CN113347131A - Improved SLM-PTS peak-to-average power ratio restraining method - Google Patents

Abstract

The invention discloses an improved SLM-PTS peak-to-average power ratio restraining method, which comprises the following steps: generating the lSLM-algorithm: the input signal sequence is duplicated into M groups in the frequency domain and multiplied by M phase twiddle factors respectively, the M groups of time domain signals are averagely divided into two groups, and the two groups of signals are combined linearly to generate M²Group 4 alternative signals; generating a WHT-IPTS algorithm: firstly, carrying out digital mapping on an input signal to generate N pieces of mapped data, inputting the mapped data into a precoder and multiplying the precoded data by a WHT matrix to obtain encoded frequency domain data, dividing the frequency domain data into V subblocks, and finding out a phase factor combination with the lowest PAPR value by using a suboptimal iterative search method; and obtaining an ISLM-mulaw-WHT-IPTS algorithm, and performing PAPR value suppression on signals in the OFDM system by adopting the ISLM-mulaw-WHT-IPTS algorithm.

Description

Improved SLM-PTS peak-to-average power ratio restraining method

Technical Field

The invention relates to the field of data signal processing, in particular to an improved SLM-PTS peak-to-average power ratio restraining method.

Background

The basic idea of OFDM is to convert a high rate serial data stream into multiple low rate parallel sub-streams by a serial-to-parallel (S/P) transform operation, which are then modulated onto multiple orthogonal sub-carriers for parallel transmission. The parallel transmission system greatly expands the pulse width of the symbol and improves the performance of resisting the severe transmission conditions such as multipath fading and the like. The OFDM system signal is formed by superimposing a plurality of subcarrier signals, so that a large Peak to Average Power Ratio (PAPR) is easily generated at the output of the OFDM system, which is also the most disturbing problem in the OFDM system. The peak-to-average ratio in the OFDM system is high, so that when a signal passes through the power amplifier, the peak value of the waveform of the OFDM signal is not in the linear dynamic range of the power amplifier, signal distortion is caused, the efficiency of the power amplifier is reduced, and the orthogonality among subcarriers is damaged, thereby affecting the overall performance of the system. The ratio of the maximum peak power to the average power within one OFDM symbol period is commonly referred to as the peak-to-average power ratio. For a discrete-time signal x (n), its PAPR formula can be expressed as

There are three conventional PAPR suppression methods: coding methods, pre-distortion methods and probability methods. The main idea of the coding method is to use a code word with low PAPR value to replace a code word with high PAPR value in an original signal by adopting a coding mode, and different coding modes can cause different PAPRs, then a group with the lowest PAPR value is selected from a large number of code groups to be used as an input signal, the signal is required to be decoded at a receiving end, and the selection of the coding mode is to perform suppression processing on the existing higher PAPR. The existing coding algorithms mainly include block coding, reed miller coding and gray complementary coding.

The predistortion method is the simplest and most direct method for realizing PAPR suppression in three types of optimization algorithms, and the basic idea of the predistortion method is to compress or directly eliminate a signal with a high instantaneous PAPR value before the signal is sent to a power amplification element, so that the signal does not exceed the dynamic transformation range of the power amplification element, and the condition that the PAPR value is too high is directly prevented. However, the predistortion operation directly adopted by the predistortion method generates certain out-of-band noise and in-band interference, thereby affecting the communication performance of the system. The most common existing predistortion techniques are the direct clipping technique and the companding technique.

The probability algorithm is to reduce the probability of occurrence of high PAPR as much as possible, and cannot optimize the PAPR of the input signal to be less than the set desired threshold. The method selects the symbol sequence with the minimum PAPR value for transmission by taking some auxiliary information into account, and does not generate any change to the signal. The probability algorithm has high computational complexity, and sideband information needs to be transmitted in the implementation process, so that the system bandwidth is increased, and the system transmission rate is reduced. The probability algorithm is considered to be the most potential method for solving the higher APPR value at present at home and abroad. The existing probability algorithm mainly includes a selective Mapping (SLM), a Partial Transmit Sequence (PTS), and an Active Constellation Extension (ACE).

The main idea of the SLM algorithm is to generate a plurality of candidate signals, and then to screen out a group with the lowest PAPR value from these sequences as a transmission signal, so as to achieve the purpose of reducing PAPR. The main idea of the PTS algorithm is to divide frequency domain data into V sub-blocks, perform weighted superposition on the sub-sequences and the phase factors obtained by each sub-block group after IFFT operation, and finally find out a phase factor combination with the lowest PAPR by traversing all the phase twiddle factors, thereby achieving the purpose of reducing PAPR.

