CN114465861B

CN114465861B - Method and device for reducing peak-to-average ratio of OFDM (orthogonal frequency division multiplexing) signals based on constellation rotation

Info

Publication number: CN114465861B
Application number: CN202210026981.1A
Authority: CN
Inventors: 谢映海; 李先怀; 崔宇昊; 胡泽鑫
Original assignee: Zhonghui Microelectronics Co ltd
Current assignee: Zhonghui Microelectronics Co ltd
Priority date: 2022-01-11
Filing date: 2022-01-11
Publication date: 2023-06-02
Anticipated expiration: 2042-01-11
Also published as: CN114465861A

Abstract

A method, a device and a storage medium for reducing the peak-to-average ratio of OFDM signals based on constellation rotation, wherein the method comprises the following processing procedures of signals at a signal receiving end: constellation mapping is carried out on binary data to obtain an original frequency domain signal, then IFFT operation is carried out on the original frequency domain signal to obtain a first alternative time domain signal, and Q intermediate vectors are simultaneously generated in the process of carrying out IFFT operation on the original frequency domain signal X; obtaining the remaining M-1 alternative time domain signals according to the following formula by using the Q intermediate vectors; and calculating the peak-to-average ratio of all the alternative time domain signals, and selecting the alternative time domain signal with the smallest peak-to-average ratio from the calculated peak-to-average ratio as a transmitting signal for transmission. The method can greatly reduce the calculated amount of the time-domain alternative signal calculation process.

Description

Method and device for reducing peak-to-average ratio of OFDM (orthogonal frequency division multiplexing) signals based on constellation rotation

Technical Field

The invention belongs to the technical field of signal processing, and particularly relates to a method for reducing peak-to-average ratio of an OFDM signal based on constellation rotation, a peak-to-average ratio reduction device and a storage medium.

Background

Due to the advantages of high spectral efficiency, multipath fading resistance, low computational complexity of channel equalization, etc., OFDM technology has been adopted for use by a number of digital wide band-pass standards such as 5G, 4G, 802.15.4G, digital Video Broadcast, etc. However, the OFDM system has a significant disadvantage that the peak-to-average ratio of the transmitted signal is high, which requires the power amplifier to maintain a relatively high linearity in a larger dynamic range, otherwise, when the peak value of the signal enters a nonlinear region, signal distortion is generated, which causes intermodulation interference and out-of-band radiation between subcarriers, destroys orthogonality between subcarriers, and reduces signal transmission performance. The problem of peak-to-average ratio has obvious influence on the communication terminal powered by the battery, and the increase of the linearity requirement of the power amplifier can bring the problems of hardware cost increase, battery energy utilization efficiency reduction and the like, so the peak-to-average ratio reduction technology is one of the key technologies of the OFDM system.

The current peak-to-average ratio reduction technology mainly has three main categories: signal distortion class techniques (Signal distortion techniques), coding class techniques (coding techniques) and multiple signal and probability class techniques (multiple signaling and probabilistic techniques). Among them, the signal distortion technology is to directly implement clipping filtering on the signal before passing through the power amplifier to reduce the peak-to-average ratio, and typical methods are clipping and filtering, peak windowing, scaling, and peak cancellation, etc. The coding techniques are to find a redundant code group with good peak-to-average ratio performance, and select a signal with minimum peak-to-average ratio from the found redundant code group to transmit, and typical coding techniques include precoding, filtering codes, low PMEPR codes, BCH codes, etc. The multi-signal and probability technology generally adopts two modes to reduce the peak-to-average ratio of signals, one mode is to generate a certain number of alternative signals, and then select the signal with the minimum peak-to-average ratio (PAPR) for transmission; another way is to modify the signal by phase shifting, adding a peak cancellation carrier, or constellation change, etc., and send the signal with minimized PAPR values by optimized selection of parameters, comparing the popular partial transmission sequence algorithm (partial transmission sequence) with the selective mapping algorithm (selective mapping).

In order to support higher rate data transmission, the number of subcarriers of many OFDM communication standards is currently increasing, such as mobile-WiMAX, LTE and DVB-T standards supporting up to 2048, 2048 and 8192 subcarriers, respectively. For most of the peak-to-average ratio reduction methods adopting the coding and multi-signal and probability techniques, the calculation complexity can be rapidly increased along with the increase of the number of subcarriers in a nonlinear manner, so that although the methods have excellent peak-to-average ratio reduction performance, the calculation time and the power consumption are increased due to the high calculation complexity, and the practical application value of the methods is greatly reduced. The signal distortion technology has simple realization and low calculation complexity, but the problem of increased transmission error rate caused by amplitude limiting and filtering of the transmitted signal can occur.

Disclosure of Invention

The invention aims to provide a low-complexity method for reducing peak-to-average ratio of an OFDM transmission signal based on constellation rotation, a peak-to-average ratio reduction device and a storage medium.

