CN117880047A - Soft output 32APSK demodulation method of hard decision threshold - Google Patents

Soft output 32APSK demodulation method of hard decision threshold Download PDF

Info

Publication number
CN117880047A
CN117880047A CN202410087588.2A CN202410087588A CN117880047A CN 117880047 A CN117880047 A CN 117880047A CN 202410087588 A CN202410087588 A CN 202410087588A CN 117880047 A CN117880047 A CN 117880047A
Authority
CN
China
Prior art keywords
hard decision
constellation point
hard
constellation
32apsk
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202410087588.2A
Other languages
Chinese (zh)
Inventor
吴广富
陈畅霖
陈莉
向碧群
涂艳丽
李云
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University of Post and Telecommunications
Original Assignee
Chongqing University of Post and Telecommunications
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chongqing University of Post and Telecommunications filed Critical Chongqing University of Post and Telecommunications
Priority to CN202410087588.2A priority Critical patent/CN117880047A/en
Publication of CN117880047A publication Critical patent/CN117880047A/en
Pending legal-status Critical Current

Links

Abstract

The invention belongs to the technical field of wireless communication, and particularly relates to a soft output 32APSK demodulation method of a hard decision threshold, which comprises the following steps: in a 32APSK Gray mapping constellation diagram, analyzing a constellation point set corresponding to each bitAnd (3) withWherein the set of constellation points for which the hard decision is "1" is defined as S 1 The method comprises the steps of carrying out a first treatment on the surface of the Defining a constellation point set with a hard decision of "0" as S 0 Determining a constellation point judgment area; set of constellation points with hard decisions of "0" and "1" in constellation point decision regionAnd (3) withThe distribution characteristics determine corresponding hard decision thresholds; and when soft information corresponding to each bit is calculated, taking absolute values of real parts or imaginary parts of received signals to map to corresponding judgment areas, calculating Euclidean distance from the received signals to a hard judgment threshold, and generating LLR information. The invention replaces the traditional soft output method based on complex curve hard decision threshold distance measurement with a simple straight line hard decision threshold, and can effectively reduce the calculation complexity.

