CN111954263A - Compression transmission method and system of 5GNR baseband digital signal - Google Patents

Compression transmission method and system of 5GNR baseband digital signal Download PDF

Info

Publication number
CN111954263A
CN111954263A CN202010663149.3A CN202010663149A CN111954263A CN 111954263 A CN111954263 A CN 111954263A CN 202010663149 A CN202010663149 A CN 202010663149A CN 111954263 A CN111954263 A CN 111954263A
Authority
CN
China
Prior art keywords
source symbol
information
symbol
information source
coding
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.)
Granted
Application number
CN202010663149.3A
Other languages
Chinese (zh)
Other versions
CN111954263B (en
Inventor
刘�东
张文忠
谭伟
余秋星
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.)
Hangzhou Honglingtong Information Technology Co Ltd
Original Assignee
Hangzhou Honglingtong Information Technology Co Ltd
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 Hangzhou Honglingtong Information Technology Co Ltd filed Critical Hangzhou Honglingtong Information Technology Co Ltd
Priority to CN202010663149.3A priority Critical patent/CN111954263B/en
Publication of CN111954263A publication Critical patent/CN111954263A/en
Application granted granted Critical
Publication of CN111954263B publication Critical patent/CN111954263B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/06—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/04—Protocols for data compression, e.g. ROHC

Abstract

The invention discloses a compression transmission method and a compression transmission system for a 5GNR baseband digital signal. The compression transmission method and the system thereof improve the statistical compression coding technology, and improve the mapping quality from the information source bit to the information source symbol by adaptively selecting the optimal information source symbol mapping set at the data transmission coding side. In addition, a statistical characteristic parameter set is obtained based on the calculation of the optimal source symbol set, the statistical characteristic parameter set is the minimum parameter set for representing the distribution characteristics of the source symbols, the size of the set is far smaller than that of a set formed by each source symbol statistic, and the latter forms the basis of general statistical coding and decoding.