However, if a better PAPR suppression effect is to be obtained, the number of candidate signals needs to be increased for the SLM algorithm. For the PTS algorithm, the number of subblocks for frequency domain data division needs to be increased. However, the larger the number of candidate signals and the number of sub-blocks, the greater the computational complexity.

Disclosure of Invention

According to the problems existing in the prior art, the invention discloses an improved SLM-PTS peak-to-average power ratio inhibition method, which specifically comprises the following steps:

generating an ISLM- μ law algorithm: the method comprises the steps of (1) duplicating an input signal sequence into M groups in a frequency domain, multiplying the M groups by M phase twiddle factors respectively to generate M groups of alternative signals, and carrying out IFFT (inverse fast Fourier transform) on the M groups of alternative signals respectively to generate M groups of time domain signals; averagely dividing M groups of time domain signals into two groups, and randomly and linearly combining the two groups of signals to generate M²The/4 groups of alternative signals are selected, and a group of signals with the minimum PAPR value are selected from the alternative signals to be subjected to companding conversion so as to generate signals to be transmitted;

generating a WHT-IPTS algorithm: firstly, carrying out digital mapping on an input signal to generate N pieces of mapped data, inputting the mapped data into a precoder and multiplying the precoded data by a WHT matrix to obtain encoded frequency domain data, dividing the frequency domain data into V sub-blocks, respectively carrying out IFFT operation on the V sub-blocks, and finding out a phase factor combination with the lowest PAPR value by using a suboptimal iterative search method;

and combining the ISLM-mulaw algorithm and the WHT-IPTS algorithm in series to obtain the ISLM-mulaw-WHT-IPTS algorithm, and performing PAPR value suppression on signals in the OFDM system by adopting the ISLM-mulaw-WHT-IPTS algorithm.

Further, when the lSLM- μ law algorithm is generated:

firstly, constellation mapping and serial-parallel conversion are carried out on original binary bit data, and an input signal sequence is copied into M groups in a frequency domain;

multiplying the M groups of signals by M phase twiddle factors respectively to generate M groups of alternative signals containing the same information;

performing IFFT on the M groups of alternative signals respectively to generate M groups of time domain signals, and averagely dividing the M groups of time domain signals into two groups, namely

X is to be^(m)And x⁽ⁿ⁾Performing linear combination to generate M²Selecting a group of signals x with the minimum PAPR value from the 4 groups of alternative signals;

carrying out companding transformation on the signal x to generate a signal x';

taking the signal x' as an input signal of the next stage after FFT operation, carrying out digital mapping on the input signal to generate N pieces of mapped data, and multiplying the N pieces of mapped data by an NxN WHT matrix through a precoder to obtain encoded data;

performing data segmentation in an interleaving segmentation mode, and segmenting frequency domain data into V subblocks;

performing IFFT transformation on the V sub-block data respectively;

using a suboptimal iterative search method, only considering b_vAnd taking the two values of-1, 1 to find out the phase factor combination with the lowest PAPR value.

Further, when a suboptimal iterative search method is adopted to find out the phase factor combination with the lowest PAPR value:

s21, dividing the frequency domain data with N number of sub-carriers into V sub-data blocks which are not overlapped with each other;

s22 is made as follows_vThe PAPR under the condition is calculated as 1(V is 1,2, …, V) and 1 (index is 1)₀Max | X '|/E (| X' |), wherein

S23 is made as follows_indexCalculating the peak-to-average power ratio (PAPR) by adopting the method in S22 as-1;

s24 PAPR₀Comparing with PAPR if PAPR is not the same₀< PAPR, then b _index1 is ═ 1; otherwise, if the PAPR is not₀＝PAPR，index＝index+1；

S25, when the index is less than or equal to V, jumping to S23, otherwise, repeating S24 until the index is V to complete the search, and finally obtaining the suboptimal twiddle factor, wherein the optimal PAPR value is min { PAPR, PAPR₀}；

And after one iteration search is finished, performing second optimization by taking the result of the first iteration search as an initial value, and repeating the steps until the required iteration times are finished.

By adopting the technical scheme, the improved SLM-PTS peak-to-average power ratio restraining method provided by the invention has the advantages of improving the traditional SLM algorithm and the traditional PTS algorithm, and combines the improved two algorithms in series, namely, the output signal of the lSLM- μ law algorithm is used as the input signal of the WHT-IPTS algorithm, so that the PAPR restraining capability is improved. The problem of overhigh signal PAPR value in the OFDM system is effectively solved, so that signal distortion is prevented, and the overall performance of the system is improved.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present application, and other drawings can be obtained by those skilled in the art without creative efforts.

Fig. 1 is a schematic block diagram of the ISLM- μ law algorithm.

FIG. 2 is a functional block diagram of the WHT-IPTS algorithm.

