CN103812817B - Peak-to-average power ratio inhibition method for orthogonal frequency division multiplexing (OFDM) signal - Google Patents

Abstract

The invention discloses a peak-to-average power ratio inhibition method for an orthogonal frequency division multiplexing (OFDM) signal. Through the method, the problems of fixation in the selection of parameters and low inhibition performance of signal peak-to-average power ratios in the prior art are solved. The method includes the following steps of: 1) up-sampling an OFDM modulating signal to obtain an original OFDM signal; 2) selecting a companding function and setting an initial value of an iteration parameter; 3) performing companding transformation on the original OFDM signal by using the companding function; 4) transforming the signal subjected to companding to a frequency domain for frequency domain filtering, transforming to a time domain to obtain a transmission signal, and calculating the peak-to-average power ratio of the transmission signal; and 5) judging whether iteration is finished or not according to the iteration parameter, if so, acquiring the transmission signal which meets the peak-to-average power ratio (PAPR) requirements of a system, and otherwise, continuing to iterate. According to the peak-to-average power ratio inhibition method, the peak-to-average power ratio of the OFDM signal can be reduced obviously under the condition of little effect on the bit error rate (BER) of the system, and thus, the peak-to-average power ratio inhibition method can be used for a new generation of various broadband OFDM wireless communication systems.

Description

Peak-to-average power ratio suppression method for Orthogonal Frequency Division Multiplexing (OFDM) signal

Technical Field

The invention belongs to the technical field of wireless communication, relates to a peak-to-average power ratio (PAPR) suppression method for wireless Orthogonal Frequency Division Multiplexing (OFDM) modulation transmission signals, and can be widely applied to various new-generation broadband OFDM wireless communication systems.

Background

The OFDM modulation technique has the advantages of high spectrum efficiency and effective suppression of multipath fading and intersymbol interference, and can solve many problems in a wireless mobile environment while effectively utilizing spectrum resources, thus becoming a popular technique in wireless communication applications. OFDM modulation divides a signal sequence into N parallel sequences, each parallel sequence modulating 1 subcarrier, each subcarrier occupying less than the associated bandwidth of the channel, so that the signal experiences a frequency non-selective channel. However, since the OFDM signal is formed by adding a plurality of independent subcarrier signals, the OFDM system has a serious disadvantage: when the phases of a plurality of subcarrier signals are the same, a high peak-to-average ratio (PAPR) is generated by mutual superposition. If the peak-to-average ratio of the transmission signal is high, in order to ensure that the signal can be linearly amplified, a transmitter power amplifier is inevitably required to have a wide linear dynamic range, and the precision of a D/A digital-to-analog converter is also required to be higher, so that the construction cost of the OFDM transmitter is greatly increased; on the contrary, although the power amplifier with narrow linear dynamic range can reduce the construction cost of the transmitter, it will cause serious distortion of the transmitted signal, so that the performance of the OFDM system is drastically reduced.

There are many existing methods for reducing the peak-to-average ratio of the OFDM signal, such as an iterative filtering method and a signal companding transform method. The main ideas of the existing various companding transformation methods are as follows: and carrying out distortion processing on the signal by using a companding function. Huang, for example, proposes a linear Companding method in the "compressing transform for reduction in peak-to-average power ratio of ofdm signals", which utilizes a segmentation function to respectively compand small signals and large signals, and can effectively reduce the peak-to-average ratio of signals under the condition of less influence on the system performance. But the method only optimizes the signal peak-to-average ratio, but cannot simultaneously control the BER performance and the out-of-band spectrum spreading performance of the system; in addition, in the method, the companding parameter is a fixed value, and real-time adaptive adjustment cannot be performed according to an actual signal. The modern iterative filtering method adopts an iterative structure, reduces the peak-to-average ratio of signals by multiple amplitude limiting filtering, and can perform joint optimization control on the performance of a plurality of systems by utilizing an iterative process, but the required iteration times are large, so that the calculation complexity of the system can be greatly increased.

Disclosure of Invention

The invention aims to provide a peak-to-average power ratio (PAPR) suppression method of Orthogonal Frequency Division Multiplexing (OFDM) signals, which combines iterative filtering and companding transformation, aiming at overcoming the defects of the prior art, so as to control the BER performance and the out-of-band spectrum spreading performance of a system while obviously reducing the PAPR of the OFDM signals, reduce the computational complexity of the system and improve the overall performance of the system.