Description

Compression transmission method and system of 5GNR baseband digital signal
Technical Field
The invention relates to the technical field of wireless communication, in particular to a compression transmission method and a compression transmission system for 5GNR baseband digital signals.
Background
In a 5GNR (5-Generation New Radio, fifth Generation New wireless) mobile communication system, there is a high data rate transmission requirement between a DU (Distributed-Unit) and an AAU (active Antenna Unit) of a base station device, and it is necessary to occupy a large transmission bandwidth resource of the base station device, thereby increasing the device manufacturing cost.
With the development of the new generation of communication technology 5G-NR, the system needs to support a larger bandwidth, more sampling data of spatial antennas and a smaller transmission delay, and further needs to transmit more digital information at the same time, which brings a larger cost problem.
In the prior art, the real-time waveform characteristics of physical signal quantity are utilized, or the amplitude of signal change is dynamically adjusted, and then linear or nonlinear quantization is performed.
For example, patent publication No. CN102291773A and publication No. CN103812608A reduce the total digital information amount, or perform transformation operation on the semaphore, for example, patent publication No. CN103348645A and patent publication No. CN105517052A, these methods not only need to add processing on both sides of transmission, but also increase the comprehensive cost, amplify the quantization noise, and damage the quality of the transmitted information; or as described in patent publication No. CN106470054A, the method directly transforms physical semaphore at the transmitting side, then reduces the transformed digital semaphore, and also utilizes the data correlation property of multi-antenna data port to perform data compression, as described in patent publication No. CN106470054A and patent publication No. CN104754644A, which requires high requirement on the transmitting side, and requires high cost for upgrading, and introduces transformation processing that will ultimately directly damage the transmission information quality, and a method for compression transmission, as described in patent publication No. CN107295568A, which directly estimates symbol source probability density at the transmitting side, then calculates compression factor for each statistical interval, corresponds to the compressed data unit, and sets the compressed data bit width, and the receiving side decompresses at different compressed data units. In order to reduce the cost of compression overhead, the statistical interval division granularity cannot be too fine, so that the data compression ratio is directly influenced.
Disclosure of Invention
The invention aims to provide a compression transmission method of a 5GNR baseband digital signal and a system thereof aiming at the defects of the prior art.
The compression transmission method comprises the following steps:
step 1, a sending processing module is sent in a sending period, original bit stream information collected successively is processed, and coded bit stream information is output successively:
step 1.1, the symbol grouping and dividing submodule and a decoding party agree M information source symbol dividing modes, wherein, the M information source symbol dividing modes are every continuous NmBinary symbol of bit b (i N)m),b(i*Nm+1),…,b((i+1)*Nm-1) } to a source symbol Xm,iThe value corresponding to the source symbol is Xm,iWherein i is …, -1,0,1, …; m is 1, …, M;
step 1.2, in a statistical period time T, the statistical characteristic calculation submodule performs statistical information source symbol set { X }m(n)|n＝0,…,2Nm-1} for each source symbol Xm(n) number of occurrences fm(n) according to fm(n) estimating a source symbol statistical property parameter { beta }m(1),…,βm(q) }, source symbol statistical property parameters calculated for the gaussian distribution of source symbols, mean μ and variance σ of the gaussian distribution, respectively2The mean μ and variance σ of the Gaussian distribution are calculated according to the following equations (1) and (2), respectively2：
Figure BDA0002579362020000021
Figure BDA0002579362020000022
With gaussian distribution, there are only two statistical parameters, i.e. q is 2, βm(1)＝μ,βm(2) When N is ═ σmWhen the data flow is gradually reduced, all the information source symbols gradually tend to be uniformly distributed in the information source symbol data flow, and only one characteristic parameter exists at the moment;
step 1.3, the optimal information source symbol selection submodule selects an information source symbol division mode with optimal coding efficiency in M information source symbol division modes to be selected according to the following formula (3), wherein the information source symbol division mode with the optimal coding efficiency is the mth*The method comprises the following steps:
Figure BDA0002579362020000023
m th*The information source symbol set corresponding to the information source symbol division mode is { Xm*(n)|n＝1,…,2Nm*The statistical property parameter set is { beta }m *(1),…βm *(q) }, the mapped source symbol stream is Xm*,i；
Step 1.4, the optimal information source symbol selection submodule selects the sub-module according to the statistical characteristic parameter set { betam *(1),…βm *(q) }, sequential symbol set { X) in increasing order of n*(n)|n＝1,…,2NCalculating the n information source symbol X ordered in the information source symbol set*Statistical probability of (n)
Figure BDA0002579362020000024
The calculation is performed from a set of statistical property parameters, i.e. the following equation (4):
Figure BDA0002579362020000031
in the formula: x*(n) the cumulative probability in the sequential symbol set is the following equation (5):
Figure BDA0002579362020000032
C0＝0，Cnsource symbol X representing the nth input encoder*(n) cumulative probabilities in the sequential symbol set;
step 1.5, compressing the coding submodule at the coding time omega>Within T, for source symbol stream Xm*And i, performing statistical compression coding, wherein the selected coding mode is arithmetic coding:
step 1.5.1, initialize L(0)＝0，H(0)＝2Q-1, scale ═ 0, Q is taken to satisfy Q>log2(Len) a positive integer, Len being the number of source symbols in the encoding time;
step 1.5.2, in the coding process, assuming that the j +1 th signal source symbol is currently input, obtaining H by iterative calculation according to the following formulas (6) and (7)(j+1)And L(j+1)：
Figure BDA0002579362020000033
Figure BDA0002579362020000034
Step 1.5.3 by comparison of H(j+1)And L(j+1)The most significant bit of the binary code element outputs or does not output binary code element '0' or '1', and the judgment and the execution are carried out according to the following sequence conditions:
when H is present(j+1)And L(j+1)Are different from each other and do not satisfy L(j+1)The most significant 2 bits of (A) are "01", H(j+1)When the most significant bit of the data is '10', the code element is not output, the next information source symbol is input, the step 1.5.2 is carried out, otherwise, the following judgment and processing are carried out in sequence according to the sequence of a) and b):
a) if H is present(j+1)And L(j+1)Is the same, outputs the most significant bit, H(j+1)And L(j+1)Left-shifted by 1bit, and the least significant bit is respectively complemented with '1' and '0', and then scale is judged>0 isIf not, outputting the complement of the most significant bit, and subtracting 1 from scale, repeating the step until exiting the step when the scale is 0, and judging b);
b) if L is(j+1)The most significant 2 bits of (A) are "01", H(j+1)Is "10", H(j+1)And L(j+1)Left shifted by 1bit, and its least significant bit is respectively complemented by '1' and '0', H(j+1)And L(j+1)Negating the most significant bit of the sequence, and adding 1 to scale by itself, and performing the step 1.5.3;
repeating the process of the steps 1.5.2-1.5.3 until the encoding input is finished, and at the time of final output, carrying out the final L iterative computation result corresponding to the binary bit data (L)(j+1))bFinal output is also included;
step 1.6, finally compressing the output bit information obtained by coding and selecting the information source symbol dividing mode parameter m*Coding total source symbol number Len, source symbol statistical characteristic identification and statistical characteristic parameter set { beta }m *(1),…,βm *(q) sending out the signals together to a receiving processing module at a receiving side;
step 2, the receiving processing module utilizes the information source symbol grouping information and the information source statistical characteristic parameter which are added at the sending end to take charge of decoding and information source recovery of the received coded digital information:
step 2.1, the information separation submodule separates the bit information output by coding from the received information, and the information source symbol division mode m corresponding to the information source symbol set indication parameter*Total source symbol coding number Len, source symbol statistical property identifier and statistical property parameter set { beta }m *(1),…,βm *(q) } information;