FIG. 3 is a schematic block diagram of an ISLM- μ law-WHT-IPTS algorithm.

FIG. 4 is a graph comparing PAPR inhibition of the ISLM-mulaw-WHT-IPTS algorithm and the algorithms before improvement.

Detailed Description

In order to make the technical solutions and advantages of the present invention clearer, the following describes the technical solutions in the embodiments of the present invention clearly and completely with reference to the drawings in the embodiments of the present invention:

an improved SLM-PTS peak-to-average ratio suppressing method as shown in fig. 1 specifically includes the following steps:

and S1, improving the traditional SLM algorithm to obtain the lSLM-mulaw algorithm.

FIG. 1 is a schematic diagram of an lSLM- μ law algorithm, in which an input signal sequence is first duplicated into M sets in the frequency domain, and the M sets are multiplied by M phase twiddle factors, respectively, to generate M sets of candidate signals. Then, M groups of candidate signals are respectively subjected to IFFT transformation to generate M groups of time domain signals, then, the M groups of time domain signals are equally divided into two groups, linear combination shown in formula (1) is arbitrarily performed between the two groups of signals, and finally, linear combination is generated

Group candidate signals, and select the PAPR value minimum therefromAnd finally, carrying out mu-law companding on the selected group of signals, and sending the generated signals to a channel for transmission.

x^(m,n)＝sinβ·x^(m)+cosβ·x⁽ⁿ⁾＝sinβ·IFFT(x·b^(m))+cosβ·IFFT(x·b⁽ⁿ⁾) (1)

S2: and improving the traditional PTS algorithm to obtain a WHT-IPTS algorithm.

Fig. 2 is a schematic block diagram of a WHT-IPTS algorithm, in which an input signal is digitally mapped to generate N mapped data, and the N mapped data are multiplied by a WHT matrix via a precoder to obtain encoded data. And then data segmentation is carried out, frequency domain data are segmented into V sub-blocks, the V sub-blocks are respectively subjected to IFFT operation, and a suboptimal iterative search method is applied to find out a phase factor combination with the lowest PAPR value. The next iteration search method comprises the following specific steps:

(1) dividing the frequency domain data with N number of subcarriers into V sub-data blocks which are not overlapped with each other

(2) Let b_vThe PAPR under the condition is calculated as 1(V is 1,2, …, V) and 1 (index is 1)₀Max | X '|/E (| X' |), wherein

(3) Let b_indexCalculating the peak-to-average power ratio (PAPR) in the same step (2) as the step-1;

(4) PAPR (Peak to average Power ratio)₀Comparing with PAPR if PAPR is not the same₀< PAPR, then b _index1 is ═ 1; otherwise, if the PAPR is not₀＝PAPR，index＝index+1；

(5) And (4) when index is less than or equal to V, jumping to the step (3), otherwise, repeating the step (4) until the index is equal to V and the search is completed. Finally obtaining suboptimal twiddle factor, wherein the optimal PAPR value is min { PAPR, PAPR₀}。

After one iteration search is finished, the result of the first iteration search is taken as an initial value to carry out second optimization, and the like, so that the required iteration times are finished, and a good effect can be achieved after 3 times of iteration times are generally finished. The suboptimal iterative algorithm only needs to carry out IFFT operation V.P times, and compared with the traditional PTS algorithm, the IPTS algorithm has greatly reduced calculation complexity.

S3: and combining the lSLM-mulaw algorithm and the WHT-IPTS algorithm in series to obtain the ISLM-mulaw-WHT-IPTS algorithm.

FIG. 3 is a schematic block diagram of an lSLM- μ law-WHT-IPTS algorithm. The specific implementation steps of the lSLM-mulaw-WHT-IPTS algorithm are as follows:

(1) firstly, constellation mapping and serial-parallel conversion are carried out on original binary bit data, and an input signal sequence is copied into M groups in a frequency domain;

(2) multiplying the M groups of signals by M phase twiddle factors respectively to generate M groups of alternative signals containing the same information;

(3) performing IFFT on the M groups of alternative signals respectively to generate M groups of time domain signals, and averagely dividing the M groups of time domain signals into two groups, namely

(4) X is to be^(m)And x⁽ⁿ⁾Linear combination of formula (1) to ultimately produce M ²4, selecting a group of signals x with the minimum PAPR value from the group of alternative signals;

(5) carrying out mu-law companding transformation on the signal x to generate a signal x';

(6) x' is used as the input signal of the next stage after FFT operation, the input signal is firstly subjected to digital mapping to generate N pieces of mapped data, and then the N pieces of mapped data are multiplied by a precoder and an NxN WHT matrix to obtain encoded data;

(7) performing data segmentation in an interleaving segmentation mode, and segmenting frequency domain data into V subblocks;

(8) performing IFFT transformation on the V sub-block data respectively;

(9) using a suboptimal iterative search method, only considering b_vAnd taking the two values of-1, 1 to find out the phase factor combination with the lowest PAPR value.

FIG. 4 is a graph comparing PAPR inhibition capability of the lSLM-mulaw-WHT-IPTS algorithm and that of the algorithms before improvement. The PAPR values of the WHT-IPTS algorithm, the lSLM- μ law algorithm and the lSLM- μ law-WHT-IPTS algorithm are respectively 7.5dB, 7dB and 6.4dB, so that the PAPR value of the improved SLM-PTS combined algorithm provided by the invention is respectively reduced by 1.1dB and 0.6dB before being improved, and the PAPR inhibition capability is obviously improved.

The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.

Claims (3)

1. An improved SLM-PTS peak-to-average ratio suppression method, comprising:

generating an ISLM- μ law algorithm: the method comprises the steps of (1) duplicating an input signal sequence into M groups in a frequency domain, multiplying the M groups by M phase twiddle factors respectively to generate M groups of alternative signals, and carrying out IFFT (inverse fast Fourier transform) on the M groups of alternative signals respectively to generate M groups of time domain signals; averagely dividing M groups of time domain signals into two groups, and randomly and linearly combining the two groups of signals to generate M²The/4 groups of alternative signals are selected, and a group of signals with the minimum PAPR value are selected from the alternative signals to be subjected to companding conversion so as to generate signals to be transmitted;

generating a WHT-IPTS algorithm: firstly, carrying out digital mapping on an input signal to generate N pieces of mapped data, inputting the mapped data into a precoder and multiplying the precoded data by a WHT matrix to obtain encoded frequency domain data, dividing the frequency domain data into V sub-blocks, respectively carrying out IFFT operation on the V sub-blocks, and finding out a phase factor combination with the lowest PAPR value by using a suboptimal iterative search method;

and the lSLM-mulaw algorithm and the WHT-IPTS algorithm are combined in series to obtain an ISLM-mulaw-WHT-IPTS algorithm, and the ISLM-mulaw-WHT-IPTS algorithm is adopted to carry out PAPR value suppression on signals in the OFDM system.

2. The improved SLM-PTS peak-to-average ratio suppression method according to claim 1, characterized by: when the ISLM- μ law algorithm is generated:

firstly, constellation mapping and serial-parallel conversion are carried out on original binary bit data, and an input signal sequence is copied into M groups in a frequency domain;

multiplying the M groups of signals by M phase twiddle factors respectively to generate M groups of alternative signals containing the same information;

performing IFFT on the M groups of alternative signals respectively to generate M groups of time domain signals, and averagely dividing the M groups of time domain signals into two groups, namely

X is to be^(m)And x⁽ⁿ⁾Performing linear combination to generate M²Selecting a group of signals x with the minimum PAPR value from the 4 groups of alternative signals;

carrying out companding transformation on the signal x to generate a signal x';

taking the signal x' as an input signal of the next stage after FFT operation, carrying out digital mapping on the input signal to generate N pieces of mapped data, and multiplying the N pieces of mapped data by an NxN WHT matrix through a precoder to obtain encoded data;

performing data segmentation in an interleaving segmentation mode, and segmenting frequency domain data into V subblocks;

performing IFFT transformation on the V sub-block data respectively;

using a suboptimal iterative search method, only considering b_vAnd taking the two values of-1, 1 to find out the phase factor combination with the lowest PAPR value.

3. The improved SLM-PTS peak-to-average ratio suppression method according to claim 1, characterized by: and when the phase factor combination with the lowest PAPR value is found out by adopting a suboptimal iterative search method:

s21, dividing the frequency domain data with N number of sub-carriers into V sub-data blocks which are not overlapped with each other;

s22 is made as follows_vThe PAPR under the condition is calculated as 1(V is 1,2, …, V) and 1 (index is 1)₀Max | X '|/E (| X' |), wherein

S23 is made as follows_index＝-1，Calculating the PAPR by adopting the mode in S22;

s24 PAPR₀Comparing with PAPR if PAPR is not the same₀< PAPR, then b_index1 is ═ 1; otherwise, if the PAPR is not₀＝PAPR，index＝index+1；

S25, when the index is less than or equal to V, jumping to S23, otherwise, repeating S24 until the index is V to complete the search, and finally obtaining the suboptimal twiddle factor, wherein the optimal PAPR value is min { PAPR, PAPR₀}；

And after one iteration search is finished, performing second optimization by taking the result of the first iteration search as an initial value, and repeating the steps until the required iteration times are finished.

Images