The basic idea for realizing the invention is as follows: the iterative idea of the iterative filtering method is applied to the companding conversion method, the peak-to-average ratio (PAPR) performance, the out-of-band spectrum spreading performance and the system Bit Error Rate (BER) performance of a signal are controlled and parameters are adjusted in real time through iteration, and the technical scheme comprises the following steps:

1. a method for suppressing a peak-to-average power ratio of an Orthogonal Frequency Division Multiplexing (OFDM) signal comprises the following steps:

(1) up-sampling OFDM modulation signal to obtain original OFDM signal x_nWherein N =0,1, …, JN-1, J denotes an upsampling factor, N denotes the number of subcarriers included in OFDM modulation, and JN denotes the number of subcarriers included in the OFDM system after upsampling;

(2) selecting a linear piecewise companding function f (x):

wherein x represents the input signal of a linear function, f (x) represents the output signal of a linear function,as a function of the first segmentSlope of x, 0<k<1 is the slope of the second segment function kx, 0<v is not more than max { | x | } is an inflection point, |, is a modulo operator, 0<V is less than or equal to | x |, which means that the input signal meeting the condition is a small signal, and | x | luminance>v represents that the input signal satisfying this condition is a large signal;

(3) enabling the iteration number M to be =1, setting a maximum iteration number M and an attenuation factor alpha according to a peak-to-average power ratio (PAPR) and a system Bit Error Rate (BER) required by the system, and starting an iteration process;

(4) from the value α, the slope k and inflection point v of the second-stage function kx are obtained by the following equations, and the first-stage function is obtainedSlope of x

WhereinIs the original OFDM signal x_nM denotes the original OFDM signal x_nAverage value of (d);

(5) applying companding function f (x) to original OFDM signal x_nPerforming companding conversion to obtain companding conversion signal y_n：

n=0,1,…,JN-1；

(6) To companded transform signal y_nFFT to obtain frequency domain signal C_n；

(7) Using frequency domain filter to process FFT frequency domain signal C_nFiltering to obtain filtered signalThen the filtered signal is processedIFFT conversion is carried out to obtain a new transmission signal

(8) Computing a transmission signal from a peak-to-average ratio definitionAnd outputting a transmission signal according to an iteration result:

if m<M, making the iteration number M = M +1, and using the transmission signalReplacing the original OFDM signal x_nSetting an attenuation factor α according to the current PAPR value, and returning to the step (4) to continue execution;

if M = M, the iteration is ended and the transmission signal is output

The invention applies the iterative idea of the amplitude limiting filtering method to the companding conversion method, realizes the combined optimization of the PAPR performance of the signal and the BER performance of the system bit error rate through iteration, and adds the processing of the frequency domain signal in the optimization process, thereby obtaining good BER and PSD performance of the system bit error rate while obviously reducing the PAPR of the OFDM signal.

Drawings

FIG. 1 is a flow chart of the present invention;

FIG. 2 is a system block diagram of a companding transformation process;

FIG. 3 is a parameter v of a linear piecewise function in the present invention_nA relation to k represents a diagram;

FIG. 4 is a graph of the simulated effect of peak-to-average ratio rejection performance of the present invention and the three prior art methods;

FIG. 5 is a diagram of the simulation effect of the frequency spectrum performance of the present invention and the three prior art methods;

FIG. 6 is a graph showing the simulation effect of the error rate in an additive white Gaussian noise channel according to the present invention and the three prior art methods;

fig. 7 is a diagram of the simulation effect of the bit error rate under the leis fading channel of the present invention and the three existing methods.

Detailed Description

Embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the protection scope of the present invention is not limited to the following embodiments.

Referring to fig. 1, the specific implementation steps of the present invention are as follows:

the method comprises the following steps: up-sampling the OFDM modulated signal to obtain original OFDM signal x_nWhere N =0,1, …, JN-1, J denotes an upsampling factor, N denotes the number of subcarriers included in the OFDM system, and JN denotes the number of subcarriers included in the OFDM system after upsampling.

Step two: a linear piecewise companding function f (x) is selected.

According to the basic idea of reducing the large signal and amplifying the small signal in the input signal, a linear piecewise companding function f (x) is selected:

wherein x represents the input signal of the linear piecewise companding function, f (x) represents the output signal of the linear piecewise companding function,x represents a first-stage function of the first-stage,representing a first-stage functionThe slope of x, kx is the second-stage function, k is the slope of the second-stage function kx, 0<k<1, v is an inflection point, 0<v is not more than max { | x | }, |, is a modulo operator, satisfies 0<The input signal of the condition that | x | < v is a small signal, and satisfies | x | < v |>The input signal of the v condition is a large signal;

step three: setting an initial value:

let M =1, and set the maximum number of iterations M and the attenuation factor α according to the peak-to-average ratio PAPR and the system bit error rate BER required by the system, and start to iterate, where M =2, α =0.92, M =3, α =0.92, M =2, α =0.95, and M =3, α =0.95 are set respectively in this example.