In order to achieve the above object, the present invention adopts the following technical solutions:

a method for reducing peak-to-average ratio of OFDM signal based on constellation rotation includes the following steps at signal receiving end:

s1-1, modulating binary data to obtain an original frequency domain signal X;

s1-2, performing IFFT operation on an original frequency domain signal X to obtain a first alternative time domain signal, and generating Q intermediate vectors simultaneously in the process of performing IFFT operation on the original frequency domain signal X: alpha ₀ 、α ₁ 、…、 α _Q-1 ；

S1-3, obtaining the remaining M-1 alternative time domain signals by using Q intermediate vectors according to the following formula:

alpha in the formula ₀ 、α ₁ 、…、α _Q-1 Respectively represent the 1 st, 2 nd, … th and Q th intermediate vectors,

for the phase sequence->

Element θ of (a) ₀ ,θ ₁ ,…,θ _N-1 For the constellation rotation angle vector θ= [ θ ] ₀ ,θ ₁ ,...,θ _N-1 ]The element value in the constellation rotation angle vector theta has periodicity and the periodicity value is Q, and the element theta in the constellation rotation angle vector theta _n The method meets the following conditions: />

S _n As the rotation angle set, a rotation angle set S _n Contains L different elements with values ranging from [ -pi, pi), and the value of the first element is 0;

s1-4, calculating peak-average ratio values of all the alternative time domain signals, and selecting the alternative time domain signal with the smallest peak-average ratio value from the calculated peak-average ratio values as a transmitting signal for transmission.

Further, the method comprises the steps of,

representing an alternative frequency domain signal,/->

Further, at the signal receiving end, after obtaining the frequency domain signal received after channel equalization, the following method is adopted to process the received signal:

receiving frequency domain sub-signals

W in the formula _n Is noise after channel equalization, ++>

To transmit frequency domain sub-signals;

the modulation constellation point set of the original frequency domain signal X is C= { C ₀ ,C ₁ ,...,C _V-1 }，

After rotation, L new star points are collected into

a) For each received frequency domain sub-signal, each element Y of the received frequency domain sub-signal is calculated one by one _n+k(Q-1) And Euclidean distance dis between L×V constellation points _n,k,l,v ，k＝0,1,…,N/Q-1：

b) Selecting the minimum Euclidean distance dis_m between each element and L new star base sets _n,k,l ：

dis_m _n,k,l ＝min([dis _n,k,l,0 ,dis _n,k,l,1 ,...,dis _n,k,l,V-1 ]),l＝0,...,L-1；

c) Calculating the sum of minimum Euclidean distances of all elements, dis_m_s _n,l ：

d) The decision mechanism for the constellation point set is as follows:

e) After the constellation point set is determined, receiving the frequency domain sub-signal according to the rotation angle omega corresponding to the constellation point set

And (3) performing inverse rotation of the corresponding angle, and equivalently recovering to a received signal of the original frequency domain signal adopting the modulation constellation point set C.

The invention also provides a peak-to-average power ratio reducing device for OFDM signals, which comprises: the device comprises a first signal processing unit, a signal transmitting unit, a second signal processing unit and a signal receiving unit;

a first signal processing unit for modulating binary data to obtain an original frequency domain signal;

the first signal processing unit is further used for performing IFFT operation on the original frequency domain signal to obtain a first alternative time domain signal, and obtaining the rest alternative time domain signal by using an intermediate vector obtained in the IFFT operation process;

the first signal processing unit is further used for calculating the peak-to-average ratio value of all the alternative time domain signals, and selecting the alternative time domain signal with the smallest peak-to-average ratio value from the peak-to-average ratio values.

The signal transmitting unit is used for transmitting the alternative time domain signal with the minimum peak-to-average ratio as a transmitting signal;

a signal receiving unit for receiving the signal transmitted by the signal transmitting unit from the transmission channel;

and the second signal processing unit is used for recovering the received signal received by the signal receiving unit to the received signal of the original frequency domain signal adopting the modulation constellation point set.

The present invention also provides a computer storage medium having stored thereon a plurality of instructions adapted to be loaded by a processor and to carry out the foregoing method steps.

According to the technical scheme, the alternative frequency domain signals are generated by applying constellation rotations with different angles to the sub-signals at different positions of the original frequency domain signals, the intermediate vectors of the IFFT recursion calculation process are utilized, the calculation process of the alternative time domain signals is simplified into linear addition of the intermediate vectors, and therefore the complex multiplication quantity is reduced to the level of about 0.005% -0.05% of the traditional SLM method; the method has the advantages that the calculation complexity of the generation process of the alternative time domain signal of the transmitting end is greatly reduced through the specific phase sequence, and sideband information is not required to be carried. On the basis, even if the transmitting end does not carry any sideband information, the receiving end can accurately identify the constellation rotation angle of each sub-signal through a constellation detection mechanism. Theoretical and simulation results show that through parameter combination between the signal rotation position and the angle, the method does not need to save a large number of phase sequences at both receiving and transmitting ends, and generates a large number of alternative signals in a very simple mode, so that the method has a large performance optimization space, and the method approximates to or even surpasses the limit performance of the traditional SLM method based on the Hadamard orthogonal matrix in terms of signal peak-to-average ratio performance, so that the method has high application value.

