed using a differential detection circuit as shown in FIG. 15 (for example, clipping is performed independently for I and Q).
Also, for the same reason as in the eighth embodiment, the correction capability of Viterbi decoding is improved as compared with the case where the delay detection circuit of FIG. 22 is used.

[Tenth Embodiment] The first embodiment according to the present invention
0, a receiver using a soft decision circuit for a modulated signal based on the π / 4 shift DQPSK method,
This will be described with reference to FIGS.

In FIG. 16, a delay detection circuit 91 includes an amplitude detection circuit 911 for obtaining an amplitude from an input IQ signal, a clipping circuit 912 for performing clipping in the amplitude direction on the input IQ signal using an output of the amplitude detection circuit 911, It comprises a symbol delay circuit 913 that delays the output of the clipping circuit 912 by one symbol, and a complex multiplication circuit 914 that performs conjugate complex multiplication of the outputs of the clipping circuit 912 and the symbol delay circuit 913. The input soft decision value of the Viterbi decoder 92 is obtained from the output of the complex multiplication circuit 914.

Here, in the case of π / 4 shift DQPSK,
The phase information after differential detection is π / 4, 3π / 4, 5π /
4, 7π / 4, which is known to correspond to QPSK mapping. Conjugate complex multiplication is equivalent to obtaining a phase difference and weighting with a product of amplitudes.

The clipping in the amplitude direction is shown in FIG.
Perform as shown. That is, when the symbol S8 having an amplitude value equal to or larger than the normalized amplitude is received, each of the IQs is divided by the amplitude value, thereby performing clipping on the unit circle in the amplitude direction. Delay detection similar to that of the related art is performed using the IQ signal after clipping.

The I "Q" signal shown in FIG. 17B is obtained by conjugate complex multiplication of the IQ signal after clipping. I "
A soft-decision value is obtained from the Q ″ signal. At this time, since the phase is shifted by π / 4 for each symbol, the phase after complex multiplication is θ = n, as in QPSK, if error-free.
π / 2 + π / 4. Therefore, a phase shift circuit becomes unnecessary.

For example, when amp0 = 0.5 and amp1 = 1.5, if clipping is not performed, it is determined that amp-factor = amp1 × amp0 = 0.75. On the other hand, if clipping is performed, amp-factor = 0.
5 (<0.75). The unreliability of the phase information due to amp0 = 0.5 is reflected in the amp-factor after clipping.

By using the delay detection circuit 91, the correction capability of Viterbi decoding is improved as compared with the case where the delay detection circuit of FIG. 22 is used.

Even when clipping other than clipping in the amplitude direction is performed, the correction capability of Viterbi decoding is improved as compared with the case where the delay detection circuit of FIG. 22 is used.

[0084]

As described above, according to the present invention, both the degree of uncertainty in the phase direction and the degree of uncertainty in data due to a small amplitude can be reflected in the soft decision value. It is possible to provide a soft-decision method and a receiving apparatus for PSK that can maximize the characteristics of the correction code.

[Brief description of the drawings]

FIG. 1 shows a DQ according to a first embodiment of the present invention.
DQ adopting soft decision method of modulated signal by PSK method
FIG. 2 is a block circuit diagram showing a configuration of a PSK receiving device.

FIG. 2 is an I "Q" plan view for explaining a point of weighting in the DQPSK soft decision of the embodiment.

FIG. 3 shows a second embodiment according to the present invention,
FIG. 9 is an IQ plan view for explaining a soft decision method of a modulation signal by the SK method.

FIG. 4 is an exemplary view for explaining a point of weighting in 8PSK soft decision of the embodiment.

FIG. 5 shows a third embodiment according to the present invention.
FIG. 4 is an IQ plan view for explaining a soft decision method of a modulation signal according to the DAPSK method.

FIG. 6 is an I′Q ′ plan view for explaining a point of the phase soft decision of the embodiment.

FIG. 7 is an exemplary view for explaining the manner of weighting in 16 DAPSK soft decision according to the embodiment;

FIG. 8 is a view for explaining various methods of the position of the normalized amplitude value and the method of engraving between them in the embodiment.

