A Research On The Method For Reducing The Nonlinear Distortion Of OFDM Signal

Orthogonal frequency division multiplexing(OFDM) is provided with the advantages on data transmission efficiency and the ability against frequency selective fading. But the OFDM signal has high peak to average power ratio(PAPR), on one hand,it requires that the high power amplifier(HPA) has a large dynamic range, which results in poor efficiency of HPA; on the other hand, the OFDM signal will be distorted when it passes through HPA, thus increase the bit error rate(BER).At present, the methods for reducing the nonlinear distortion can be divided into two categories, namely, distortion restraint and distortion compensation. Among them,the distortion restraint technologies reduce the PAPR by signal preprocessing, thus reducing the distortion caused by HPA. The methods for PAPR reduction include probabilistic, clipping and coding. In the first category, both partial transmit sequence(PTS) and selected mapping(SLM) can reduce PAPR efficiently without additional distortion. However, the conventional SLM and PTS method have two obvious shortcomings. Firstly, the transmitter has high computational complexity. Secondly, the additional side information(SI) need to be transmitted to recover the original signal,resulting in the loss of data transmission efficiency. The distortion compensation technologies reduce the nonlinear distortion by estimating and adding the compensation signal, and both the compressive sensing(CS) algorithm and compensation algorithm based inverse model are used commonly.In this paper, for one thing, the low computational complexity algorithms can be proposed to reduce the PAPR and the nonlinear distortion caused by HPA; for another thing, the reasonable compensation algorithms can be designed to reduce the distortion and improve the BER performance. The main work of this paper is organized as follows:(1) A low complexity PTS algorithm based on cyclic shifting partial time-domain signal will be proposed in the first section. Utilizing the characteristic of inverse fast fourier transform(IFFT), the transmitter needs only one IFFT operation to acquire alternative sequences, and the receiver can restore the cyclic shift factor for blind detection by comparing the distance between the inverse rotation sequences and its nearest constellation points.(2) A low complexity partitioning SLM algorithm based on blind detection will be present in the second section. The transmitter only needs a few low-dimensional IFFT operations to obtain a lot of alternative sequences, at the same time, the receiver uses a modified blind detection with low complexity to achieve the restoration of the signal.(3) A compensation algorithm based on clipping and compressive sensing will be given in the third section. The transmitter can reduce the PAPR by clipping. The receiver uses a modified compensation method based on the inverse model of HPA to reduce the nonlinear distortion caused by the HPA, and then uses compressive sensing to counteract the distortion caused by clipping.(4) A low complexity PTS algorithm for SCMA system will be proposed in the last section. According to the characteristic of the SCMA codebooks, the transmitter will use a few low-dimensional IFFTs and sub-signal combination to generate alternative sequences. The receiver will apply the rotated codebooks and massage pass algorithm(MPA) to recovering the original signal.