Drawings

In order to more clearly illustrate the embodiments of the present invention, the following description will briefly explain the embodiments or the drawings required to be used in the description of the prior art, it being apparent that the drawings in the following description are only some embodiments of the present invention and that other drawings may be obtained from these drawings without inventive effort to those skilled in the art.

FIG. 1 is a signal processing flow chart of a conventional SLM peak-to-average-value-decreasing method at a signal transmitting end;

FIG. 2 is a schematic diagram of an IFFT recursion tree operation process;

FIG. 3 is a flow chart of signal processing at the signal transmitting end of the method of the present invention;

fig. 4 is a tree calculation diagram when q=16;

fig. 5 is a schematic diagram of a set of l=3 new constellation points during BPSK modulation;

fig. 6 is a diagram of l=2 new constellation point sets at the time of QPSK modulation;

FIG. 7 is a graph of performance contrast between the constellation detection error rate of the scheme of FIG. 5 and the packet error rate of the BPSK+1/2Turbo scheme;

FIG. 8 is a graph of performance contrast between the constellation detection error rate of the scheme of FIG. 6 and the packet error rate of the QPSK+1/2Turbo scheme;

FIG. 9 is a graph of a simulation of the peak-to-average ratio performance of a conventional SLM method and the method of the present invention for signals modulated with BPSK;

FIG. 10 is a graph of a simulation of the peak-to-average power ratio performance of a conventional SLM method and the inventive method for a signal modulated with QPSK;

fig. 11 is a block diagram of a peak-to-average power ratio reducing device according to an embodiment of the present invention.

Detailed Description

To make the above and other objects, features and advantages of the present invention more apparent, the following detailed description of the embodiments of the present invention will be given with reference to the accompanying drawings.

The SLM (selective mapping) method is a method with better peak-to-average power ratio reducing performance in a multi-signal and probability technology. As shown in FIG. 1 (symbols in FIG. 1)

The dot product operation representing two vectors), the signal processing flow of the conventional SLM peak-to-average-reduction ratio method at the signal transmitting end is as follows: after constellation mapping (Constellation mapping) is carried out on a Binary data stream (Binary data), an original frequency domain signal is obtained, and the original frequency domain signal is combined with R (r=2 ^r ) Phase sequence p= [ P ] ₀ ,P ₁ ,…,P _R-1 ]And respectively performing point multiplication to generate R alternative frequency domain signals, respectively performing IFFT operation on the alternative frequency domain signals, and finally selecting the OFDM signal with the minimum peak-to-average ratio from all operation results to transmit.

In order for the receiving end to correctly demodulate the original frequency domain signal, the transmitting end needs to transmit the phase sequence information corresponding to the transmitting signal to the receiving end, namely, the receiving end needs to carry sideband information of r bits, and more coding redundancy than data information needs to be adopted to protect the transmission of the information, so that the information transmission efficiency of the signal can be reduced. In addition, in order to obtain better peak-to-average power ratio performance, the number of phase sequences needs to be increased, so that the number of times of IFFT operation needed to be performed by a transmitting end is correspondingly increased, and the calculation complexity is linearly increased along with the value of R. Based on the above, the invention is to improve the existing SLM method by reducing the operation times of IFFT in the generation process of the alternative time domain signals, so as to realize low calculation complexity and avoid carrying sideband information.

In the process of generating an OFDM signal, the calculation amount is mainly focused on a calculation process of transforming a frequency domain signal into a time domain, that is, the following discrete inverse fourier transform process (hereinafter referred to as IDFT):

wherein, x= [ X (0), X (1),. The term X (N-1)]Representing the original frequency domain signal, x= [ x (0), x (1), x (N-1)]Represents a time domain signal corresponding to the original frequency domain signal, k=0, 1, …, N is the number of subcarriers in the OFDM signal, and IDFT (X) represents the discrete inverse fourier transform result.