step 2.2, the compression decoding submodule obtains m according to step 2.1*Recovering the ordered source symbol set { X }*(n)|n＝1,…,2NAnd (4) calculating the statistical probability of each source symbol according to the statistical characteristic identifier and the statistical characteristic parameter set obtained in the step 2.1 in the same manner as the formula (4) and further according to the formula (5)Computing a set of ordered source symbols, each X*(n) cumulative probabilities in the sequential symbol set;
and 2.3, the compression decoding submodule decodes the coded bit stream and decodes and outputs an information source symbol data stream, and the decoding process comprises the following steps:
step 2.3.1, initialize L(0)＝0，H(0)＝2Q-1, Q is a positive integer, the decoded output length Len is 0, and the source symbol partitioning mode parameter m*Indication and corresponding coding convention, i.e. m bits are used as a source symbol, let t(0)Binary values corresponding to the previous m bits in the bit stream;
step 2.3.2, when Len < L, repeating the following steps:
step 2.3.2.1, calculating the cumulative probability of the source symbol using the following equation (8):
Figure BDA0002579362020000041
step 2.3.2.2, according to the information source symbol accumulated probability value obtained in step 2.3.2.1, the accumulated probability table in the sequence symbol set obtained in step 2.2 is checked reversely to obtain an information source decoding output;
step 2.3.2.3, based on the source decoded output obtained in step 2.3.2.2, updating L according to equations (6) and (7) above(j)And H(j)：
When L is(j)And H(j)Are different from each other and do not satisfy L(j)The most significant 2 bits of (A) are "01", H(j)If the most significant bit is '10', then step 2.3.2 is entered, otherwise the following judgment and processing are carried out according to the sequence of the step a) and the step b):
a) if L is(j)And H(j)Are the same in the most significant bit, L(j)、H(j)And t(j)All are shifted to the left by one bit and are complemented with '1', '0' and the 1-bit of the input bit stream at the least significant bit respectively;
b) if L is(j)The most significant 2 bits of (A) are "01", H(j)Is the most important ofThe upper two significant bits are "10", L(j)、H(j)And t(j)All shifted to the left by one bit and complemented by a "1", "0" and the 1-bit, L of the input bit stream at the least significant bit, respectively(j)、H(j)And t(j)Negating the most significant bit of the data;
step 2.4, the decoding symbol mapping submodule estimates the information source symbol output by the compression decoding submodule and combines the information source symbol dividing mode parameter m*And restoring the original bit stream information of the transmitting side according to the source symbol coding mapping convention.
The compression transmission system comprises a sending processing module, and the compression transmission module is used for processing the successively collected original bit stream information and successively outputting the coded bit stream information in a sending period;
and the receiving processing module is used for decoding and recovering the information source of the received coded digital information by utilizing the information source symbol grouping information and the information source statistical characteristic parameters which are added at the transmitting end.
Furthermore, the sending processing module comprises a symbol grouping division submodule, a statistical characteristic calculation submodule, an optimal information source symbol set selection submodule and a compression coding submodule;
the symbol grouping and dividing submodule is responsible for carrying out bit stream grouping and dividing on digital information to be transmitted according to M modes and outputting M information source symbol data streams corresponding to different information source symbol sets;
the statistical characteristic calculation submodule is used for receiving M kinds of information source symbol data streams output by the symbol division submodule, calculating the occurrence frequency of various different symbols of the M kinds of information source symbol data streams in a statistical period respectively, and calculating the statistical characteristic parameter of each kind of information source symbol data stream;
the optimal information source symbol set selection submodule receives the output of the statistical characteristic calculation submodule, calculates and analyzes the coding efficiency of M information source symbol data streams corresponding to M statistical characteristic parameters, selects the information source symbol data stream with the optimal coding efficiency, and calculates the statistical characteristic parameters on the information source data stream with the optimal coding efficiency;
and the compression coding sub-module is used for carrying out compression coding on the information source symbol stream based on the information source symbol data stream with the optimal coding efficiency and the corresponding coding statistical characteristic parameters, and finally outputting the coded bit stream information.
Furthermore, the receiving processing module comprises an information separation sub-module, a compression decoding sub-module and a de-symbol mapping sub-module;
an information separation submodule for separating the received information into compressed coded information, information source symbol statistic characteristic identification and statistic characteristic parameter set { beta }m *(1),…,βm *(q) } and total coding symbol number Len, which sends the separated compression coding information to the compression decoding submodule and indicates the parameter m of the separated information source symbol set*Sending to a de-symbol mapping submodule;
the compression decoding submodule decodes the input compression coding bit stream by utilizing the information source symbol statistical characteristic identifier, the statistical characteristic parameter set and the total coding symbol number, outputs a decoding symbol stream and sends the decoding symbol stream to the de-symbol mapping submodule;
and the de-symbol mapping submodule is used for indicating parameters according to the information source symbol set and recovering the input decoding symbol stream into bit stream information according to the convention of the transmitting party and the receiving party.
Compared with the prior art, the compression transmission method and the system thereof have the following remarkable advantages:
1, the invention firstly improves the statistical compression coding technology, and on the data transmitting and coding side, the invention can adaptively select the optimal source symbol mapping set, thereby improving the mapping quality from the source bit to the source symbol. In addition, it introduces the statistical characteristic parameter set obtained based on the calculation of the aforementioned optimal source symbol set, the parameter set is the minimum parameter set characterizing the source symbol distribution, the size of the parameter set is generally much smaller than the set formed by each source symbol statistic, the latter forms the basis of the general statistical coding and decoding, therefore, the former is transmitted to the receiving and decoding party along with the coded bit stream, and the dependency of the coding and decoding on each source symbol statistic in the source symbol set can be reduced, that is, the transmission overhead after compression only contains the source symbol distribution characteristic parameter, but not the distribution of each source symbol, which greatly reduces the transmission overhead between the coding, transmission and the receiving and decoding. The invention can realize the compression of the total data transmission amount between the DU and AAU of the base station equipment, and can realize that the receiving side can obtain the original bit information of the transmitting side without errors.
2, before the coding of the transmitting end, a plurality of information source symbol dividing modes exist, and the optimal information source symbol set can be obtained through statistical calculation from the information source symbol obtaining modes.
And 3, on the premise of ensuring the compression coding efficiency as much as possible, the statistical characteristic parameters of the information source symbols of the transmitting end are used for replacing the statistical characteristics of each information source symbol, so that the requirement of receiving and decoding on the prior information of the coding side is reduced, and the transmission overhead of the system is further reduced.
4, the period of the distribution statistics of the source symbols at the encoding side is separated from the period of the source compression encoding, and the former can be before or at the head time end or the middle time of the latter in terms of time, which can be determined according to specific application scenarios and use conditions, so that the two key processes of the compression encoding and the source symbol statistics can be flexibly deployed in terms of processing time sequence.