Step four, determining the slope k and the inflection point v of a second-stage function kx in the linear piecewise companding function and the first-stage function according to the value αSlope of x

4.1) constructing an approximate relation between the slope k and the inflection point v of the second segment function kx according to the characteristic that the average power of the input signal x and the average power of the output signal f (x) of the companding function are equal:

wherein | x | represents the input signal amplitude, | f (x) | represents the output signal amplitude, E [ | x | non-volatile memory²]Representing the average power of the input signal x, E [ | f (x) & gtdoes not luminance²]Represents the average power of the output signal f (x), E [ ·]A desired operator is represented as a result of the operation,is the original OFDM signal x_nThe standard deviation of the (c) signal,is a normalized threshold value;

the above formula is simulated to obtain v_nThe relationship with k is shown in FIG. 3, where v is obtained from FIG. 3_nApproximately linear with k:

according to v_nDefinition of (a) and v above_nAnd k is approximate to a linear relation, and the obtained approximate relation between the slope k and the inflection point v is as follows:

4.2) solving the slope k of the second-stage function kx and the first-stage function according to the definition of the attenuation factor αSlope of xThe solving process is as follows:

wherein,the probability density function of the input signal amplitude | x | of the companding function is represented, and the corresponding relation between α and k is given in table 1;

TABLE 1 list of corresponding values of attenuation factor alpha and parameter k in companding function

Step five: from the original OFDM signal x_nAmplitude | x of_nSize of | for the original OFDM signal x_nPerforming companding conversion to obtain companding conversion signal y_n：

If 0<|x_nIf | is less than or equal to v, then the original OFDM signal x_nMultiplication byObtaining companded transformed signalsn=0,1,…,JN-1；

If x_n|>v, then the original OFDM signal x_nMultiplying by k to obtain companded transform signal y_n=k·x_n,n=0,1,…,JN-1。

Step six: according to the definition of Fast Fourier Transform (FFT), companding the transform signal y_nPerforming JN point FFT to obtain frequency domain signal C_n。

Step seven: from the frequency-domain signal C_nObtaining a transmission signalAnd calculates its peak-to-average ratio PAPR.

7a) Using frequency-domain filter to pair frequency-domain signal C_nFiltering, i.e. the frequency domain signal C_nResponse function H to frequency domain filter_nPerforming dot multiplication to obtain filtered signal

n=0,1,…,JN-1，

Wherein the response function H of the frequency domain filter_nComprises the following steps:

7b) the filtered signal is subjected to IFFTCarrying out IFFT transformation on JN points to obtain transmission signals

7c) Calculating a transmission signal according to a peak-to-average ratio (PAPR) definitionPAPR of (d):

wherein max {. means the operator of taking the maximum value.

Step eight: and setting an attenuation factor alpha according to the PAPR to obtain a transmission signal meeting the PAPR performance requirement of the system.

8a) If M < M, let the number of iterations M = M +1, and set the attenuation factor α according to the value of the peak-to-average ratio PAPR:

if the signal is transmittedWhen the PAPR is more than 5dB, the transmission signal is usedReplacing the original OFDM signal x_nAnd α = α -0.01, and the fourth step is returned;

if the signal is transmittedWhen the PAPR is less than or equal to 5dB, the transmission signal is usedReplacing the original OFDM signal x_nKeeping α unchanged, returning to step four,

8b) if M = M, namely the iteration is ended, the transmission signal obtained in the step sevenNamely, the signal meeting the PAPR performance requirement of the system peak-to-average power ratio is output.

The effect of the present invention can be further illustrated by simulation.

1) Simulation conditions are as follows: in the OFDM modulation, the number of OFDM signal symbols is selected to be 1000, the number of subcarriers is N =1024, a signal constellation is in a QPSK mode, and the modulation system does not perform any other channel coding processing.

2) Simulation content and results:

simulation 1, the original OFDM signal is companded and transformed by using the present invention and the existing clipping filtering method, exponential companding method and conventional linear companding method, and the obtained PAPR suppression performance is shown in fig. 4.

Simulation 2, the original OFDM signal is companded and transformed by using the present invention and the existing clipping filtering method, exponential companding method and conventional linear companding method, and the obtained out-of-band spectrum performance is shown in fig. 5.

Simulation 3, under an additive white gaussian noise channel, the original OFDM signal is companded and transformed by using the present invention and the existing clipping filtering method, exponential companding method and conventional linear companding method, and the obtained BER performance is shown in fig. 6.

Simulation 4, under the rice fading channel, the bit error rate BER performance obtained by performing companding conversion on the original OFDM signal by using the present invention, the existing clipping filtering method, the existing exponential companding method, and the existing linear companding method is shown in fig. 7.

As can be seen from fig. 4, the PAPR of the present invention is significantly better than the conventional linear companding method, but slightly worse than the exponential companding method and the limiting filtering method.