Description

Soft output 32APSK demodulation method of hard decision threshold
Technical Field
The invention belongs to the technical field of wireless communication, and particularly relates to a soft output 32APSK demodulation method of a hard decision threshold.
Background
32APSK, due to its unique constellation structure, small spectral spread and high frequency characteristics of fast roll-off, can effectively solve the problem of low-orbit Satellite channel nonlinearity, and has been applied to the european telecommunication standardization institute second generation Satellite broadcasting standard (Digital Video Broadcasting-Satellite-Second Generation, DVB-S2).
During signal demodulation, the demodulator needs to obtain accurate log-likelihood ratio information (Log Likelihood Ratio, LLR) as an effective input to the decoder. The algorithms for solving LLR information are mainly three: MAP algorithm, max-MAP algorithm, and soft output algorithm based on hard decision threshold distance metric. The MAP algorithm is an optimal soft demodulation algorithm, can obtain the most accurate LLR information, but involves more square sums, exponential operations and logarithmic operations, and has higher implementation complexity; in the Max-MAP algorithm, the exponential operation and the logarithmic operation are effectively simplified, more square sum calculation is still needed, and the complexity is still higher; as mentioned in literature "an M-APSK soft decision demapping scheme based on hard decision boundary" by a learner Lv Tingting, the soft output algorithm based on the hard decision threshold distance metric uses the boundary between constellation points of which hard decision is "1" and "0" as a hard decision threshold, uses the euclidean distance from a received signal to the constellation point hard decision threshold as a metric, and multiplies the metric value by a noise variance constant to obtain LLR information, but for a high-order 32APSK modulation system, the problem that the hard decision threshold is complex is existed, so that more comparison operation is needed.
Disclosure of Invention
In order to solve the technical problems, the invention provides a soft output 32APSK demodulation method of a hard decision threshold, which comprises the following steps:
s1: in the 32APSK gray mapping constellation, a set of constellation points with hard decisions of "1" is defined asDefine the constellation point set with hard decision "0" as +.>Wherein n= {1,2, …,5} represents the bit number of the 32APSK signal, and the constellation point set corresponding to each bit is analyzed>And->Determining a constellation point decision region;
s2: set of constellation points with hard decisions of "0" and "1" in constellation point decision regionAnd->The distribution characteristics determine corresponding hard decision thresholds;
s3: and when soft information corresponding to each bit is calculated, taking absolute values of real parts or imaginary parts of received signals to map to corresponding judgment areas, calculating Euclidean distance from the received signals to a hard judgment threshold, and generating LLR information.
Preferably, the constellation point set corresponding to each bit is analyzedAnd->Determining a constellation point decision region, comprising:
s11: if constellation point setAnd->The decision area is set as a first quadrant if the received signal is positioned in other quadrants, the received signal can be correspondingly mapped to the first quadrant for correlation processing, and LLR information values of the received signal cannot be changed;
s12: if constellation point setAnd->Only symmetric about the real or imaginary axis, respectively, the decision region can be reduced to a single dimension.
Preferably, the constellation point sets are hard-decided as "0" and "1" according to the constellation point decision regionAnd->The distribution characteristics, confirm the corresponding hard decision threshold, include:
s21: if the decision region is in the first quadrant, the hard decision threshold is set toAnd->A curve formed by the midpoint position of the constellation point is simplified;
s22: if the decision region can be reduced to a single dimension, i.e. if the constellation point setAnd->The hard decision threshold is set as the virtual axis only about the real axis symmetry respectively; if constellation point set->And->The hard decision threshold is set to the real axis only symmetrically about the imaginary axis, respectively.
Preferably, simplifying the curve hard decision threshold of the decision region in the first quadrant includes:
for the curve hard decision threshold of the decision region in the first quadrant, the curve hard decision threshold of the decision region in the first quadrant can be reduced to a straight line parallel to the real axis or the imaginary axis by using the average value of the real part value or the imaginary part value of the constellation points at the two sides of the decision threshold.
The invention has the beneficial effects that:
the invention replaces the traditional soft output method based on complex curve hard decision threshold distance measurement with a simple straight line hard decision threshold, and can effectively reduce the calculation complexity, in the embodiment, the calculation complexity of the method is only about 45% of that of the traditional method by verification, and the loss of the bit error rate performance is about 0.03dB compared with that of the traditional method.