5, the transmitting end can adaptively select the optimal source symbol dividing mode according to the source statistical characteristics to obtain the optimal compression codebook, in addition, the distribution characteristic parameters of the source symbols are counted at the transmitting end, the requirement of the receiving side decoder on the source symbol statistical characteristics is reduced, the transmission overhead after compression only comprises the source symbol distribution characteristic parameters instead of the distribution of each source symbol, the larger the codebook set is, the saved transmission overhead is saved, which is very important for saving the transmission cost between DU and AAU.
Drawings
Fig. 1 is a schematic structural diagram of a sending processing module of the compression transmission method and the system thereof according to the present invention;
FIG. 2 is a schematic structural diagram of a receiving processing module of the compression transmission method and the system thereof according to the present invention;
fig. 3 is a schematic diagram of the transmission of the uplink bit stream of the 5GNR according to the compression transmission method and the system thereof of the present invention;
fig. 4 is a schematic diagram of 5GNR uplink compression transmission according to the compression transmission method and the system thereof of the present invention.
Detailed Description
The invention is described in further detail below with reference to the drawings and the detailed description.
The method is suitable for any actual scene in which the information source is compressed and sent to the receiving side. Specifically, taking uplink data transmission between a CU and an AAS of 5GNR as an example, as shown in fig. 3 and 4, the number of antennas configured in the AAS and the system bandwidth are both greatly increased, which results in a greatly increased requirement for the amount of data to be transmitted between the AAS and the CU. In order to greatly reduce the total transmission data volume and ensure that the transmission information does not have variation, double-end deployment is carried out on the receiving and transmitting sides of the compression transmission system, the deployment has little change on the original processing flow of the compression transmission system, and the influence on the time sequence is ignored.
The multi-user signals received by the AAU are considered to be independent in statistics, each user experiences independent multipath Rayleigh (Rayleigh) fading, received time domain signals can be approximately considered to meet Gaussian distribution, and the compression transmission method and the system thereof can be directly utilized.
As shown in fig. 1 and 2, the compression transmission method includes the following steps:
the compression transmission method comprises the following steps:
step 1, a sending processing module is sent in a sending period, original bit stream information collected successively is processed, and coded bit stream information is output successively:
step 1.1, the symbol grouping and dividing submodule and a decoding party agree M information source symbol dividing modes, wherein, the M information source symbol dividing modes are every continuous NmBinary symbol of bit b (i N)m),b(i*Nm+1),…,b((i+1)*Nm-1) } to a source symbol Xm,iThe value corresponding to the source symbol is Xm,iWherein i is …, -1,0,1, …;m＝1,…,M；
step 1.2, in a statistical period time T, the statistical characteristic calculation submodule performs statistical information source symbol set { X }m(n)|n＝0,…,2Nm-1} for each source symbol Xm(n) number of occurrences fm(n) according to fm(n) estimating a source symbol statistical property parameter { beta }m(1),…,βm(q) }, source symbol statistical property parameters calculated for the gaussian distribution of source symbols, mean μ and variance σ of the gaussian distribution, respectively2The mean μ and variance σ of the Gaussian distribution are calculated according to the following equations (1) and (2), respectively2：
Figure BDA0002579362020000071
Figure BDA0002579362020000072
With gaussian distribution, there are only two statistical parameters, i.e. q is 2, βm(1)＝μ,βm(2) When N is ═ σmWhen the data flow is gradually reduced, all the information source symbols gradually tend to be uniformly distributed in the information source symbol data flow, and only one characteristic parameter exists at the moment;
step 1.3, the optimal information source symbol selection submodule selects an information source symbol division mode with optimal coding efficiency in M information source symbol division modes to be selected according to the following formula (3), wherein the information source symbol division mode with the optimal coding efficiency is the mth*The method comprises the following steps:
Figure BDA0002579362020000073
m th*The information source symbol set corresponding to the information source symbol division mode is { Xm*(n)|n＝1,…,2Nm*The statistical property parameter set is { beta }m *(1),…βm *(q) }, the mapped source symbol stream is Xm*,i；
In the step 1.4, the method comprises the following steps of,the optimal source symbol selection submodule selects the optimal source symbol according to the statistical characteristic parameter set [ beta ]m *(1),…βm *(q) }, sequential symbol set { X) in increasing order of n*(n)|n＝1,…,2NCalculating the n information source symbol X ordered in the information source symbol set*Statistical probability of (n)
Figure BDA0002579362020000085
The calculation is performed from a set of statistical property parameters, i.e. the following equation (4):
Figure BDA0002579362020000081
in the formula: x*(n) the cumulative probability in the sequential symbol set is the following equation (5):
Figure BDA0002579362020000082
C0＝0，Cnsource symbol X representing the nth input encoder*(n) cumulative probabilities in the sequential symbol set;
step 1.5, compressing the coding submodule at the coding time omega>Within T, for source symbol stream Xm*And i, performing statistical compression coding, wherein the selected coding mode is arithmetic coding:
step 1.5.1, initialize L(0)＝0，H(0)＝2Q-1, scale ═ 0, Q is taken to satisfy Q>log2(Len) a positive integer, Len being the number of source symbols in the encoding time;
step 1.5.2, in the coding process, assuming that the j +1 th signal source symbol is currently input, obtaining H by iterative calculation according to the following formulas (6) and (7)(j+1)And L(j+1)：
Figure BDA0002579362020000083
Figure BDA0002579362020000084
Step 1.5.3 by comparison of H(j+1)And L(j+1)The most significant bit of the binary code element outputs or does not output binary code element '0' or '1', and the judgment and the execution are carried out according to the following sequence conditions:
when H is present(j+1)And L(j+1)Are different from each other and do not satisfy L(j+1)The most significant 2 bits of (A) are "01", H(j+1)When the most significant bit of the data is '10', the code element is not output, the next information source symbol is input, the step 1.5.2 is carried out, otherwise, the following judgment and processing are carried out in sequence according to the sequence of a) and b):
c) if H is present(j+1)And L(j+1)Is the same, outputs the most significant bit, H(j+1)And L(j+1)Left-shifted by 1bit, and the least significant bit is respectively complemented with '1' and '0', and then scale is judged>If 0 is true, outputting the complement of the most significant bit, and subtracting 1 from scale, repeating the step until exiting the step when scale is 0, and judging b);
d) if L is(j+1)The most significant 2 bits of (A) are "01", H(j+1)Is "10", H(j+1)And L(j+1)Left shifted by 1bit, and its least significant bit is respectively complemented by '1' and '0', H(j+1)And L(j+1)Negating the most significant bit of the sequence, and adding 1 to scale by itself, and performing the step 1.5.3;
repeating the process of the steps 1.5.2-1.5.3 until the encoding input is finished, and at the time of final output, carrying out the final L iterative computation result corresponding to the binary bit data (L)(j+1))bFinal output is also included;
step 1.6, finally compressing the output bit information obtained by coding and selecting the information source symbol dividing mode parameter m*Coding total source symbol number Len, source symbol statistical characteristic identification and statistical characteristic parameter set { beta }m *(1),…,βm *(q) are sent out together and sent to a receiving processing module at a receiving side；
Step 2, the receiving processing module utilizes the information source symbol grouping information and the information source statistical characteristic parameter which are added at the sending end to take charge of decoding and information source recovery of the received coded digital information:
step 2.1, the information separation submodule separates the bit information output by coding from the received information, and the information source symbol division mode m corresponding to the information source symbol set indication parameter*Total source symbol coding number Len, source symbol statistical property identifier and statistical property parameter set { beta }m *(1),…,βm *(q) } information;
