基于伪随机序列处理技术的ZP-OFDM系统  

A Performance Improvement of Zero Padding OFDM System with Pseudorandom Sequence Processing Technique

作　　者：范达[1] 曹志刚[1] 樊平毅[1] 

机构地区：[1]清华大学电子工程系,北京100084

出　　处：《北京邮电大学学报》2005年第6期8-12,共5页Journal of Beijing University of Posts and Telecommunications

基　　金：高等学校博士学科点专项科研基金课题(20030003039)

摘　　要：针对补零正交频分复用(ZP-OFDM)系统中基于快速傅里叶变换(FFT)的接收机性能较差的情形,提出了一种伪随机序列处理技术.该技术将一段伪随机序列的元素分别乘在快速傅里叶逆变换(IFFT)矩阵的各行,则可以改善IFFT矩阵各列之间的相关特性,从而提高OFDM系统的性能.并且由于伪随机序列的元素为±1,所以改进后的IFFT矩阵仍然保持归一性和正交性.利用该技术可以对ZP-OFDM、ZP-OFDM-OLA(overlap add,重叠相加)以及ZP-OFDM-FAST等方法进行改进.仿真部分对改进方法和原方法进行了性能对比,采用Hiperlan2的A、E 2种信道模型,在误比特率为10-4时,改进方法在性能上至少提高了2.2 dB.仿真结果验证了伪随机序列处理技术的有效性.ZP-OFDM（zero padding orthogonal frequency-division multiplexing）has the disadvantage that the simple FFT （fast fourier transform） based receiver does not perform well. Therefore, a pseudorandom sequence processing technique that is each row of the IFFT （inverse fast fourier transform） matrix multiplies with a bit of a pseudorandom sequence is proposed. The technique changes the correlation properties of the IFFT matrix columns, and hence improves the performance over multi-path channels. The modified IFFT matrix is still unitary and orthogonal, since the multiplication by ±1 of the IFFT unitary matrix rows doesn＇t destroy its unitarity and orthogonality. This technique is analized theoretically and applied to ZP-OFDM, ZP-OFDM-OLA and ZP-OFDM-FAST. Simulation resuits verify the improved performance of the modified methods using this technique. The Hiperlan2 Channel models A and E are considered in our simulation. For the two channel models, the modified methods exhibit at least 2.2 dB advantage in SNR at 10^-4 BER （bit error rate）.

关 键 词：正交频分复用 伪随机序列 相关特性 信道状态信息 误比特率 

分 类 号：TN911.7[电子电信—通信与信息系统]