FIG. 9 shows a fourth embodiment according to the present invention.
FIG. 2 is a block circuit diagram showing a configuration of a receiving apparatus using a soft decision circuit for a modulated signal based on the PSK method.

FIG. 10 shows a fifth embodiment according to the present invention.
FIG. 3 is a block circuit diagram showing a configuration of a receiving apparatus using a soft decision circuit for a modulated signal based on the QPSK method.

FIG. 11 shows a sixth embodiment according to the present invention.
FIG. 3 is a block circuit diagram showing a configuration of a receiving apparatus using a soft decision circuit for a modulated signal based on the QPSK method.

FIG. 12 shows a seventh embodiment according to the present invention,
FIG. 2 is a block circuit diagram showing a configuration of a receiving apparatus using a soft decision circuit for a modulated signal based on the PSK method.

FIG. 13 shows an eighth embodiment according to the present invention;
FIG. 3 is a block circuit diagram showing a configuration of a receiving apparatus using a soft decision circuit for a modulated signal based on the QPSK method.

FIG. 14 is an IQ plan view and an I′Q ′ plan view for explaining a point of the phase soft decision in the embodiment.

FIG. 15 shows a ninth embodiment according to the present invention,
FIG. 3 is a block circuit diagram showing a configuration of a receiving apparatus using a soft decision circuit for a modulated signal based on the QPSK method.

FIG. 16 shows a tenth embodiment according to the present invention.
FIG. 2 is a block circuit diagram showing a configuration of a receiving apparatus using a soft decision circuit for a modulated signal based on the π / 4 shift DQPSK method.

FIG. 17 is an IQ plan view and an I ″ Q ″ plan view for explaining a point of the phase soft decision in the embodiment.

FIG. 18 is an IQ plan view and I ′ for describing soft decision of a phase modulation signal according to a conventional DQPSK method.
Q 'plan view.

FIG. 19 is a block circuit diagram for describing a configuration of a receiving apparatus according to a conventional QPSK soft decision method.

FIG. 20 is an I′Q ′ plan view and an I ″ Q ″ plan view showing an example of soft decision in the receiving apparatus.

FIG. 21 is an image diagram of the reliability of phase information due to a change in a reception level in the receiving device.

FIG. 22 is a block circuit diagram showing a configuration of a PSK receiving apparatus using a conventional delay detection circuit.

[Explanation of symbols]

11: delay detection circuit, 111: IQ → phase conversion circuit, 1
12 ... Symbol delay circuit, 113 ... Subtractor, 114 ... Phase soft decision circuit, 12 ... Viterbi decoder for QPSK, 21 ...
Delay detection circuit, 211... IQ → phase-amplitude conversion circuit, 21
2,217 ... symbol delay circuit, 213 ... subtractor, 21
4 ... Phase soft decision circuit, 215, 216 ... Multiplier, 218
... weighting coefficient generation circuit, 22 ... DQPSK Viterbi decoder, 31 ... soft decision circuit, 311 ... phase-amplitude conversion circuit,
312: phase soft decision circuit, 3121: phase delay circuit, 3
122 ... Subtractor, 3123 ... Phase difference soft decision circuit, 313
... weight generation circuit, 3131 ... amplitude delay circuit, 313
2 ... weighting coefficient generation operation circuit, 314 ... weighting circuit, 41 ... soft decision circuit, 411 ... phase soft decision circuit, 41
2 ... weight generation circuit, 413 ... weighting circuit, 42 ...
Viterbi decoder, 51: Soft decision circuit, 511: Soft phase decision circuit, 512: Weighting circuit, 52: Viterbi decoder, 6
1: Soft decision circuit, 611: Phase / amplitude conversion circuit, 612:
Phase soft decision circuit, 613: weighting circuit, 62: speed conversion circuit, 63: Viterbi decoder, 71: delay detection circuit, 7
11: amplitude detection circuit, 712: clipping circuit, 71
3 ... Symbol delay circuit, 714 ... Complex multiplication circuit, 72 ...
Phase shift circuit, 73 Viterbi decoder, 81 Delay detection circuit, 811 Clipping circuit, 812 Symbol delay circuit, 813 Complex multiplication circuit, 82 Phase shift circuit, 83 Viterbi decoder, 91 Delay detection circuit , 911
... amplitude detection circuit, 912 ... clipping circuit, 913 ...
Symbol delay circuit, 914 complex multiplying circuit, 92 Viterbi decoder.

 ──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Masami Aizawa 5-2-2-8 Akasaka, Minato-ku, Tokyo Inside the Next Generation Digital Television Broadcasting System Research Laboratories (72) Inventor Hidenori Tsuboi Shinsugita, Isogo-ku, Yokohama-shi, Kanagawa 8 Tochiba Multimedia Technology Research Laboratories

Claims (10)

[Claims] 1. PSK for transmitting information by a phase component
When receiving a signal of the (Phase Shift Keying) system, the soft decision value of each bit is weighted by the weighting coefficient obtained from the amplitude information to the phase soft decision value of each bit obtained from the phase component. PSK characterized by the following:
Soft decision method. 2. The received signal is multi-level PSK, multi-level D
PSK (Differential Phase Shift Keying), multi-level APSK (Amplitude Phase Shift Keying)
: Differential amplitude phase shift keying), multi-level DAPSK (Differe
2. The weighting coefficient is an amplitude value obtained by clipping a value equal to or greater than a normalized amplitude when modulated by any one of ntial Amplitude Phase Shift Keying (differential amplitude phase shift keying). PSK soft decision method. 3. When the received signal is modulated by a multi-level DPSK or a multi-level DAPSK method, the weighting coefficient is calculated by calculating an arithmetic mean of two symbol amplitudes, a geometric mean of two symbols, and a product of two symbol amplitudes. 2. The P according to claim 1, wherein the amplitude is one of a smaller one of the two symbols and a current amplitude of the two symbols.
SK soft decision method. 4. An amplitude used as the weighting coefficient,
4. The soft decision method of PSK according to claim 3, wherein a value equal to or greater than the normalized amplitude is a clipped amplitude value. 5. PSK for transmitting information by a phase component
(Phase Shift Keying) phase-amplitude conversion means for extracting phase information and amplitude information from a signal by the method, and phase soft decision means for obtaining a phase soft decision value for each bit from the phase information obtained by this means And a weighting means for weighting the phase soft decision value obtained by the means with a predetermined weighting coefficient and outputting the weighted value. 6. PSK for transmitting information by a phase component
(Phase Shift Keying) phase-amplitude conversion means for extracting phase information and amplitude information from a signal by the method, and phase soft decision means for obtaining a phase soft decision value for each bit from the phase information obtained by this means Weighting coefficient generation means for generating a weighting coefficient based on the amplitude information obtained by the phase amplitude conversion means; and a phase soft decision value obtained by the phase soft decision means obtained by the weighting coefficient generation means. A weighting means for weighting with a weighting coefficient and outputting a soft decision value, the receiving apparatus using a PSK soft decision circuit. 7. When the signal to be received is modulated by a multi-level DPSK or a multi-level DAPSK system, the weighting coefficient generation means calculates an arithmetic mean of amplitudes of two symbols, a geometric mean of two symbols, and a symbol average of two symbols. 7. The receiver according to claim 6, wherein the product of the amplitudes is one of the smaller one of the two symbols and the current amplitude of the two symbols. 8. The receiving apparatus using a PSK soft-decision circuit according to claim 6, wherein said weighting coefficient generating means sets the amplitude used as the weighting coefficient to an amplitude value obtained by clipping a normalized amplitude or more. 9. Clipping means for clipping an input signal, delay means for delaying the output of the clipping means, and complex multiplication means for performing conjugate complex multiplication of the output of the clipping means and the output of the delay means. A delay detection circuit, comprising: 10. An amplitude detecting means for obtaining an amplitude from an input signal, a clipping means for performing clipping on the input signal using an output of the amplitude detecting means, a delay means for delaying an output of the clipping means, and a clipping means. And a complex multiplying means for performing a conjugate complex multiplication of an output of the delay detecting means and an output of the delay means.