step 2.2, the compression decoding submodule obtains m according to step 2.1*Recovering the ordered source symbol set { X }*(n)|n＝1,…,2NAnd (3) calculating the statistical probability of each information source symbol according to the statistical characteristic identifier and the statistical characteristic parameter set obtained in the step 2.1 in the same manner as the calculation manner shown in the formula (4), and further calculating the ordered information source symbol set according to the formula (5), wherein each X is an X*(n) cumulative probabilities in the sequential symbol set;
and 2.3, the compression decoding submodule decodes the coded bit stream and decodes and outputs an information source symbol data stream, and the decoding process comprises the following steps:
step 2.3.1, initialize L(0)＝0，H(0)＝2Q-1, Q is a positive integer, the decoded output length Len is 0, and the source symbol partitioning mode parameter m*Indication and corresponding coding convention, i.e. m bits are used as a source symbol, let t(0)Binary values corresponding to the previous m bits in the bit stream;
step 2.3.2, when Len < L, repeating the following steps:
step 2.3.2.1, calculating the cumulative probability of the source symbol using the following equation (8):
Figure BDA0002579362020000091
step 2.3.2.2, according to the information source symbol accumulated probability value obtained in step 2.3.2.1, the accumulated probability table in the sequence symbol set obtained in step 2.2 is checked reversely to obtain an information source decoding output;
step 2.3.2.3, based on the source decoded output obtained in step 2.3.2.2, updating L according to equations (6) and (7) above(j)And H(j)：
When L is(j)And H(j)Are different from each other and do not satisfy L(j)The most significant 2 bits of (A) are "01", H(j)If the most significant bit is '10', then step 2.3.2 is entered, otherwise the following judgment and processing are carried out according to the sequence of the step a) and the step b):
c) if L is(j)And H(j)Are the same in the most significant bit, L(j)、H(j)And t(j)All are shifted to the left by one bit and are complemented with '1', '0' and the 1-bit of the input bit stream at the least significant bit respectively;
d) if L is(j)The most significant 2 bits of (A) are "01", H(j)Is "10", L(j)、H(j)And t(j)All shifted to the left by one bit and complemented by a "1", "0" and the 1-bit, L of the input bit stream at the least significant bit, respectively(j)、H(j)And t(j)Negating the most significant bit of the data;
step 2.4, the decoding symbol mapping submodule estimates the information source symbol output by the compression decoding submodule and combines the information source symbol dividing mode parameter m*And restoring the original bit stream information of the transmitting side according to the source symbol coding mapping convention.
The compression transmission system comprises a sending processing module, and the compression transmission module is used for processing the successively collected original bit stream information and successively outputting the coded bit stream information in a sending period;
and the receiving processing module is used for decoding and recovering the information source of the received coded digital information by utilizing the information source symbol grouping information and the information source statistical characteristic parameters which are added at the transmitting end.
Furthermore, the sending processing module comprises a symbol grouping division submodule, a statistical characteristic calculation submodule, an optimal information source symbol set selection submodule and a compression coding submodule;
the symbol grouping and dividing submodule is responsible for carrying out bit stream grouping and dividing on digital information to be transmitted according to M modes and outputting M information source symbol data streams corresponding to different information source symbol sets;
the statistical characteristic calculation submodule is used for receiving M kinds of information source symbol data streams output by the symbol division submodule, calculating the occurrence frequency of various different symbols of the M kinds of information source symbol data streams in a statistical period respectively, and calculating the statistical characteristic parameter of each kind of information source symbol data stream;
the optimal information source symbol set selection submodule receives the output of the statistical characteristic calculation submodule, calculates and analyzes the coding efficiency of M information source symbol data streams corresponding to M statistical characteristic parameters, selects the information source symbol data stream with the optimal coding efficiency, and calculates the statistical characteristic parameters on the information source data stream with the optimal coding efficiency;
and the compression coding sub-module is used for carrying out compression coding on the information source symbol stream based on the information source symbol data stream with the optimal coding efficiency and the corresponding coding statistical characteristic parameters, and finally outputting the coded bit stream information.
Furthermore, the receiving processing module comprises an information separation sub-module, a compression decoding sub-module and a de-symbol mapping sub-module;
the information separation submodule is used for separating the received information into compressed coding information, information source symbol statistical characteristic marks, statistical characteristic parameter sets { beta m (1), … beta m (q) } and the total coding symbol number Len according to a convention mode, sending the separated compressed coding information to the compression decoding submodule and sending the separated information source symbol set indication parameter m to the de-symbol mapping submodule;
the compression decoding submodule decodes the input compression coding bit stream by utilizing the information source symbol statistical characteristic identifier, the statistical characteristic parameter set and the total coding symbol number, outputs a decoding symbol stream and sends the decoding symbol stream to the de-symbol mapping submodule;
and the de-symbol mapping submodule is used for indicating parameters according to the information source symbol set and recovering the input decoding symbol stream into bit stream information according to the convention of the transmitting party and the receiving party.
It should be noted that the whole deployment and processing are very direct, and through the above deployment and processing according to steps, the lossless compression transmission of the data volume transmitted from the original AAS to the CU can be realized, which not only ensures that the quality of the transmitted bit stream is not affected, but also relieves the traffic pressure of the data transmission between the 5GNR CU and the AAS.
Further, assuming that under the condition of 5GNR, there are 4 cells (cells), each cell is configured with an uplink 4-antenna (ant) for reception, and for the bandwidth condition of 100MHz, according to I/Q (In-phase and Quadrature quadratures) two paths of 16-bit FFT, 4096-point FFT, and subcarrier spacing of 30KHz, the peak value raw data transmission flow between AAU and DU is estimated:
4(Cell)*4(Ant)*(4096*30KHz)*(16bit+16bit)≈62.9Gbps。
if the technical scheme of the invention is adopted, at this time, it is assumed that the optimal information source symbol set has 2^16 information source symbols, the distribution of the optimal information source symbols meets Gaussian distribution, the mean value and the variance of the optimal information source symbols are both quantized to 16 bits, the coding length information occupies 16 bits (L <2^16), the optimal information source symbol set indicates 4 bits to estimate statistical characteristics within the time T equal to 0.5ms, after compression within omega equal to 5ms, the average compression ratio is set to 0.6, and the transmission flow of peak compression data between AAU and DU is estimated:
4(Cell)*4(Ant)*(4096*30KHz)*(16bit+16bit)*0.6≈37.74Gbps。
if all the source symbol distributions are directly transmitted, the transmission overhead is estimated:
4(Cell)*4(Ant)*(2^16*16bit)/5ms≈3.36Gbps。
the size of the signal source symbol set is directly related to the size of the signal source symbol set, the size of the signal source symbol set increases exponentially along with the number N of the signal source bit stream partitions, and the overhead still increases as the compression coding time omega decreases.
The compression transmission method and the system thereof of the invention have the following steps:
4(Cell)*4(Ant)*(16bit+16bit+16bit+4bit)/5ms＝166Kbps。
it has no relation with the size of the source symbol set, only directly relates to the characteristic parameter of the optimal source symbol set, and is more suitable for the transmission setting with shorter compression coding time and more flexibility due to the relatively small overhead.
The above description is only for the preferred embodiment of the present invention and should not be construed as limiting the present invention, and various modifications and changes can be made by those skilled in the art without departing from the spirit and principle of the present invention, and any modifications, equivalents, improvements, etc. should be included in the scope of the claims of the present invention.