Drawings
FIG. 1 is a 32APSK constellation point distribution diagram of the present invention;
FIG. 2 is a schematic diagram of the present invention b 1 Bit hard decision threshold diagram;
FIG. 3 is a schematic diagram of the present invention b 2 Bit hard decision threshold diagram;
FIG. 4 is a schematic diagram of the present invention b 3 Bit hard decision threshold diagram;
FIG. 5 is a schematic diagram of the present invention b 4 Bit hard decision threshold diagram;
FIG. 6 is a schematic diagram of the present invention b 5 Bit hard decision threshold diagram;
FIG. 7 is a schematic diagram of the present invention b 2 A bit-simplified hard decision threshold diagram;
FIG. 8 is a schematic diagram of the present invention b 3 A bit-simplified hard decision threshold diagram;
fig. 9 is a graph of bit error rate performance versus the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
A soft output 32APSK demodulation method of hard decision threshold includes:
s1: in the 32APSK gray mapping constellation, a set of constellation points with hard decisions of "1" is defined asDefine the constellation point set with hard decision "0" as +.>Wherein n= {1,2, …,5} represents the bit number of the 32APSK signal, and the constellation point set corresponding to each bit is analyzed>And->Determining a constellation point decision region;
s2: set of constellation points with hard decisions of "0" and "1" in constellation point decision regionAnd->The distribution characteristics determine corresponding hard decision thresholds;
s3: and when soft information corresponding to each bit is calculated, taking absolute values of real parts or imaginary parts of received signals to map to corresponding judgment areas, calculating Euclidean distance from the received signals to a hard judgment threshold, and generating LLR information.
A 32APSK symbol is formed of 5 bits, denoted b 1 b 2 b 3 b 4 b 5 Wherein b 1 Representing the most significant bit, b 5 Representing the least significant bit. The set of 32 constellation points is denoted s= { S 1 ,s 2 ,…,s 32 Set of constellation points where hard decision is "1" is defined as S 1 The method comprises the steps of carrying out a first treatment on the surface of the Defining a constellation point set with a hard decision of "0" as S 0 。
The 32APSK constellation selected in the present case is shown in figure 1, and the radius value of the first loop of the 32APSK is r 1 The radius value of the second ring is r 2 The radius value of the third ring is r 3 The ratio of the radius of each ring relative to the radius of the first ring is respectively selected as [1,2.61,4.53 ]]The FEC coding scheme is selected as a Turbo code with a code rate of 1/3, the code length is 3840, the wireless channel is selected as an additive Gaussian white noise channel, and the noise variance is sigma 2 。
A 32APSK signal received by a receiving end is expressed as α=γ+jλ, where γ and λ represent a real part and an imaginary part of the 32APSK signal, respectively, and j is a complex unit. Order theRepresenting the amplitude value of a 32APSK signal, +.>Representing the phase angle after mapping the absolute value of the real and imaginary parts of the received signal to the first quadrant. Demodulation is performed by:
s1: analyzing a constellation corresponding to each bit in a 32APSK constellation diagramPoint setAnd->And determining the constellation point decision region. Fig. 2 to 6 respectively show constellation point distribution cases of 32APSK five-bit bits in the embodiment, wherein constellation points marked with "·" represent that their hard decisions are "0"; the constellation points marked with an "x" represent their hard decisions as "1".
b 1 b 2 b 3 Bit of bits: FIGS. 2, 3 and 4 show b, respectively 1 b 2 b 3 Bit constellation point distribution feature, whereinAre all satisfied symmetrically about the real and imaginary axes, and thus for b 1 b 2 b 3 The bit can set the decision area as the first quadrant, and if the received signal is in other quadrants, the received signal can be mapped to the first quadrant correspondingly for correlation processing, and the LLR information value of the received signal is not changed.
b 4 Bit of bits: FIG. 5 shows b 4 Bit constellation point distribution feature, whereinAnd->Satisfy the term about the imaginary axis, and thus for b 4 The bit bits may reduce their decision region to a single dimension.
b 5 Bit of bits: FIG. 6 shows b 5 Bit constellation point distribution feature, whereinAnd->Satisfying the relationshipSymmetrical about the real axis, thus for b 5 The bit bits may reduce their decision region to a single dimension.
S2: set of constellation points with hard decisions of "0" and "1" in constellation point decision regionAnd->And (5) distributing the characteristics and determining a corresponding hard decision threshold.
b 1 Bit of bits: the decision area is located in the first quadrant, seeThe boundary line is a circular ring formed by the median value of the radii of the second ring and the third ring, so that the hard decision threshold can be expressed as +.>θ represents the phase angle, as shown by the curve marked with a dash-dot line in fig. 2.
b 2 Bit of bits: the decision area is in the first quadrant, and the hard decision threshold is selected asAnd->Curve T formed by the midpoint position of constellation points 2 ' and T 2 "as shown by the curve marked with a dash-dot line in fig. 3.
b 3 Bit of bits: its decision regionThe domain is located in the first quadrant, and its hard decision threshold is selected asAnd->Curve T formed by the midpoint position of constellation points 3 ' and T 3 "as shown by the curve marked with a dash-dot line in fig. 4.
b 4 Bit of bits: because of its constellation pointsAnd->With respect to virtual axis symmetry, the decision region can be reduced to a single dimension, so the hard decision threshold T 4 The real axis is chosen as indicated by the curve marked with a dash-dot line in fig. 5.
b 5 Bit of bits: because of its constellation pointsAnd->With respect to real axis symmetry, the decision region can be reduced to a single dimension, so the hard decision threshold T 5 The virtual axis is selected as indicated by the curve marked with a dash-dot line in fig. 6.
B in S2 2 And b 3 The hard decision threshold of (2) is complex, and the hard decision threshold can be simplified into the average value of the real part value or the imaginary part value of constellation points at both sides of the decision thresholdA straight line parallel to the real or imaginary axis.
b 2 Bit of bits: for T' 2 With the size parallel to the real axis as the constellation point s 8 ,s 12 ,s 16 ,s 20 ,s 28 Straight line E 'of imaginary mean' 2 Alternatively, the method may be used; for T' 2 Divided into two sections, the first section being formed by constellation points s parallel to the real axis 0 ,s 4 ,s 8 ,s 12 Straight line E' of imaginary mean value 2 Alternatively, the method may be used; the second segment is parallel to the virtual axis and has a constellation point s 4 ,s 24 Straight line E 'of real part mean' 2 Alternatively, as shown by the curve marked with a dash-dot line in fig. 7. The values of the simplification threshold may be expressed as follows:
E′ 2 ＝imag(s 8 +s 12 +s 16 +s 20 +s 28 )/5
E″ 2 ＝imag(s 0 +s 4 +s 8 +s 12 )/4
E″′ 2 ＝real(s 4 +s 24 )/2
b 3 bit of bits: for T' 3 With the size parallel to the imaginary axis as the constellation point s 0 ,s 4 ,s 8 ,s 12 ,s 16 ,s 20 Straight line E 'of real part mean' 3 Alternatively, the method may be used; for T' 3 Divided into two sections, the first section is divided into constellation points s parallel to the imaginary axis 4 ,s 24 Straight line E' of real part mean value 3 Alternatively, the method may be used; the second segment is parallel to the real axis and has a constellation point s 24 ,s 28 Straight line E 'of imaginary mean' 3 Alternatively, as shown by the curve marked with a dash-dot line in fig. 8. The values of the simplification threshold may be expressed as follows:
E′ 3 ＝real(s 0 +s 4 +s 8 +s 12 +s 16 +s 20 )/6
E″ 3 ＝real(s 4 +s 24 )/2
E″′ 3 ＝imag(s 24 +s 28 )/2
s3: and (3) calculating soft information corresponding to each bit according to the hard decision threshold designed in the step (S2). The real part or the imaginary part of the received signal alpha is first taken as an absolute value to be mapped to a corresponding decision region, and then the euclidean distance of the received signal to a hard decision threshold is calculated to generate LLR information.
For b 1 Bit from amplitude value beta of APSK signal to median radius of second ring and third ringAs LLR information:
for b 2 And the bit is used for taking absolute values of a real part and an imaginary part of the APSK signal so as to map to the first quadrant. When the hard decision threshold is not simplified, its LLR information can be expressed as:
in the above formula, |·| represents modulo arithmetic. After simplifying the hard decision threshold, the LLR information can be expressed as:
for b 3 The bit, taking absolute value of both real part and imaginary part of the APSK signal to map to the first quadrant, when the hard decision threshold is not simplified, the LLR information can be expressed as:
after simplifying the hard decision threshold, the LLR information can be expressed as:
for b 4 Bit, hard decision threshold T 4 As a real axis, its LLR information can be expressed as:
for b 5 Bit, hard decision threshold T 5 As an imaginary axis, its LLR information can be expressed as:
thus, LLR information of the APSK signal is obtained.
The present invention simplifies the more complex hard decision threshold and compares the proposed scheme with the existing scheme in table 1:
TABLE 1
The proposal can be seen to keep the advantages of the traditional soft output proposal based on hard decision threshold distance measurement, wherein the complex exponential operation and logarithmic operation are not involved, and the simplified hard decision threshold requires less comparison operation times, and the proposal can effectively reduce the calculation complexity of APSK demodulation. Under the condition that the data consumed by the square and contrast operation are the same, the calculation complexity of the proposed algorithm is only about 45% of that of a traditional soft output scheme based on the hard decision threshold distance measurement. Fig. 9 shows the comparison of the error rate performance of the scheme with the prior scheme, and the error rate performance of the scheme is only reduced by about 0.03dB compared with the conventional soft output scheme based on the hard decision threshold distance measurement on the premise of greatly reducing the computational complexity.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.

Claims (4)

1. A soft output 32APSK demodulation method for a hard decision threshold, comprising:
s1: in the 32APSK gray mapping constellation, a set of constellation points with hard decisions of "1" is defined asDefine the constellation point set with hard decision "0" as +.>Wherein n= {1,2, …,5} represents the bit number of the 32APSK signal, and the constellation point set corresponding to each bit is analyzed>And->Determining a constellation point decision region;
s2: set of constellation points with hard decisions of "0" and "1" in constellation point decision regionAnd->The distribution characteristics determine corresponding hard decision thresholds;
s3: and when soft information corresponding to each bit is calculated, taking absolute values of real parts or imaginary parts of received signals to map to corresponding judgment areas, calculating Euclidean distance from the received signals to a hard judgment threshold, and generating LLR information.
2. The soft output 32APSK demodulation method of claim 1 wherein a constellation point set corresponding to each bit is analyzedAnd->Determining a constellation point decision region, comprising:
s11: if constellation point setAnd->The decision area is set as a first quadrant if the received signal is positioned in other quadrants, the received signal can be correspondingly mapped to the first quadrant for correlation processing, and LLR information values of the received signal cannot be changed;
s12: if constellation point setAnd->Only symmetric about the real or imaginary axis, respectively, the decision region can be reduced to a single dimension.
3. The soft output 32APSK demodulation method of claim 1 wherein the set of constellation points for hard decisions of "0" and "1" in the constellation point decision regionAnd->The distribution characteristics, confirm the corresponding hard decision threshold, include:
s21: if the decision region is in the first quadrant, the hard decision threshold is set toAnd->A curve formed by the midpoint position of the constellation point is simplified;
s22: if the decision region can be reduced to a single dimension, i.e. if the constellation point setAnd->The hard decision threshold is set as the virtual axis only about the real axis symmetry respectively; if constellation point set->And->The hard decision threshold is set to the real axis only symmetrically about the imaginary axis, respectively.
4. The soft output 32APSK demodulation method of claim 1 wherein simplifying a curvilinear hard decision threshold with a decision region in a first quadrant comprises:
for the curve hard decision threshold of the decision region in the first quadrant, the curve hard decision threshold of the decision region in the first quadrant can be reduced to a straight line parallel to the real axis or the imaginary axis by using the average value of the real part value or the imaginary part value of the constellation points at the two sides of the decision threshold.
CN202410087588.2A 2024-01-22 2024-01-22 Soft output 32APSK demodulation method of hard decision threshold Pending CN117880047A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410087588.2A CN117880047A (en) 2024-01-22 2024-01-22 Soft output 32APSK demodulation method of hard decision threshold

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410087588.2A CN117880047A (en) 2024-01-22 2024-01-22 Soft output 32APSK demodulation method of hard decision threshold

Publications (1)

Publication Number Publication Date
CN117880047A true CN117880047A (en) 2024-04-12

Family

ID=90590166

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410087588.2A Pending CN117880047A (en) 2024-01-22 2024-01-22 Soft output 32APSK demodulation method of hard decision threshold

Country Status (1)

Country Link
CN (1) CN117880047A (en)

Similar Documents

Publication Publication Date Title
US7313750B1 (en) Efficient soft decision demapper to minimize viterbi decoder complexity
US6862552B2 (en) Methods, apparatus, and systems employing soft decision decoding
JP3741703B2 (en) Demodulation apparatus and method in communication system using 16-aryQAM modulation system
CN111277536B (en) Soft de-mapping method of DVB-S2X system and digital signal processing system
EP2281376B1 (en) Method and digital communication device for calculating soft information for received QAM symbols
AU2002321932A1 (en) Demodulation apparatus and method in a communication system employing 16-ary QAM
CN113965438A (en) Method for solving soft information under 16APSK high-order modulation mode
JP2005503725A (en) Apparatus and method for calculating input softness value of channel decoder in data communication system
KR100706618B1 (en) Soft decision method on the high order modulation for the iterative decoder and error correction apparatus using it
KR20040023863A (en) Soft Decision Decoder, Apparatus of Generalized Log Likelihood Ratio considered Channel Estimation Errors for Soft Decision Decoding at QAM Signals and method thereof
Ogundile et al. Improved reliability information for rectangular 16-QAM over flat Rayleigh fading channels
CN113225284A (en) 8FSK incoherent soft decision demodulation method for high-performance channel coding and decoding
US5425037A (en) Data transmission apparatus
CN117880047A (en) Soft output 32APSK demodulation method of hard decision threshold
CN115987745A (en) Low-complexity quadrature amplitude modulation cross constellation demapping method
JP2004512742A (en) Method for generating soft bit information from gray coded signal
CN113411279B (en) Reusable Q power demapping method and system based on DVB-S2 system
CN106101052B (en) Low complex degree 128APSK soft de-mapped method based on judgement domain
KR100800882B1 (en) Demodulation apparatus and method in a communication system employing 8-ary psk modulation
CN111865867B (en) 256APSK modulation signal demodulation method, system and device
US7139335B2 (en) Optimal decision metric approximation in bit-soft decisions
US7580480B2 (en) Demodulation method using soft decision for quadrature amplitude modulation and apparatus thereof
KR100556448B1 (en) Method and apparatus for demapping
KR101799849B1 (en) Method for calculating soft decision value for qam signals
TWI387274B (en) A demodulation method using soft decision for quadrature amplitude modulation and apparatus thereof

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination