CN103441769B - PTS method for reducing PAPR of OFDM system - Google Patents

Abstract

The invention belongs to the technical field of wireless communication, and particularly relates to a PTS method for reducing the PAPR of an OFDM system. After an OFDA signal is partitioned, autocorrelation with different autocorrelation coefficients is respectively added according to the plus or minus characteristic of a first data point real part of each piece, so that the aims that the PAPR of the OFDM signal is reduced and a receiving end detects the signal without sideband auxiliary information are achieved. The receiving end recovers a correlation coefficient according to the plus and minus characteristic of an autocorrelation value of each piece, a piece signal without autocorrelation and the autocorrelation coefficient corresponding to the piece signal are multiplied, and the phase rotation sequence timing the piece is judged according to the plus and minus characteristic of a product. The receiving end can recover an original signal without the sideband auxiliary information while the PAPR of the OFDM signal is effectively reduced through the self-correlation matching-PTS, so that band width is saved, and frequency spectrum utilization efficiency is effectively improved.

Description

A PTS method for reducing PAPR of OFDM system

technical field

The invention belongs to the technical field of wireless communication, and in particular relates to a PTS method for reducing PAPR of an OFDM system.

Background technique

OFDM (Orthogonal Frequency Division Multiplexing, OFDM) is a multi-carrier modulation technology with superior performance, which has been widely valued because it can effectively combat inter-symbol interference (Inter-symbol Interference, ISI) and multipath delay. There are multiple orthogonal sub-carriers in the OFDM system, and its output signal is the superposition of multiple sub-signals, so there will be a high peak-to-average power ratio (PAPR). High PAPR will cause out-of-band power radiation, hardware nonlinear distortion, and mutual interference between subcarriers, etc., which will seriously deteriorate system performance. Therefore, it is of great significance to reduce the PAPR of the OFDM system.

At present, the methods for reducing the PAPR of OFDM systems can be roughly divided into three categories: signal pre-distortion technology, coding methods and probability methods. Probabilistic methods are the focus of current research. The probabilistic method does not really reduce the peak power of the OFDM signal, but achieves the purpose of reducing the occurrence probability of high peak power by implementing linear segmentation or linear transformation on the original OFDM signal. Probabilistic methods include many algorithms, such as subcarrier reservation (Tone Reservation, TR), selective mapping (Selective Mapping, SLM), partial transmission sequence (Partial Transmit Sequence, PTS), etc. In order to restore the original signal at the receiving end, the PTS technology needs to transmit certain sideband side information, which reduces the spectrum utilization efficiency of the system. In order to accurately recover the sideband side information at the receiving end, additional techniques such as coding techniques are usually used to process the sideband side information, which may cause the system PAPR to increase again, and further reduce the system spectrum utilization rate. The autocorrelation matching-PTS method proposed in this paper not only effectively reduces the PAPR of the system, but also embeds the information of the rotational phase factor into the autocorrelation of the signal, and can realize the recovery of the signal without the transmission of sideband information, which greatly improves the Spectrum utilization efficiency.

Contents of the invention

The purpose of the present invention is to provide a PTS method for reducing the PAPR of the OFDM system, implanting the discriminant information of the rotating phase sequence into the autocorrelation of the signal, and recovering the original signal without transmitting sideband information, thereby improving the frequency spectrum usage efficiency.

The purpose of the present invention is achieved through the following technical solutions:

S1. Let the frequency-domain signal after serial-to-parallel conversion and baseband modulation be X, divide it into V sub-blocks X ^(m) according to the traditional PTS division rules, where m=1,2,...,V, and then All sub-blocks judge the correlation coefficient G(z) through the autocorrelation coefficient discriminator;

S2, passing the autocorrelation coefficient G(z) identified by S1 and all the subblocks described in S1 through the autocorrelation signal generator to obtain the autocorrelation frequency domain subblock signal;

S3. Perform IFFT transformation on the autocorrelation frequency domain sub-block signal obtained in S2, transform the frequency domain autocorrelation sub-block signal into the time domain, and multiply all sub-block signals after the transformation by the phase rotation factors b(m), b(m) )∈(+1,-1), m=1,2,…,V, assuming that the sub-block signal in frequency domain and time domain is Then combine b(m) with Multiply to get the alternative transmission sequence

S4. Calculate each candidate transmission sequence according to the PAPR calculation formula PAPR, and choose the smallest signal among which the smallest PAPR for transmission;

S5. Suppose the signal received at the receiving end is y, Perform FFT operation on y and transform it into a frequency domain signal;

S6. According to the block method described in S1, y is divided into blocks, divided into V sub-blocks, each sub-block is expressed as Y ^(m) , m=1, 2, ..., V, Perform an autocorrelation operation on Y ^(m) , restore the autocorrelation coefficient and the rotation phase multiplied by each sub-block, and obtain the signal after eliminating the autocorrelation 1≤i≤V, the autocorrelation operation rule is, if the autocorrelation value of a sub-block is positive, then the autocorrelation coefficient of the sub-block is restored to k (k>0); if its autocorrelation value is negative, Then the autocorrelation value of the sub-block is restored to -k (k>0);

S7, use the income from S6 with the The corresponding rotation phase 1≤i≤V multiplied to obtain the final restored sub-block signals, adding the final restored sub-block signals to obtain the final restored data block Y ^* ,

S8. Perform baseband demodulation and parallel-to-serial conversion on Y ^* obtained in S7 to obtain a finally recovered signal.

Further, the number of candidate transmission sequences in S3 is n=2 ^V-1 .

The beneficial effects of the present invention are: after the OFDM signal is divided into blocks, the autocorrelations with different autocorrelation coefficients are added according to the positive and negative of the real part of the first data point of each sub-block, so as to reduce the OFDM signal PAPR and receive The end does not need sideband side information for detection signal purposes. At the receiving end, the autocorrelation coefficient is restored according to the sign of the autocorrelation value of each sub-block, and the sub-block signal after the autocorrelation is eliminated is multiplied by the corresponding autocorrelation coefficient, and the sub-block is judged according to the sign of the product Phase rotation sequence to multiply by. Autocorrelation matching-PTS can effectively reduce the PAPR of the OFDM signal, and at the same time, the original signal can be recovered at the receiving end without sideband side information, which saves bandwidth and effectively improves the efficiency of spectrum utilization.

Description of drawings

Fig. 1 is a system block diagram of PTS-OFDM.

Figure 2 is a block diagram of the autocorrelation matching-PTS system.

Detailed ways

The specific embodiment of the present invention is described below in conjunction with accompanying drawing:

A typical OFDM signal can be represented as follows:

Among them, N is the number of subcarriers in the OFDM system, X(n) is the baseband modulated frequency domain data transmitted on the nth subcarrier, f _n is the frequency of the nth subcarrier, and T is the period of an OFDM symbol .

The PAPR of a signal is defined as the ratio of the instantaneous peak power to the average power of the signal:

$\mathrm{<span\; class="notranslate">\; PAPR</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; max</span>}<span\; class="notranslate">\; \{</span>{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \}</span>}{\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; \{</span>{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \}</span>}$

When expressed in dB, rewrite as follows:

Among them, |x(t)| ² represents the instantaneous power of the signal, max[·] represents the maximum value, and E[·] represents the average value.

It is customary to use the complementary cumulative distribution function (Complementary Cumulative Distribution Function, CCDF) to measure the PAPR distribution in the OFDM system, that is, the sum of the probability of occurrence of the part of the sampling values greater than the threshold in all PAPR sampling values: P{PAPR>z}= 1-P{PAPR≤z}=1-(1-e ^-z ) ^N , where z represents a threshold of PAPR. It can be seen that the larger the number N of subcarriers is, the larger P(PAPR>z) will be.

The traditional PTS technology is shown in Fig. 1, N sub-carriers in one OFDM data are divided into V sub-blocks X ^(m) , m=1,2,...,V. The division rules are as follows; first, the vector length of each sub-block is equal; second, for each sub-block, the position of the original OFDM symbol information that is not divided into this block is supplemented with 0. Each sub-block contains data of N/V different frequency points in X.

In the traditional PTS method, X is expressed as a vector in the form of X=[X(0),X(1),…,X(N-1)], and X can be divided according to three division rules: random, adjacent and interleaved. Divide into V sub-blocks, express each sub-block as X ^(m) , m=1,2,...,V, then X can be expressed as: Then each sub-block is multiplied by an appropriate phase rotation factor, and a set of phase factors that make the transmitted signal have the minimum peak-to-average power ratio is selected Wherein, b(m), m=1, 2,..., V are phase rotation factors, b(m)=exp(jφ(m))m=1, 2,...,V. In particular, when the rotation phase factor is limited to b(m)∈(+1,-1), m=1,2,...,V, U=2 ^V-1 alternative transmission sequences can be obtained, and the transmission The bits of the sideband side information are log ₂ U = log ₂ 2 ^V-1 = V-1.

The time domain signal is obtained from the frequency domain signal through Inverse Fast Fourier Transform (IFFT): $\hat{x}=\mathrm{IFFT}\left(\underset{m=1}{\overset{v}{\mathrm{\&Sigma;}}}b\left(m\right)x^{\left(m\right)}\right)=\underset{m=1}{\overset{v}{\mathrm{\&Sigma;}}}b\left(m\right)*\mathrm{IFFT}\left(x^{\left(m\right)}\right)=\underset{m=1}{\overset{v}{\mathrm{\&Sigma;}}}b\left(m\right)x^{\left(m\right)}$ according to $\mathrm{PAPR}\left(\mathrm{dB}\right)=10\log 10\frac{\max \left\{{\left|x\left(t\right)\right|}^2\right\}}{\mathrm{E.}\left\{\right|{x\left(t\right)|}^2\}}$ The PAPR of each candidate signal is calculated separately, and the signal with the smallest PAPR is selected for transmission, and the sideband side information identifying the phase rotation sequence is transmitted at the same time.

The system block diagram of the autocorrelation matching-PTS method proposed by the present invention is shown in FIG. 2 . Assuming that the frequency domain signal after serial-to-parallel conversion and baseband modulation is X, it is divided into V sub-blocks according to the traditional PTS division rule, X ^(m) , m=1,2,...,V. All sub-blocks are then passed through an autocorrelation coefficient discriminator. The autocorrelation coefficient discriminator determines the autocorrelation coefficient of the sub-block according to the positive or negative of the real part of the first data point in each sub-block, if the real part is positive, then the autocorrelation coefficient of the sub-block is determined as α=k, (k>0), otherwise, the autocorrelation coefficient of the sub-block is judged as α=-k, (k>0).

The transfer function of the autocorrelation signal generator used among the present invention is: G(z)=1+α ^* z ^-1 , wherein, z ^-1 is the backward transfer operation, as z ^-1 s(t)=s( t-1). All the sub-blocks and the correlation coefficients judged by the auto-correlation coefficient discriminator pass through the auto-correlation signal generator to obtain auto-correlation frequency-domain sub-block signals. Then we use the traditional PTS method to process the autocorrelated frequency-domain sub-block signal in the subsequent processing after block. First, the frequency-domain autocorrelation sub-block signals are transformed into the time domain by IFFT transformation, and then each sub-block is multiplied by the phase rotation factor. At this time, we limit b(m)∈(+1,-1), m=1,2,...,V, so according to the analysis of the traditional PTS method, we can get U=2 ^V-1 alternative transmission signals . in accordance with |x(t)| ² represents the instantaneous power of the signal, max[ ] represents the maximum value, E[ ] represents the average value, and the PAPR of the signal is defined as the ratio of the instantaneous peak power to the average power of the signal:

$\mathrm{<span\; class="notranslate">\; PAPR</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; max</span>}<span\; class="notranslate">\; \{</span>{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \}</span>}{\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; \{</span>{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \}</span>}$

When expressed in dB, rewrite as follows:

$\mathrm{<span\; class="notranslate">\; PAPR</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; dB</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 10</span>}\mathrm{<span\; class="notranslate">\; log</span>}\mathrm{<span\; class="notranslate">\; 10</span>}\frac{\mathrm{<span\; class="notranslate">\; max</span>}<span\; class="notranslate">\; \{</span>{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \}</span>}{\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; \{</span><span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \}</span>}$

Calculate the PAPR of each candidate transmission signal separately, and select the signal with the smallest PAPR for transmission. We have embedded the information of the rotating phase sequence into the autocorrelation of the signal implicitly, so there is no need to transmit sideband side information, which saves bandwidth and improves spectrum utilization efficiency.

Suppose the signal received at the receiver is y, which can be expressed as After the received signal undergoes FFT operation, it is transformed into a frequency domain signal.

Divide the received signal into V sub-blocks according to the same block method as that at the sending end. Each sub-block is expressed as Y ^(m) , m=1, 2, ..., V, then Y can be expressed as: Then, each sub-block is passed through an auto-correlation signal matcher, and an auto-correlation operation is performed on each sub-block signal, so as to restore the auto-correlation coefficient and the rotation phase multiplied by each sub-block. The restoration rules of the autocorrelation coefficient are as follows: if the autocorrelation value of a sub-block is positive, the autocorrelation coefficient of the sub-block is restored to k (k>0); if the autocorrelation value is negative, the autocorrelation coefficient of the sub-block is The correlation value is restored to -k (k>0). Then the recovered sub-block signal Y _i , 1≤i≤V and the recovered corresponding autocorrelation coefficient 1≤i≤V and through the autocorrelation signal inverse transformation generator, the signal after eliminating the autocorrelation can be obtained 1≤i≤V.

Immediately afterwards, each sub-block signal after eliminating the autocorrelation 1≤i≤V and its corresponding autocorrelation coefficient 1≤i≤V while passing through the rotating phase discriminator. The rotation phase discrimination rule corresponding to each sub-block is as follows: the autocorrelation coefficient of each sub-block 1≤i≤V and the sub-block signal Multiply the real part of the first data point in 1≤i≤V, if the product is greater than 0, the rotation phase sequence corresponding to the sub-block 1≤i≤V is +1; otherwise 1≤i≤V is -1.

Afterwards, each sub-block signal with autocorrelation eliminated 1≤i≤V and its corresponding rotation phase 1≤i≤V multiplied to obtain the final restored sub-block signals, and then add the sub-block signals to obtain the final restored data block Y ^* , namely After baseband demodulation and parallel-to-serial conversion, the final recovered signal can be obtained.

Claims (2)

1. reduce a PTS method of the PAPR of ofdm system, it is characterized in that: its step is as described below:

S1, set the frequency-region signal after serial to parallel conversion and baseband modulation as X, traditionally the division rule of PTS is divided into V sub-block X ^(m), wherein, m=1,2 ..., V, then passes through auto-correlation coefficient arbiter by all sub-blocks, judges coefficient correlation G (z);

S2, auto-correlation coefficient G (z) determined by S1 pass through autocorrelation signal generator together with sub-blocks all described in S1, obtain auto-correlation frequency domain sub-block signal;

S3, IFFT conversion is carried out to S2 gained auto-correlation frequency domain sub-block signal, frequency domain auto-correlation sub-block signal is transformed to time domain, all sub-block signals after conversion are multiplied by phase rotation coefficient b (m) respectively, b (m) ∈ (+1 ,-1), m=1,2,, V, supposes that time domain frequency domain sub-block signal is then by b (m) with be multiplied and can obtain alternate transmission sequence

S4, calculate each alternate transmission sequence according to PAPR computing formula pAPR, and the signal selecting wherein minimum PAPR minimum for transmission;

S5, suppose that the signal received at receiving terminal is y, fFT operation is carried out to y, is transformed into frequency-region signal;

S6, carry out piecemeal according to the division rule of conventional P TS described in S1 to y, be divided into V sub-block, each sub-block is expressed as Y ^(m), m=1,2 ..., V, to Y ^(m)carry out auto-correlation computation, recover the rotatable phase that auto-correlation coefficient is taken advantage of with each sub-block, the signal after the autocorrelation that is eliminated described auto-correlation computation rule is, if the autocorrelation value of a sub-block is just, then the auto-correlation coefficient of this sub-block reverts to k (k > 0); If its autocorrelation value is negative, then the autocorrelation value of this sub-block reverts to-k (k > 0);

S7, by S6 gained with this corresponding rotatable phase be multiplied, obtain the final each sub-block signal recovered, by each sub-block signal plus of described final recovery, obtain the final data block Y recovered ^*,

S8, to S7 gained Y ^*carry out base band demodulating and parallel serial conversion, the signal finally recovered.

2. a kind of PTS method reducing the PAPR of ofdm system according to claim 1, is characterized in that: described in S3, the number of alternate transmission sequence is n=2 ^v-1.