Claims (8)

1. A compression transmission method of a 5GNR baseband digital signal is characterized by comprising the following steps:
step 1, a processing module is sent in a sending period, and the encoded bit stream information is output successively after the processing of the successively collected original bit stream information;
and 2, the receiving processing module is responsible for decoding and recovering the information source of the received coded digital information by utilizing the information source symbol grouping information and the information source statistical characteristic parameters which are added at the sending end.
2. The method for compressed transmission of 5GNR baseband digital signals according to claim 1, further comprising the following steps in step 1:
step 1.1, the symbol grouping and dividing submodule and a decoding party agree M information source symbol dividing modes, wherein, the M information source symbol dividing modes are every continuous NmBinary symbol of bit b (i N)m),b(i*Nm+1),…,b((i+1)*Nm-1) } to a source symbol Xm,iThe value corresponding to the source symbol is Xm,iWherein i is …, -1,0,1, …; m is 1, …, M;
step 1.2, in a statistical period time T, the statistical characteristic calculation submodule performs statistical information source symbol set { X }m(n)|n＝0,…,2Nm-1} for each source symbol Xm(n) number of occurrences fm(n) according to fm(n) estimating a source symbol statistical property parameter { beta }m(1),…,βm(q)}The statistical characteristic parameters of the source symbols calculated for the Gaussian distribution of the source symbols are the mean mu and the variance sigma of the Gaussian distribution2The mean μ and variance σ of the Gaussian distribution are calculated according to the following equations (1) and (2), respectively2：
Figure FDA0002579362010000011
Figure FDA0002579362010000012
With gaussian distribution, there are only two statistical parameters, i.e. q is 2, βm(1)＝μ,βm(2) When N is ═ σmWhen the data flow is gradually reduced, all the information source symbols gradually tend to be uniformly distributed in the information source symbol data flow, and only one characteristic parameter exists at the moment;
step 1.3, the optimal information source symbol selection submodule selects an information source symbol division mode with optimal coding efficiency in M information source symbol division modes to be selected according to the following formula (3), wherein the information source symbol division mode with the optimal coding efficiency is the mth*The method comprises the following steps:
Figure FDA0002579362010000013
m th*The source symbol set corresponding to the source symbol dividing mode is
Figure FDA0002579362010000015
The statistical property parameter set is { betam *(1),…βm *(q) }, the source symbol stream mapped as
Figure FDA0002579362010000016
Step 1.4, the optimal information source symbol selection submodule selects the sub-module according to the statistical characteristic parameter set { betam *(1),…βm *(q) }, sequential symbol set { X) in increasing order of n*(n)|n＝1,…,2NCalculating the n information source symbol X ordered in the information source symbol set*Statistical probability of (n)
Figure FDA0002579362010000014
The calculation is performed from a set of statistical property parameters, i.e. the following equation (4):
Figure FDA0002579362010000021
in the formula: x*(n) the cumulative probability in the sequential symbol set is the following equation (5):
Figure FDA0002579362010000022
C0＝0，Cnsource symbol X representing the nth input encoder*(n) cumulative probabilities in the sequential symbol set;
step 1.5, compressing the coding submodule at the coding time omega>Within T, for source symbol stream
Figure FDA0002579362010000025
Carrying out statistical compression coding, wherein the selected coding mode is arithmetic coding;
step 1.6, finally compressing the output bit information obtained by coding and selecting the information source symbol dividing mode parameter m*Coding total source symbol number Len, source symbol statistical characteristic identification and statistical characteristic parameter set { beta }m *(1),…,βm *(q) are sent out together and sent to a receiving processing module at the receiving side.
3. The method for compressed transmission of 5GNR baseband digital signals according to claim 2, characterized in that in step 1.5, it further comprises the steps of:
step 1.5.1, initialChanging L(0)＝0，H(0)＝2Q-1, scale ═ 0, Q is taken to satisfy Q>log2(Len) a positive integer, Len being the number of source symbols in the encoding time;
step 1.5.2, in the coding process, assuming that the j +1 th signal source symbol is currently input, obtaining H by iterative calculation according to the following formulas (6) and (7)(j+1)And L(j+1)：
Figure FDA0002579362010000023
Figure FDA0002579362010000024
Step 1.5.3 by comparison of H(j+1)And L(j+1)The most significant bit of the binary code element outputs or does not output binary code element '0' or '1', and the judgment and the execution are carried out according to the following sequence conditions:
when H is present(j+1)And L(j+1)Are different from each other and do not satisfy L(j+1)The most significant 2 bits of (A) are "01", H(j+1)When the most significant bit of the data is '10', the code element is not output, the next information source symbol is input, the step 1.5.2 is carried out, otherwise, the following judgment and processing are carried out in sequence according to the sequence of a) and b):
a) if H is present(j+1)And L(j+1)Is the same, outputs the most significant bit, H(j+1)And L(j+1)Left-shifted by 1bit, and the least significant bit is respectively complemented with '1' and '0', and then scale is judged>If 0 is true, outputting the complement of the most significant bit, and subtracting 1 from scale, repeating the step until exiting the step when scale is 0, and judging b);
b) if L is(j+1)The most significant 2 bits of (A) are "01", H(j+1)Is "10", H(j+1)And L(j +1)Left shifted by 1bit, and its least significant bit is respectively complemented by '1' and '0', H(j+1)And L(j+1)Negating the most significant bit of the sequence, and adding 1 to scale by itself, and performing the step 1.5.3;
repeating the process of the steps 1.5.2-1.5.3 until the encoding input is finished, and at the time of final output, carrying out the final L iterative computation result corresponding to the binary bit data (L)(j+1))bWith the final output.
4. The method for compressed transmission of 5GNR baseband digital signals according to claim 1, further comprising the following steps in step 2:
step 2.1, the information separation submodule separates the bit information output by coding from the received information, and the information source symbol division mode m corresponding to the information source symbol set indication parameter*Total source symbol coding number Len, source symbol statistical property identifier and statistical property parameter set { beta }m *(1),…,βm *(q) } information;
