KR101011499B1 - System for 16/32 apsk modulation and demodulation by using pragmatic trellis coding - Google Patents

Abstract

PURPOSE: A 16/32 APSK modulation method and a demodulation system thereof are provided to improve the performance of a channel resource in an allocated channel environment by correcting errors regarding the channel environment during a demodulation process. CONSTITUTION: A convolution coder(100) outputs a convolution signal having a plurality of bits by convolution-coding at least one bit coded bit signal. An APSK modulator(110) modulates a non-coded bit signal among the convolution signal and the input data by an APSK method. The coded bit signal comprises a least significant bit among the input data. The same number of constellation points is arranged in the inner circuit and the outer circle on a constellation when the convolution signal and the non-coded bit signal are modulated to 16APSK.

Description

SYSTEM FOR 16/32 APSK MODULATION AND DEMODULATION BY USING PRAGMATIC TRELLIS CODING}

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a wireless communication system, and more particularly, to a demodulation system of an amplitude phase shift keying (APSK) system, which is mainly used in a nonlinear channel environment, and a pragmatic convolutional encoder (K = 7). Trellis Coded Modulation (TCM) method with R = 1/2 and G = [171oct 133oct]) secures additional coding gain without increasing bandwidth or loss of data rate. To correct the error.

Non-linear channel environments generally cause transmission performance degradation in communication systems, especially in high-density modulation schemes such as QAM. In the case of the APSK-based modulation scheme, which has recently been adopted as a standard for satellite communication systems, it has an advantage of reducing performance deterioration in a nonlinear channel environment due to the symbol arrangement form circularly distributed on constellation. Accordingly, in case of APSK, constellation distortion occurring in nonlinear channel environment can be alleviated. As such, APSK is more widely used than QAM in nonlinear channel environment.

Meanwhile, in a wireless communication system such as mobile communication or satellite communication, an error correcting code method for correcting an error occurring in a channel when a wireless signal is transmitted and received, and when an error occurs in a channel after adding an additional symbol to the wireless signal, The Forward Error Correction (FEC) scheme, which detects or corrects channel errors using algebraic properties at the receiving end, is mainly used.

The FEC can be divided into a block code and a convolution code. A convolutional code using a convolution code uses a method of encoding based on a correlation between a current input and a past input. As an encoder, it is effective for correcting random errors.

In addition, additional information for error correction is added to data that is error corrected and encoded by the convolutional encoder, and thus the amount of data that can be transmitted per limited bandwidth is reduced. Therefore, the encoding and modulation are performed separately. By combining, the Trellis Coded Modulation (TCM) scheme is used as a transmission method that can obtain a large code gain without reducing the data rate or increasing the bandwidth.

Accordingly, the present invention intends to apply a Pragmatic TCM technique to a coding scheme for correcting an error in an APSK modulation scheme for a nonlinear channel environment.

The present invention is the least significant bit of three-bit or four-bit data in a wireless communication system, such as mobile communication, satellite communication using the Trellis Coded Modulation (TCM) method to correct errors caused by the channel environment when transmitting a radio signal By convolutionally encoding only (LSB: Least, Significant Bit) into 2 bits and combining them with unencoded bit signals, they are modulated into 16 APSK and 32 APSK, respectively, to provide a system that improves the performance of allocated channel resources. There is a purpose.

In addition, an object of the present invention is to provide a demodulation system for an APSK modulated system using a TCM code, and to provide a system capable of correcting an error occurring during APSK modulation.

The 16/32 amplitude phase shift modulation system using the first invention Pragmatic TCM of the present application for achieving the above object comprises convolutional coding (K = 7, R = 1/2, A convolutional encoder for outputting a convolutional signal of 2 bits by G = [171oct 133oct]); And an APSK modulator for modulating an unsigned bit signal of the convolution signal and the input data into an APSK.

In addition, the 16/32 amplitude phase shift demodulation system using the second invention TCM of the present invention is an extraction unit for extracting the amplitude and phase information of the symbol through the I and Q value of the received symbol; A zone determination unit that determines a position within the constellation diagram of the symbol based on the amplitude and phase information of the symbol extracted by the extraction unit; A soft decision metric determination unit for receiving the amplitude and phase information of the extraction unit and the zone determination result of the zone determination unit, and giving a soft decision value for the amplitude and phase of the symbol; A Viterbi decoder which receives the soft decision values of amplitude and phase of the soft decision metric determination unit and decodes them into one bit of information bits; A convolution encoder for receiving the information bits from the non-coder decoder and outputting the convolution signals of two bits; An amplitude phase decoder configured to receive the determination result of the zone determination unit and the convolution signal to determine an amplitude and a phase of the symbol; And a determination unit which receives the amplitude and phase data of the amplitude phase decoder and determines information data corresponding thereto.

As described above, the present invention can correct errors caused by the channel environment during demodulation by using the Pragmatic TCM method in the signal transmission process using APSK modulation, and improve the performance of channel resources in the allocated channel environment. It can be effective.

1 is a block diagram of an APSK modulator using Pragmatic TCM according to an embodiment of the present invention;
2 is a constellation diagram according to 16APSK modulation results using Pragmatic TCM according to an embodiment of the present invention;
3 is a constellation diagram according to a 32APSK modulation result using Pragmatic TCM according to an embodiment of the present invention;
4 is a block diagram of an APSK demodulator using Pragmatic TCM, in accordance with an embodiment of the present invention;
5 is a segmentation based on the radius (signal magnitude) in 16APSK,
6 is a section based on phase in 16APSK,
7 is a segmentation based on the radius in 32 APSK,
8 is a section based on phase at 32APSK,
9 is a method for setting an amplitude level for soft decision decoding in 16APSK according to an embodiment of the present invention;
10 is a method for setting an amplitude level for soft decision decoding in 32 APSK according to an embodiment of the present invention;
11 is a diagram illustrating a method for setting phase levels within constellations for soft decision decoding in 16APSK according to one embodiment of the present invention;
12 is a diagram illustrating a method of setting phase levels within constellations for soft decision decoding in 32APSK according to one embodiment of the present invention;
13 is a method of determining an amplitude using a region of a received symbol and coding bits;
FIG. 14 is a diagram illustrating Signal to Noise Ratio (Eb / No) according to BER (Bit Error Rate) of 16/32 APSK and uncoded 8PSK using TCM.

Hereinafter, exemplary embodiments of the present invention will be described with reference to the accompanying drawings. In adding reference numerals to components of the following drawings, it is determined that the same components have the same reference numerals as much as possible even if displayed on different drawings, and it is determined that they may unnecessarily obscure the subject matter of the present invention. Detailed descriptions of well-known functions and configurations will be omitted.

1 is a block diagram of an APSK modulator using a TCM according to an embodiment of the present invention, the present invention includes a convolutional encoder 100 and an APSK modulator 110 for modulating with APSK.

As shown in the figure, the least significant bit (LSB: Least Significant Bit) of the 3-bit input data is defined as a 1-bit encoded signal E1 to distinguish it from the remaining unencoded bit signals I0 and I1. The encoder 100 may output the 1-bit encoded bit signal E1 as the 2-bit convolution signals b0 and b1.

The characteristic of the convolutional encoder 100 is preferably K = 7, R = 1/2, G = [171oct, 133oct], where K is the constraint length and R is the encoder rate. ), G stands for Generator Polynomial.

The APSK modulator 110 combines the convolution signals b0 and b1 with the unsigned bit signals I0 and I1 to modulate 16 APSK. Accordingly, the upper two bits of the 4-bit data according to the 16APSK modulation may be the information bits I0 and I1, and the lower two bits may be arranged the convolution signals b0 and b1.

If the input data is 4 bits, the APSK modulator 110 performs 32 APSK modulation. That is, the least significant bit (LSB: Least Significant Bit) is used as the coded signal E1, and the 2-bit convolution signals b0 and b1 are output through the convolutional encoder 100, and the unsigned bit signals I0, In combination with I1 and I2), it modulates to 32 APSK. Accordingly, the upper three bits of the 5-bit data according to the 32APSK modulation may be the information bits I0, I1, and I2, and the lower two bits may include the convolution signals b0 and b1.

The input data can be implemented in constellations by outputting the modulated signals of the I and Q channels through APSK modulation and displaying them in complex coordinates.

2 shows 16APSK constellation using TCM according to an embodiment of the present invention.

As shown in the figure, in the 16APSK constellation, the first convolution signal b0 is mapped to the amplitude (or amplitude) of the constellation, the inner circle is 0, and the outer circle is 1. Or vice versa. The second convolution signal b1 is mapped to the phase of the constellation and modulated, and the b1 value of the constellation point having an arbitrary phase is designated as 0 or 1, so that adjacent constellation points of circles having the same radius are respectively. Define to have different values. In this case, eight constellation points are arranged at the inner circle and eight at the outer circle, so that the phase difference between the constellation points in the circle having the same radius is π / 4.

In this case, the first convolution signal b0 or the second convolution signal b1 may be mapped between the same neighboring constellation points so as to have a gray code, so that data errors during demodulation may be minimized. That is, the constellations have the same phase but different radii, and have the same convolutional points with the same constellation point but the same constellation with the same constellation but the same phase, and the second convolution signal b1 has the same gray code. You can deploy a store.

Each constellation point has a different amplitude and phase, and can be distinguished by complex coordinates I and Q.

The radius ratio between the source and the source in the constellation may be 3: 5. This is to maximize the demodulation performance in the demodulation process. If the standard deviation of the normalized constellation point spacing is large, the probability of error in determining the region during demodulation increases. Therefore, it is necessary to determine the radius of an appropriate value, and demodulation performance when the distance d1 between neighboring constellation points of the inner circle and the outer circle and the minimum distance d2 that can distinguish the phase from the neighboring symbol in the inner circle symbol are the same. This can be maximum. At this time, the radius ratio between the source and the source is 2.9: 5. In this case, in consideration of the advantages of signal processing, when the radius is an integer ratio and the radius of the inner circle and the outer circle is set to 3 and 5, the deviation between d1 and d2 may be very small, 0.12.

3 shows constellations according to 32APSK modulation results using TCM according to an embodiment of the present invention.

Similar to the 16APSK constellation, the first convolution signal b0 is mapped to the amplitude of the constellation, and the second convolution signal b1 is mapped to the phase of the constellation and modulated. However, since 32APSK modulation has four circles in constellation, unlike in the case of 16APSK, 0, 1, 0, 1 or vice versa is sequentially assigned from the innermost circle toward the outermost circle.

If the four circles in the 32APSK constellation are defined as the first to fourth circles, starting with the innermost circle and the outermost circle, respectively, four constellation points are defined for the first and fourth circles. Twelve constellation points can be placed in circles and third circles. Thus, in the first and fourth circles, the distance between neighboring constellation points is π / 2, and in the second and third circles, the distance between neighboring constellation points is π / 6. Like 16APSK, b1 is mapped to the phase of the constellation, and b1 of a constellation point having an arbitrary phase is designated as 0 or 1, so that neighboring constellation points of a circle having the same radius have different values.

In addition, the first convolution signal b0 or the second convolution signal b1 may be mapped to have a gray code between neighboring constellation points to minimize data errors during demodulation.

At this time, the radius ratio of each circle in the constellation may be 1: 2: 3: 4, preferably, the size may be 2, 4, 6, 8, respectively. Like the 16APSK, if the standard deviation of the normalized constellation point spacing is large, the probability of error is increased in determining the demodulation zone, so that it has a minimum standard deviation value.

Each constellation point has a different amplitude and phase, and can be distinguished by complex coordinates I and Q.

4 is a block diagram of an APSK demodulator using a TCM according to an embodiment of the present invention.

According to the present embodiment, the APSK demodulator using the TCM includes an extractor 400, a zone determiner 410, a soft decision metric determiner 420, a Viterbi decoder 430, and a convolution encoder 440. , A delay unit 450, an amplitude phase decoder 460, a determination unit 470, and a memory unit 480.

The extractor 400 may extract amplitude and phase information of constellation points in the constellation using complex coordinates I and Q of the received symbol. The extractor 400 may include an amplitude extractor 401 and a phase extractor 402. The amplitude extractor 401 determines the amplitude of the received symbol through the magnitude of the complex coordinate, that is, (I ² + Q ² ) ^1/2 , and the phase extractor 402 determines arctan (Q / I). Through the phase of the received symbol (Phase) can be determined.

The zone determiner 410 receives the amplitude and phase information of the received symbol extracted by the extractor 400 and determines a region in the constellation of the received symbol. The zone determiner 410 may include a radius zone determiner 411 and a phase zone determiner 412.

5 and 6 show zone division based on radius and phase in 16APSK, respectively. FIG. 5 is a zone division based on a radius. When a zone divided according to a radius is referred to as an r zone, two circles of an inner circle and an outer circle exist in 16APSK, and thus, three zones of r0, r1, and r2 exist.

FIG. 6 is a phase-based zone classification. When the zones classified according to the phase are p zones, eight constellation points are disposed at the 16APSK inner and outer circles, and thus there are eight zones from p0 to p7. .

7 and 8 show zone division based on radius and phase in 32APSK, respectively. Since 32 APSK has four circles from the first to the fourth circle, five zones exist from r0 to r4. In addition, since there are four constellation points in the first and fourth circles, there are four zones from p0 to p3, and there are four zones in the second and third circles, so there are twelve zones from p0 to p11. exist.

The radius zone determination unit 411 and the phase zone determination unit 412 determine the r zone and the p zone of the constellation points based on the amplitude and phase results, respectively.

The soft decision metric determination unit 420 receives the amplitude and phase and the zone determination result of the zone determination unit 410 to determine the soft decision metric so that the Viterbi decoder 430 can perform soft decision decoding. do. In this case, the soft decision metric may use a 3-bit 8-level metric.

The soft decision metric determination unit 420 may include a radius soft decision metric determination unit 421 and a phase soft decision metric determination unit 422. The radius soft decision metric determination unit 421 sets the outer zone of the outermost circle to the seventh level and the inner zone of the innermost circle to zero level, and the area between the outermost circle and the innermost circle is repeated from 0 to 7 levels. To give. The soft decision metric is determined based on the given level, amplitude, and r-zone judgment result.

9 and 10 show the level setting according to the amplitude in the constellation to give the soft decision metric in 16APSK and 32APSK, respectively, according to an embodiment of the present invention. The space between each circle in the constellation can be divided into 8 levels from 0 to 7, and the soft decision metric can be given a level by repeating the ascending and descending order based on each circle.

Since the constellation points no longer exist inside or outside the outer circle, the level may not be subdivided.

The phase soft decision metric determination unit 422 divides the zone between adjacent constellation points of the same distal radius from 0 to 7 levels, and determines the soft decision metric based on the phase and p zone determination results.

11 and 12 illustrate level settings according to phases within constellations for imparting soft decision metrics in 16APSK and 32APSK, respectively, according to an embodiment of the present invention. As in the case of amplitude, the ascending and descending order is repeatedly used to give soft decision metrics. In the case of 16APSK, there are eight constellation points in the inner circle and the outer circle, so both the inner circle and the outer circle divide the π / 4 size into eight levels. In the case of 32 APSK, since four constellation points exist in the innermost circle and the outermost circle, and 12 constellation points exist in the circle between them, the size of? / 2 and? / 6 is divided into 8 levels.

The radius soft decision metric determination unit 421 and the phase soft decision metric determination unit 422 beat the 3-bit soft decision metric for b0 and b1 based on the data of the extraction unit 400 and the zone determination unit 410. Transmit to the non-decoder 430.

However, the soft decision metric generated by the soft decision metric determination unit 420 is not limited to 3 bits, and may be added or subtracted according to the number of constellation points and the number of bits of the input data.

The Viterbi decoder 430 receives a soft decision value for b0 and b1 from the soft decision metric determination unit 420, and decodes it into one bit of information bits E1. The Viterbi decoder 430 accumulates and adds the Eucladian distance between the received sequence and the possible sequence, and determines the values of b0 and b1 by selecting the shortest path, and decodes the original. The information bit E1 can be decoded.

The Viterbi decoder 430 transmits the demodulated information bits back to the convolutional encoder 440. The convolutional encoder 440 may code the received information bits E1 into 2-bit convolution signals b0 and b1 and transmit them to the amplitude decoder 461 and the phase decoder 462, respectively. Since the error of the information bits received from the convolutional encoder 440 is corrected through the Viterbi decoder 430, the information of the remaining bits is also demodulated by determining the amplitude and phase by correcting the error of the received convolutional signal. It is.

The delay unit 450 delays the amplitude and phase region determination results of the region determination unit 410 by the decoding time of the Viterbi decoder 430 and the encoding time of the convolutional encoder 440 to the amplitude phase decoder 460. Should be sent. This is to synchronize the convolution signal of the symbol with the symbol to be demodulated when determining the amplitude and phase in the amplitude phase decoder 460.

The amplitude phase decoder 460 may include an amplitude decoder 461 and a phase decoder 462. The amplitude decoder 461 receives an amplitude region determination result and an encoding bit b0 to determine an amplitude, and the phase decoder 462 receives a phase region determination result and an encoding bit b1 to determine a phase. .

FIG. 13 illustrates a method of determining amplitude using a region of a received symbol and coding bits.

According to FIG. 13, assuming that the reception symbol exists in the r2 region, when the first convolution signal b0 is 1, the reception symbol is in the second circle, and when the encoding bit b0 is 0, the reception symbol is in the third circle. It can be determined. If the received symbol exists in the r3 region, it may be determined that the received symbol is in the fourth circle when the first convolution signal b0 is 1 and that the received symbol is in the third circle when b0 is 0. That is, since the circles adjacent to each other during APSK modulation are determined so that the values of the first convolution signal b0 are not equal to each other, the amplitude of the symbol can be determined using the region where the symbols exist and the encoding bit b0.

Similarly, since the second convolution signal b1 is determined so that the value of the second convolution signal b1 is not the same even in the phase difference in the circle of the same amplitude during APSK modulation, the phase symbol of the received symbol and the second convolution signal b1 are used to determine the received symbol. The phase can be determined.

The determination unit 470 may receive the amplitude and phase determined by the amplitude phase decoder 460 and determine an information bit corresponding thereto.

Also, according to another embodiment of the present invention, the APSK demodulation system using the TCM may further include a memory unit in which the determination unit 470 stores a look up table (LUT) for extracting information bits. . The determination unit 470 may extract information bits corresponding to an amplitude and a phase from the LUT of the memory unit.

The information bits stored in the LUT may be stored except the LSB. In this case, demodulation may be completed by receiving the decoded information bits from the Viterbi decoder 430.

FIG. 14 illustrates a bit error rate (BER) according to Signal to Noise Ratio (Eb / No) for performance comparison between 16/32 APSK and uncoded 8PSK using Pragmatic TCM as a computer simulation result.

As shown, when BER is 10 ^-6 , it can be seen that 16APSK is about 2.5dB compared to 8PSK, 32APSK is about 1.0dB SNR compared to 8PSK.

100: convolutional encoder 110: APSK modulator
400: extractor 401: amplitude extractor
402: phase extraction unit 410: zone determination unit
411: radius zone determination unit 412: phase zone determination unit
420: soft decision metric determination unit 421: radius soft decision metric determination unit
422: phase soft decision metric determination unit 430: Viterbi decoder
440: convolution encoder 450: delay unit
460: amplitude phase decoder 461: amplitude decoder
462: phase decoder 470: determination unit

Claims (11)

A convolution encoder which convolutionally encodes at least one bit of the input data and outputs the convolutional signal as a plurality of convolution signals;
An APSK modulator for modulating an unsigned bit signal of the convolution signal and the input data into an APSK
16/32 amplitude phase shift modulation system using a pragmatic TCM comprising a.
The method of claim 1,
The coded bit signal comprises a least significant bit (LSB) of the input data 16/32 amplitude phase shift modulation system using a pragmatic TCM.
The method of claim 1,
When the convolutional signal and the unsigned bit signal are modulated by 16 APSK, 16/32 amplitude phase shift using a pragmatic TCM is characterized in that the same number of constellation points are arranged in each of the inner and outer circles of the constellation diagram. Modulation system.
The method of claim 1,
When the convolutional signal and the unsigned bit signal are modulated with 32 APSK, the same number of constellation points are arranged in each of the innermost and outermost circles in the constellation diagram, and each circle between the innermost and outermost circles 16/32 amplitude phase shift modulation system using a pragmatic TCM, characterized in that the same number of constellation points are arranged.
The method of claim 1,
When the convolution signal and the unsigned bit signal are modulated with 16 APSK, the radius ratio between the source and the source in the constellation is 3: 5, characterized in that the 16/32 amplitude phase shift modulation system using a pragmatic TCM .
The method of claim 1,
When the convolution signal and the unsigned bit signal are modulated with 32 APSK, the radius ratio of four circles in the constellation is 1: 2: 3: 4, which is 16/32 amplitude using a pragmatic TCM. Phase Shift Keying System.
The method of claim 1,
16/32 amplitude phase shifting modulation system using a pragmatic TCM characterized in that the first bit of the convolution signal is mapped to the amplitude, the remaining bits are mapped to the phase and modulated by APSK.
The method of claim 1,
16/32 amplitude phase shifting modulation system using a pragmatic TCM, characterized in that the constellation according to the APSK modulation result, gray coding between the nearest constellation point of the same radius or phase.
An extraction unit for extracting amplitude and phase information of the symbol through I and Q values of a received symbol;
A zone determination unit that determines a position within the constellation diagram of the symbol based on the amplitude and phase information of the symbol extracted by the extraction unit;
A soft decision metric determination unit configured to receive amplitude and phase information of the extraction unit and zone determination results of the zone determination unit, and to give a soft decision value for the amplitude and phase of the symbol;
A Viterbi decoder which receives the soft decision values of amplitude and phase of the soft decision metric determination unit and decodes them into one bit of information bits;
A convolution encoder for receiving the information bits from the non-coder decoder and outputting the convolution signals of two bits;
An amplitude phase decoder configured to receive the determination result of the zone determination unit and the convolution signal to determine an amplitude and a phase of the symbol; And
Determination unit for receiving the amplitude and phase data of the amplitude phase decoder to determine the corresponding information data
16/32 amplitude phase shift demodulation system using a pragmatic TCM.
The method of claim 9,
In transmitting the determination result of the zone determination unit to the amplitude phase decoder, a delay unit for delaying the determination result by the operation time of the Viterbi decoder and the convolution encoder 16/32 using a Fragmatic TCM Amplitude Phase Shift Demodulation System.
The method of claim 9,
And a memory unit for storing a lookup table (LUT) in which the information data is mapped with the amplitude and phase data, wherein the determination unit determines the information data based on the lookup table (LUT). 16/32 amplitude phase shift demodulation system using Matic TCM.

Images