In order to reduce the calculation amount of IDFT, inverse Fast Fourier Transform (inverse fast fourier transform, IFFT for short) uses the concept of recursive calculation to successively decompose long-sequence IDFT into shorter-sequence IDFT, thereby letting the calculation complexity of time-domain signal x (k) be changed from O (N) ² ) To be reduced to

Level. That is, for a signal with the number of subcarriers being N, the IDFT operation of N points may be split into 2 IFFT operations of N/2 points, the IFFT operation of N/2 points may be split into 2 IDFT operations of N/4, and so on, the recursive computation procedure between the IFFT operation of N points and the IFFT operation of 2N/2 points is as follows:

x in the formula ₁ An even signal, X, which is the original frequency domain signal ₂ An odd sub-signal, X, of the original frequency domain signal ₁ ＝X(0:2:N-2)＝[X(0),X(2),...,X(N-2)]，X ₂ ＝X(1:2:N-1)＝[X(1),X(3),...,X(N-1)]；

Due to

And->

The above formula can therefore be equivalently reduced to:

is->

Then:

h is a complex constant vector of length N and known value.

Although the IFFT can effectively reduce the amount of computation of the signal transformed from the frequency domain to the time domain, the amount of computation is still at a high level with a large value of N. Therefore, the number of IFFT operations is reduced as much as possible in the generation process of the alternative time domain signal, which becomes a main direction of improvement of the SLM method, and meanwhile, the process flow of signal transmission and reception is simplified by further considering avoiding the carrying of sideband information, so that the data transmission efficiency is improved.

In the process of performing IFFT recursive computation on IDFT (X), numerous known intermediate vectors are generated, and a strict functional relationship exists between the intermediate vectors and IDFT (X).

For the original frequency domain signal x= [ X (0), X (1), X (N-1)]In the discrete inverse Fourier transform process, the following Q is generated by IFFT recursive computation of IDFT (X)

q is positive integer) of intermediate vectors: IDFT%X ₀ ),IDFT(X ₁ ),...,IDFT(X _Q-1 ) Wherein, the method comprises the steps of, wherein,

the sum of these Q intermediate vectors is equal to the discrete inverse fourier transform result IDFT (X), i.e.:

specifically, the process of generating Q intermediate vectors in the IFFT recursive computation is shown in fig. 2, the open nodes in fig. 2 represent complex constant vectors with known values, the solid nodes represent intermediate vectors that accompany the IDFT (X) computation,

[.] ^T representing the transpose of the vector. As shown in fig. 2, +.o-in fig. 2 represents the dot product operation of two vectors, IDFT (X) operation of N points is split into 2N/2 points +.>

Calculation and splitting into 4N/4 points +.>

Calculating, recursively dividing the calculation into +.>

2 points of

The operation is carried out by the method,

from the recursive tree calculation process from bottom to top of FIG. 2, the slave

Solid nodes in all levels to level 1, including q=2 of level Q ^q Are each independently->

Is an intermediate vector that accompanies the IDFT (X) computation process. The values of column vectors of length N/Q corresponding to the Q solid nodes of level Q in FIG. 2 are a in sequence from left to right ₀ ,a ₁ ,...,a _Q-1 Based on

It can be seen that 2 of the level q-1 above ^q-1 The column vectors with the lengths of 2N/Q corresponding to the solid nodes are sequentially as follows from left to right: />

2 of the level q-2 ^q-2 Column vector of length 4N/Q corresponding to each solid node, i-th vector psi from left to right _q-2,i The method comprises the following steps:

Ω _4N/Q and omega _2N/Q Complex constant vectors, all of which are known in value, the i-th vector can be reduced to:

wherein delta is ₁ ，Δ ₂ And delta ₃ Are complex constant column vectors with the length of 4N/Q and known values;

will be psi _q-2,i Further recursively up to level 0, from the perspective of the polynomial function, the Q column vectors a of length N/Q of IDFT (X) and level Q ₀ ,a ₁ ,...,a _Q-1 There is a strict functional relationship between them, specifically:

g (a) _i ) Representing a column vector a of length N/Q _i Repeating Q times to form a new column vector with length N, f _i (. Cndot.) represents the point multiplication of the vector in brackets with a known complex constant vectorCalculating, e.g. +.>

Representing the sum of the vector z and the known complex constant vector Φ _i Performing a dot product operation, i=1, 2, …, Q-1; then:

/>

thus, in the IFFT recursive computation process of IDFT (X), Q intermediate vectors IDFT with length of N are generatedX ₀ ),IDFT(X ₁ ),...,IDFT(X _Q-1 ) And the sum of these intermediate vectors is equal to IDFT (X), i.e

This is true.

Based on the conclusion, in order to avoid generating alternative time domain signals through IFFT operation with higher computational complexity, the invention improves the prior SLM method, carries out constellation mapping on binary data to obtain original frequency domain signals, carries out dot multiplication on the original frequency domain signals and phase sequences to generate alternative frequency domain signals

The phase sequence is +.>

Alternative frequency domain signal

θ in the formula ₀ ,θ ₁ ,…,θ _N-1 As an element in the constellation rotation angle vector θ, the constellation rotation angle vector θ= [ θ ] ₀ ,θ ₁ ,...,θ _N-1 ]The element values in the constellation rotation angle vector theta have periodicity and the periodicity value is Q, namely [ theta ] ₀ ,θ ₁ ,...,θ _Q-1 ]＝[θ _Q ,θ _Q+1 ,...,θ _2Q-1 ]＝...＝[θ _N-Q ,θ _Q+1 ,...,θ _N-1 ]Element θ of vector θ _n The following conditions are satisfied:

S _n as the rotation angle set, a rotation angle set S _n Contains L different elements with values ranging from [ -pi, pi), and the value of the first element is 0.

From this, the transmitting end divides the original frequency domain signal into q=2 ^q The number of constellation rotation angles selectable by each sub-signal is L, and M=L can be generated in total ^Q And the individual alternative frequency domain signals. For the aforementioned alternative frequency domain signal

Corresponding +.>

The method meets the following conditions:

IDFT(X ₀ )、IDFT(X ₁ ),...,IDFT(X _Q-1 ) Representing a number of intermediate vectors generated during the generation of the alternative frequency domain signal using IFFT operations.

Due to the strict functional relationship between the intermediate vector and IDFT (X), i.e. vectors of length N/Q

And a vector IDFT of length NX _k ) There is a strict functional relationship between +.>

Then:

γ _k is a known complex constant vector of length N;

while

Thus there is

While

Therefore there is->

As can be seen from the above description, for the aforementioned alternative frequency domain signal

When the signal is converted into the time domain, the calculation process can be equivalently simplified to the steps that the intermediate vectors with known Q number values are multiplied by a complex rotation coefficient with a module of 1, and then vector addition is carried out, so that the conversion between the signal frequency domain and the time domain is not needed to be carried out by using IFFT operation with higher complexity. The method of the present invention will be described with reference to fig. 3, and the steps of the method of the present invention are as follows:

at the signal transmitting end, the signal processing process is as follows:

s1-1, constellation mapping is carried out on binary data to obtain an original frequency domain signal X;

s1-2, performing IFFT operation on an original frequency domain signal X to obtain a first alternative time domain signal, and generating Q intermediate vectors simultaneously in the process of performing IFFT operation on the original frequency domain signal X: IDFT%X ₀ )＝α ₀ ,IDFT(X ₁ )＝α ₁ ,...,IDFT(X _Q-1 )＝α _Q-1 ；

S1-3, obtaining remaining M-1 alternative time domain signals according to the following formula by using Q intermediate vectors obtained in the IFFT operation process of the original frequency domain signal X in the step S1-2:

for the phase sequence->

Element θ of (a) ₀ ,θ ₁ ,…,θ _N-1 As an element in the constellation rotation angle vector θ, the constellation rotation angle vector θ= [ θ ] ₀ ,θ ₁ ,...,θ _N-1 ]The element values in the constellation rotation angle vector theta have periodicity and the periodicity value is Q, namely [ theta ] ₀ ,θ ₁ ,...,θ _Q-1 ]＝[θ _Q ,θ _Q+1 ,...,θ _2Q-1 ]＝...＝[θ _N-Q ,θ _Q+1 ,...,θ _N-1 ]Element theta in constellation rotation angle vector theta _n The method meets the following conditions: />

S _n For the rotation angle set S _n Contains L different elements with values of [ -pi, pi), and the value of the first element is 0 +.>

For an alternative frequency domain signal, < >>

S1-4, respectively calculating the generated total M=L ^Q And selecting one candidate time domain signal with the smallest peak-to-average ratio value from the peak-to-average ratio values of the candidate time domain signals as a transmitting signal for transmission. The method for calculating the peak-to-average ratio is a conventional method, and is not an innovation of the present invention, and will not be described herein.

After the original frequency domain signal is subjected to IFFT operation to obtain a first alternative time domain signal, the Q intermediate vectors generated in the calculation process of the first alternative time domain signal are utilized to multiply a coefficient respectively, then the residual alternative time domain signals are obtained by adding, a group of coefficient values correspondingly generate an alternative time domain signal, and L is the sum ^Q-1 The coefficient values are assembled to obtain M-1 alternative time domain signals. Signal flow from the above signal transmitting endThe method of the invention fully utilizes the intermediate vector generated in the IFFT recursive computation process of the 1 st alternative frequency domain signal (namely the original frequency domain signal, and the constellation rotation angle vector theta is zero vector at the moment), and reduces the equivalent computation process of transforming the rest alternative frequency domain signals from the frequency domain to the time domain into the linear addition of the intermediate vector, thereby greatly reducing the complex multiplication and addition quantity and having ultralow computation complexity. Moreover, the transmitting end performs constellation rotation on the corresponding frequency domain sub-signals respectively, and the number of selectable rotation angles is limited (typical values of L are 2 and 4), so that the receiving end can accurately judge the rotation angle corresponding to each sub-signal through a constellation detection mechanism, can accurately analyze the received signal under the condition that no sideband information is carried, and can avoid the carrying of the sideband information. The method of the invention has excellent peak-to-average ratio reduction performance, for example, when L=2 and Q=16, the generated alternative quantity can reach M=2 ¹⁶ The number of the components is 65536, a larger performance optimization space is brought, and an ideal peak-to-average power ratio reduction effect can be obtained.

Because the signal sent by the signal sending end of the method does not carry sideband information, on the basis of the technical scheme, a constellation detection mechanism is adopted at the signal receiving end to recover the received signal corresponding to the original frequency domain signal. At the signal receiving end, the signal processing process is as follows:

the signal receiving end receives signals, and the received frequency domain signals obtained after channel equalization are Y= [ Y (0), Y (1),. The number is Y (N-1)]Receiving frequency domain sub-signals

Receiving frequency-domain sub-signals>

And transmit the frequency domain sub-signal +.>

The relationship between them is as follows: />

W in the formula _n Noise after channel equalization;

I.e. L new star point sets after rotation

The intersection of any two sets is an empty set, so that the resolvable property between the new star seat sets is ensured;

for receiving frequency domain sub-signals

The method judges which constellation set is adopted by N/Q elements during modulation through a minimum distance mechanism, and comprises the following specific processes:

a) For each received frequency domain sub-signal, each element Y of the received frequency domain sub-signal is calculated one by one _n+k(Q-1) And Euclidean distance dis between L×V constellation points in new set of constellation points Θ _n,k,l,v ， k＝0,1,…,N/Q-1：

c) Calculating the sum of minimum Euclidean distances of all N/Q elements, dis_m_s _n,l ：

d) The decision mechanism for the constellation point set is as follows:

The invention improves the traditional SLM method, and when the carrier number of the OFDM signal is N=2 in terms of computational complexity ^K When K is a positive integer, the traditional SLM method completes one IFFT operation to respectively carry out the requirements

Multiply by complex number and->

And (5) performing complex addition. Thus if the conventional SLM method is used to generate m=l ^Q The individual alternative signals, the whole process in total requires +.>

Multiply by complex number and->

And (5) performing complex addition.

The method of the invention only needs to perform a complete IFFT operation in the 1 st alternative time domain signal generation process, thus the method needs to

Multiply by complex number and->

And (5) performing complex addition. In the calculation of the remaining Yu Bei time domain signals, each intermediate vector alpha can be calculated and stored _n And then multiplying the value by L-1 rotation coefficients, adding alpha without calculation _n (the 1 st rotation coefficient value is fixed to 1) to obtain a vector set

And then respectively taking out one vector from the Q vector sets for addition, thus completing the calculation process of an alternative frequency domain signal. Thus remaining L ^Q The number of complex multiplications required for the calculation process of the 1 alternative time domain signals in total is num_mul=q (L-1) N; in terms of the number of complex additions, the repetition of computations can be avoided by a tree computation method, thereby minimizing the number of additions required.

Fig. 4 is a schematic diagram of tree computation when q=16, and the tree computation method specifically includes: aggregating Q vectors eta _n N=0, 1..q-1 is divided into Q/2 groups, then two sets of vectors each containing L vectors for each group are each taken out of one vector and added, and then the combination produces one vector containing L ² The new vector set of individual vectors can form Q/2 each containing L after this layer is calculated ² The vector set of individual vectors, denoted η _0,1 ,η _2,3 ,...,η _Q-2,Q-1 The total number of complex additions required for this process is

Then the Q/2 each containing L ² The vector set of the individual vectors is divided into Q/4 groups, each of which then contains L ² Two vector sets of vectors are each added together to form a vector containing L ⁴ A new vector set of individual vectors, which can form Q/4 each containing L after completion of the calculation ⁴ Vector set of individual vectors, denoted η _0,1,2,3 ,...,η _{Q-4,Q-3,Q-2,Q-1} The total number of complex additions required for this procedure is +.>

Repeating the above steps until the last layer is completed to form 1 film containing L ^Q Vector set of individual vectors, the total number of complex additions required for this process is 1 XL ^Q X N. Thus, the total number of complex additions to complete the whole calculation process is

The calculated reduction rate CNRR (computational number reduction ratio) between the method of the present invention and the conventional SLM method is:

cnrr= (1-calculated amount of the method of the present invention/calculated amount of the conventional SLM method) ×100.

Table 1 shows the rate of reduction of the multiplications and additions calculated for the method of the invention compared to the conventional SLM method, under some typical parameters.

TABLE 1

As can be seen from the results of table 1, the method of the present invention reduces the calculation process of the remaining alternative time domain signals from the IFFT operation equivalent with larger calculation amount to the linear addition of Q known intermediate vectors with smaller number, and reduces the complex multiplication number to the level of about 0.005% -0.05% of the conventional SLM method in case of more carriers, thereby generating a huge number of alternative signals for the transmitting end to realize the improvement of the peak-to-average ratio performance, and eliminating the biggest obstacle.

Based on the constellation rotation angle detection mechanism of the sub-signals at the signal receiving end, if the constellation detection result of a certain sub-signal is wrong, the method is equivalent to introducing some phase interference noise to the N/Q received signals, thereby reducing the overall decoding performance. In order to simplify the analysis, the error rate performance analysis is performed by taking one OFDM symbol as a basic unit, and it is assumed that only more than or equal to 1 error occurs in Q constellation detection results in one OFDM signal, so that the information transmission of the whole OFDM symbol is wrong. Under the same channel transmission condition, setting the transmission error rate of one OFDM symbol without adopting the peak-to-average ratio reduction technology and adopting the peak-to-average ratio reduction technology as Pe ₁ And Pe, the constellation detection error rate of 1 sub-signal is Pe ₂ The following steps are:

Pe＝1-(1-Pe ₂ ) ^Q (1-Pe ₁ )≈QPe ₂ +Pe ₁ ,if Pe ₁ →0,Pe ₂ 0, and

Pe≈Pe ₁ ,if QPe ₂ ＜＜Pe ₁ ；

namely, under the condition of the same signal-to-noise ratio, the error rate of the constellation detection at the receiving end is required to be far lower than the signal transmission error rate performance when the peak-to-average ratio reduction technology is not adopted, so that the method can ensure that the signal transmission error rate performance is basically not influenced.

The transmission error rate performance of the method of the present invention is verified by simulation experiments. When the signal adopts BPSK modulation, the rotation angle set is S _n = {0,3 pi/4, -3 pi/4 }, n=0, 1..q-1, the resulting schematic diagram of the l=3 new set of star points is shown in fig. 5. When the signal adopts QPSK modulation, the rotation angle set is that

The resulting l=2 new set of star points is shown in fig. 6.

In the case of AWGN channel with N/q=256, fig. 7 and fig. 8 show the performance comparison between the constellation detection error rate of the schemes in fig. 5 and fig. 6 and the packet error rate of the bpsk+1/2Turbo coding scheme and the qpsk+1/2Turbo coding scheme based on the 4G physical layer communication standard, respectively, and it can be seen from the results in fig. 7 and fig. 8 that, in the case of the same channel and signal-to-noise ratio, the constellation detection error rate of the method of the present invention is far lower than the signal transmission packet error rate performance without the peak-to-average ratio technique, i.e., the condition QPe ₂ ＜＜Pe ₁ Sufficient satisfaction can be obtained so that the transmission error rate performance of the method of the present invention is not degraded.

The peak-to-average ratio reduction effect of the method is verified through a simulation experiment. The distribution of the peak-to-average ratio (PAPR) of a signal is described by using a time-domain Complementary Cumulative Distribution Function (CCDF), and the mathematical calculation formula is Pr (PAPR > z) =1-Pr (PAPR. Ltoreq.z). Fig. 9 is a graph of peak-to-average ratio performance for the inventive method and a conventional SLM method based on Hadamard orthogonal matrix and using all its vectors to generate a backup signal when the constellation rotation scheme of fig. 5 is used and q=8 when the number of signal subcarriers n=2048 and BPSK modulation technique is employed. Fig. 10 is a graph of peak-to-average ratio performance for the inventive method and the conventional SLM method based on Hadamard orthogonal matrix with the number of back-up signals M reaching its upper limit (m=n=4096) using the constellation rotation scheme of fig. 6 and q=16 when using the QPSK modulation technique.

As can be seen from FIG. 9, the phase sequence used in the method of the present invention for dot multiplication with the original frequency domain signal, compared with the conventional SLM method

The periodicity needs to be strictly met, the element value range is also greatly limited, so that mutual independence among different alternative signals is weak, and the optimization selection of signal peak-to-average ratio is not facilitated, so that the peak-to-average reduction performance of the method is equivalent to that of a traditional SLM method and slightly worse under the condition that the number of generated backup signals is about 3 times that of the traditional SLM method based on a Hadamard orthogonal matrix; however, as can be seen from fig. 10, in the case that the number of generated backup signals is 16 times that of the conventional SLM method, the peak-to-average power ratio performance of the method of the present invention has exceeded the upper limit performance of the conventional SLM method. In a combined way, the method has the advantage of ultralow computational complexity, and can make up the defect of weak independence between backup signals under the condition of more subcarriers, thereby obtaining very excellent peak-to-average power ratio performance.

The following are device embodiments of the present application, which may be used to perform method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

As shown in fig. 11, a block diagram of a peak-to-average power ratio reducing apparatus for OFDM signals according to an exemplary embodiment of the present application is shown. The apparatus may be implemented in software, hardware or a combination of both. The peak-to-average power ratio reducing device for the OFDM signal comprises a first signal processing unit, a signal transmitting unit, a second signal processing unit and a signal receiving unit.

The first signal processing unit is used for carrying out constellation mapping modulation on binary data so as to obtain an original frequency domain signal X;

the first signal processing unit is further used for performing IFFT operation on the original frequency domain signal X to obtain a first alternative time domain signal, and obtaining the remaining M-1 alternative time domain signals by using an intermediate vector obtained in the IFFT operation process;

the first signal processing unit is also used for calculating the peak-to-average ratio value of all the alternative time domain signals and selecting the alternative time domain signal with the minimum peak-to-average ratio value from the peak-to-average ratio values;

In one or more possible embodiments, the second signal processing unit is further configured to employ a constellation detection mechanism on the received frequency domain sub-signals to recover the received signals to the original frequency domain signals employing the set of modulation constellation points.

It should be noted that, when the apparatus provided in the foregoing embodiment performs the foregoing peak-to-average ratio reduction method, only the division of the foregoing functional modules is used as an example, in practical application, the foregoing functional allocation may be performed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the peak-to-average ratio reduction device for the OFDM signal and the method embodiment for reducing the peak-to-average ratio of the OFDM signal provided in the foregoing embodiments belong to the same concept, and the implementation process is shown in the method embodiment, which is not repeated here.

The embodiment of the present application further provides a computer storage medium, where a plurality of instructions may be stored, where the instructions are adapted to be loaded by a processor and execute the method steps of the embodiment shown in fig. 3, and the specific execution process may refer to the specific description of the embodiment shown in fig. 3, which is not repeated herein.

The concept of the present embodiment is the same as that of the method embodiment of fig. 3, and the technical effects thereof are the same, and the specific process may refer to the description of the embodiment of fig. 3, which is not repeated here.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a read-only memory, a random access memory, or the like.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The method for reducing the peak-to-average ratio of the OFDM signal based on constellation rotation is characterized by comprising the following steps of:

at the signal transmitting end, the signal processing process is as follows:

s1-1, modulating binary data to obtain an original frequency domain signal;

s1-2, performing IFFT operation on the original frequency domain signal to obtain a first alternative time domain signal, and generating Q intermediate vectors in the process of performing IFFT operation on the original frequency domain signal: alpha ₀ 、α ₁ 、…、α _Q-1 ；

for the phase sequence->

Element θ of (a) ₀ ,θ ₁ ,…,θ _N-1 For the constellation rotation angle vector θ= [ θ ] ₀ ,θ ₁ ,...,θ _N-1 ]The element value in the constellation rotation angle vector theta has periodicity and the period value is Q, and the element theta in the constellation rotation angle vector theta _n The method meets the following conditions: />

2. The method for reducing peak-to-average ratio of an OFDM signal according to claim 1, wherein:

representing an alternative frequency domain signal,/->

X (0), X (1), …, X (N-1) is the original frequency domain signal x= [ X (0), X (1),. The term X (N-1)]The 1 st, 2 nd, … th element, N is the number of subcarriers in the OFDM signal.

3. The method for reducing peak-to-average ratio of an OFDM signal according to claim 1, wherein: at a signal receiving end, after receiving a frequency domain signal after channel equalization, processing the received signal by adopting the following method:

receiving frequency domain sub-signals

W in the formula _n Is noise after channel equalization, ++>

To transmit frequency domain sub-signals;

After rotation, L new star points are collected into

d) The decision mechanism for the constellation point set is as follows:

e) After the constellation point set is determined, receiving the frequency domain sub-signals according to the rotation angle omega corresponding to the constellation point set

And (3) performing inverse rotation of the corresponding angle, and equivalently recovering the received signal of the original frequency domain signal adopting the modulation constellation point set C.

4. A peak-to-average power ratio reduction apparatus for an OFDM signal, comprising: the device comprises a first signal processing unit, a signal transmitting unit, a second signal processing unit and a signal receiving unit;

the first signal processing unit is further configured to perform an IFFT operation on the original frequency domain signal to obtain a first alternative time domain signal, and obtain a remaining alternative time domain signal according to the following formula by using an intermediate vector obtained in the IFFT operation process:

for the phase sequence->

5. The peak-to-average power ratio reducing apparatus for OFDM signal as set forth in claim 4, wherein: at the signal receiving end, after obtaining the received frequency domain signal after channel equalization, the processing procedure of the second signal processing unit on the received signal is as follows:

receiving frequency domain sub-signals

W in the formula _n Is noise after channel equalization, ++>

To transmit frequency domain sub-signals;

After rotation, L new star points are collected into

/>

d) The decision mechanism for the constellation point set is as follows:

6. A computer storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform the method steps of any one of claims 1 to 3.