step 2.2, the compression decoding submodule obtains m according to step 2.1*Recovering the ordered source symbol set { X }*(n)|n＝1,…,2NAnd (3) calculating the statistical probability of each information source symbol according to the statistical characteristic identifier and the statistical characteristic parameter set obtained in the step 2.1 in the same manner as the calculation manner shown in the formula (4), and further calculating the ordered information source symbol set according to the formula (5), wherein each X is an X*(n) cumulative probabilities in the sequential symbol set;
step 2.3, the compression decoding submodule decodes the coded bit stream and decodes and outputs an information source symbol data stream;
step 2.4, the decoding symbol mapping submodule estimates the information source symbol output by the compression decoding submodule and combines the information source symbol dividing mode parameter m*And restoring the original bit stream information of the transmitting side according to the source symbol coding mapping convention.
5. A method for compressed transmission of 5GNR baseband digital signals according to claim 4, further comprising, in step 2.3, the steps of:
step 2.3.1, initialize L(0)＝0，H(0)＝2Q-1, Q is a positive integer, the decoded output length Len is 0, and the source symbol partitioning mode parameter m*Indication and corresponding coding convention, i.e. m bits are used as a source symbol, let t(0)Binary values corresponding to the previous m bits in the bit stream;
step 2.3.2, when Len < L, repeating the following steps:
step 2.3.2.1, calculating the cumulative probability of the source symbol using the following equation (8):
Figure FDA0002579362010000031
step 2.3.2.2, according to the information source symbol accumulated probability value obtained in step 2.3.2.1, the accumulated probability table in the sequence symbol set obtained in step 2.2 is checked reversely to obtain an information source decoding output;
step 2.3.2.3, based on the source decoded output obtained in step 2.3.2.2, updating L according to equations (6) and (7) above(j)And H(j)：
When L is(j)And H(j)Are different from each other and do not satisfy L(j)The most significant 2 bits of (A) are "01", H(j)If the most significant bit is '10', then step 2.3.2 is entered, otherwise the following judgment and processing are carried out according to the sequence of the step a) and the step b):
a) if L is(j)And H(j)Are the same in the most significant bit, L(j)、H(j)And t(j)All are shifted to the left by one bit and are complemented with '1', '0' and the 1-bit of the input bit stream at the least significant bit respectively;
b) if L is(j)The most significant 2 bits of (A) are "01", H(j)Is "10", L(j)、H(j)And t(j)All shifted to the left by one bit and complemented by a "1", "0" and the 1-bit, L of the input bit stream at the least significant bit, respectively(j)、H(j)And t(j)The most significant bit of (a) is inverted.
6. A system for compressed transmission of a 5GNR baseband digital signal, comprising:
the sending processing module is used for processing the successively collected original bit stream information and successively outputting the coded bit stream information in a sending period;
and the receiving processing module is used for decoding and recovering the information source of the received coded digital information by utilizing the information source symbol grouping information and the information source statistical characteristic parameters which are added at the transmitting end.
7. The system for compressed transmission of 5GNR baseband digital signals according to claim 6, wherein the transmission processing module comprises a symbol grouping division sub-module, a statistical characteristic calculation sub-module, an optimal source symbol set selection sub-module and a compression coding sub-module;
the symbol grouping and dividing submodule is responsible for carrying out bit stream grouping and dividing on digital information to be transmitted according to M modes and outputting M information source symbol data streams corresponding to different information source symbol sets;
the statistical characteristic calculation submodule is used for receiving M kinds of information source symbol data streams output by the symbol division submodule, calculating the occurrence frequency of various different symbols of the M kinds of information source symbol data streams in a statistical period respectively, and calculating the statistical characteristic parameter of each kind of information source symbol data stream;
the optimal information source symbol set selection submodule receives the output of the statistical characteristic calculation submodule, calculates and analyzes the coding efficiency of M information source symbol data streams corresponding to M statistical characteristic parameters, selects the information source symbol data stream with the optimal coding efficiency, and calculates the statistical characteristic parameters on the information source data stream with the optimal coding efficiency;
and the compression coding sub-module is used for carrying out compression coding on the information source symbol stream based on the information source symbol data stream with the optimal coding efficiency and the corresponding coding statistical characteristic parameters, and finally outputting the coded bit stream information.
8. The system for compressed transmission of 5GNR baseband digital signals according to claim 6, wherein the receiving processing module comprises an information separation sub-module, a compression decoding sub-module and a de-symbol mapping sub-module;
an information separation submodule for separating the received information into compressed coded information, information source symbol statistic characteristic identification and statistic characteristic parameter set { beta }m *(1),…,βm *(q) } and total coding symbol number Len, which sends the separated compression coding information to the compression decoding submodule and indicates the parameter m of the separated information source symbol set*Sending to a de-symbol mapping submodule;
the compression decoding submodule decodes the input compression coding bit stream by utilizing the information source symbol statistical characteristic identifier, the statistical characteristic parameter set and the total coding symbol number, outputs a decoding symbol stream and sends the decoding symbol stream to the de-symbol mapping submodule;
and the de-symbol mapping submodule is used for indicating parameters according to the information source symbol set and recovering the input decoding symbol stream into bit stream information according to the convention of the transmitting party and the receiving party.
CN202010663149.3A 2020-07-10 2020-07-10 Compression transmission method and system for 5GNR baseband digital signal Active CN111954263B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010663149.3A CN111954263B (en) 2020-07-10 2020-07-10 Compression transmission method and system for 5GNR baseband digital signal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010663149.3A CN111954263B (en) 2020-07-10 2020-07-10 Compression transmission method and system for 5GNR baseband digital signal