As can be seen from fig. 5, the present invention can obtain a PSD graph of power spectral density almost the same as that of original OFDM, which is significantly better than the exponential companding method and the conventional linear companding method.

As can be seen from fig. 6, under an additive white gaussian noise channel, the BER performance of the present invention is significantly better than that of the exponential companding method and the clipping filtering method, and compared with the conventional linear companding method, the BER performance of the present invention can be obtained as good.

As can be seen from fig. 7, under the rice fading channel, the BER performance of the invention is superior to that of the clipping filtering method, the exponential companding method and the conventional linear companding method.

As can be seen from fig. 4, fig. 5 and fig. 6, under an additive white gaussian noise channel, the present invention can obtain a peak-to-average power ratio (PAPR) performance superior to that of the conventional linear companding method, and a Bit Error Rate (BER) performance superior to that of the exponential companding method and the clipping filtering method, and can obtain a spectrum performance almost the same as that of the original OFDM signal, and the overall performance is superior to that of the clipping filtering method and the existing companding methods.

As can be seen from fig. 4, fig. 5 and fig. 7, under the rice fading channel, the BER performance of the invention is superior to that of the existing methods, and the PAPR performance of the invention is superior to that of the conventional linear companding method and the spectrum performance almost the same as that of the original OFDM signal can be obtained, and the overall performance is superior to that of the existing clipping filtering method, exponential companding method and conventional linear companding method.

Claims (2)

1. A method for suppressing a peak-to-average power ratio of an Orthogonal Frequency Division Multiplexing (OFDM) signal comprises the following steps:

(1) up-sampling OFDM modulation signal to obtain original OFDM signal x_nWhere N is 0,1, …, JN-1, J denotes an upsampling factor, N denotes the number of subcarriers included in OFDM modulation, and JN denotes the number of subcarriers included in the OFDM system after upsampling;

(2) selecting a linear piecewise companding function f (x):

$f\left(x\right)=\left\{\begin{array}{cc}\frac{1}{k}x,& 0<\left|x\right|\&le;v\\ kx,& \left|x\right|>v\end{array}\right.,$

wherein x represents the input signal of a linear function, f (x) represents the output signal of a linear function,as a function of the first segmentK is the slope of the second segment function kx, 0<k<1; v is an inflection point, 0<v is not more than max { | x | }, |, is a modulo operator, satisfies 0<The input signal of the condition that | x | < v is a small signal, and satisfies | x | < v |>The input signal of the v condition is a large signal;

(3) setting the iteration number M to be 1, setting the maximum iteration number M and an attenuation factor alpha according to the peak-to-average power ratio (PAPR) and the system Bit Error Rate (BER) required by the system, and starting an iteration process;

(4) from the value α, the slope k and inflection point v of the second-stage function kx are obtained by the following equations, and the first-stage function kx is obtainedSegment functionSlope of (2)

$\mathrm{\&alpha;}=\frac{1}{k}+\left(k-\frac{1}{k}\right)\left(1+\frac{5}{3}k\right)e^{\left(-\frac{5}{3}k\right)},$

$v=\sqrt{\frac{5}{3}{\mathrm{k\&sigma;}}^2},$

WhereinIs the original OFDM signal x_nIs the standard deviation of (a), mu represents the original OFDM signal x_nAverage value of (d);

(5) applying companding function f (x) to original OFDM signal x_nPerforming companding conversion to obtain companding conversion signal y_n：

$y_n=\{\begin{array}{cc}\frac{1}{k}\&CenterDot;x_n,& 0<\left|x_n\right|\&le;v\\ k\&CenterDot;x_n,& \left|x_n\right|>v\end{array},n=0,1,...,JN-1;$

(6) To companded transform signal y_nFFT to obtain frequency domain signal C_n；

(7) Using frequency domain filter to process FFT frequency domain signal C_nFiltering to obtain filtered signalThen the filtered signal is processedIFFT conversion is carried out to obtain a new transmission signal

(8) Computing a transmission signal from a peak-to-average ratio definitionAnd outputting a transmission signal according to an iteration result:

if m<M, making the iteration number M equal to M +1, and using transmission signalReplacing the original OFDM signal x_nSetting an attenuation factor α according to the current PAPR value, and returning to the step (4) to continue execution;

if M is equal to M, the iteration is ended, and the transmission signal is output

2. The method for peak-to-average power ratio suppression of an orthogonal frequency division multiplexing, OFDM, signal as set forth in claim 1, wherein the frequency domain filter in said step (7) has a response function H_nComprises the following steps:

$H_n=\left\{\begin{array}{cc}1,& 0\&le;n\&le;N\\ 0,& N+1\&le;n\&le;JN-1\end{array}\right.,$

where N is 0,1, …, JN-1, J denotes an upsampling factor, and N denotes the number of subcarriers included in OFDM modulation.