Publications (2)

Publication Number Publication Date
CN111954263A true CN111954263A (en) 2020-11-17
CN111954263B CN111954263B (en) 2023-09-19

Family

ID=73341483

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010663149.3A Active CN111954263B (en) 2020-07-10 2020-07-10 Compression transmission method and system for 5GNR baseband digital signal

Country Status (1)

Country Link
CN (1) CN111954263B (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110052087A1 (en) * 2009-08-27 2011-03-03 Debargha Mukherjee Method and system for coding images
CN103107858A (en) * 2011-11-15 2013-05-15 中国科学院研究生院 Coding and decoding method of length-variable symbol level joint information source channel and multi-dimensional modulation
CN104539393A (en) * 2015-01-07 2015-04-22 北京邮电大学 Source coding method based on polar codes
CN106603081A (en) * 2016-11-02 2017-04-26 深圳市大语科技有限公司 General arithmetic coding method and decoding method thereof
US20170295263A1 (en) * 2013-01-31 2017-10-12 Excalibur Ip, Llc System and method for applying an efficient data compression scheme to url parameters
CN110808739A (en) * 2019-10-23 2020-02-18 中国人民解放军战略支援部队航天工程大学 Binary coding method and device with unknown source symbol probability distribution
CN111050291A (en) * 2019-12-27 2020-04-21 北京理工大学 High-coding-sign-degree value broadcasting method based on system online fountain codes
US20200218713A1 (en) * 2019-01-08 2020-07-09 Live Earth, LLC Data structure and format for efficient storage or transmission of objects

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110052087A1 (en) * 2009-08-27 2011-03-03 Debargha Mukherjee Method and system for coding images
CN103107858A (en) * 2011-11-15 2013-05-15 中国科学院研究生院 Coding and decoding method of length-variable symbol level joint information source channel and multi-dimensional modulation
US20170295263A1 (en) * 2013-01-31 2017-10-12 Excalibur Ip, Llc System and method for applying an efficient data compression scheme to url parameters
CN104539393A (en) * 2015-01-07 2015-04-22 北京邮电大学 Source coding method based on polar codes
CN106603081A (en) * 2016-11-02 2017-04-26 深圳市大语科技有限公司 General arithmetic coding method and decoding method thereof
US20200218713A1 (en) * 2019-01-08 2020-07-09 Live Earth, LLC Data structure and format for efficient storage or transmission of objects
CN110808739A (en) * 2019-10-23 2020-02-18 中国人民解放军战略支援部队航天工程大学 Binary coding method and device with unknown source symbol probability distribution
CN111050291A (en) * 2019-12-27 2020-04-21 北京理工大学 High-coding-sign-degree value broadcasting method based on system online fountain codes

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
MAROUA TAGHOUTI: "Reduction of padding overhead for RLNC Media distribution with variable size packets", IEEE TRANSACTIONS ON BROADCASTING(VOLUME:65, ISSUE:3, SEPTEMBER 2019) *
刘东: "基于SLWE的自适应熵编码器概率估计模型应用研究", 中国优秀硕士学位论文全文数据库 *
梁飞: "带宽紧缺条件下混合数模视频传输方法的研究", 中国优秀博士学位论文全文数据库 *

Also Published As

Publication number Publication date
CN111954263B (en) 2023-09-19

Similar Documents

Publication Publication Date Title
Oien et al. Impact of channel prediction on adaptive coded modulation performance in Rayleigh fading
EP2807747B1 (en) A method and apparatus for soft information transfer between constituent processor circuits in a soft-value processing apparatus
CN101965704B (en) Feedback with unequal error protection
US9722677B2 (en) Backhaul signal compression through spatial-temporal linear prediction
EP2689539B1 (en) Signal compression for backhaul communications using linear transformations
EP2888849B1 (en) Soft metrics compressing method
CN1777041A (en) Receiver system and method for soft-decision decoding of retracted convolutional codes
US8989257B1 (en) Method and apparatus for providing near-zero jitter real-time compression in a communication system
US9553954B1 (en) Method and apparatus utilizing packet segment compression parameters for compression in a communication system
WO2001050613A1 (en) Multi-rate, channel-optimized trellis-coded quantization
CN103905149A (en) Methods for designing codebook, communication device and codebook training method
CN107437976B (en) Data processing method and equipment
US20110096852A1 (en) Radio reception device, radio transmission device, and feedback method
CN101931498A (en) Method and device for transmitting channel state information
CN102307372B (en) Lloyd-Max-quantizer-based data compression method and equipment
CN111954263B (en) Compression transmission method and system for 5GNR baseband digital signal
US20030138058A1 (en) Diversity coded OFDM for high data-rate communication
WO2009125591A1 (en) Wireless receiver, wireless transmitter, and feedback method
CN113556159A (en) Channel feedback method of large-scale MIMO multi-user system
CN104113394B (en) The compression of communication modulation signal and decompressing method
CN114157722A (en) Data transmission method and device
Tembhurne et al. Alamouti based on spatial modulation for data transmission
CN106533519B (en) multi-hop cooperative transmission method based on dynamic antenna selection
WO2018228704A1 (en) A network device, a user equipment and a method for wireless data transmission
EP4047829B1 (en) System and method for uplink coordinated transmission in an open radio access network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant