Journal of Systems Engineering and Electronics ›› 2010, Vol. 32 ›› Issue (8): 1578-1581.doi: 10.3969/j.issn.1001-506X.2010.08.05

• 电子技术 • 上一篇    下一篇

降低MIMO-OFDM系统峰均比的分解并行选择映射算法

高静,汪晋宽,解志斌   

  1. (东北大学信息科学与工程学院, 辽宁 沈阳 110004)
  • 出版日期:2010-08-13 发布日期:2010-01-03

Decomposed concurrent selected mapping to reduce the PAPR in MIMO-OFDM systems

 GAO Jing, WANG Jin-kuan,  XIE Zhi-bin   

  1. (School of Information Science & Engineering, Northeastern Univ., Shenyang 110004, China)
  • Online:2010-08-13 Published:2010-01-03

摘要:

高峰均功率比(peak-to-average power ratio, PAPR)问题是多输入多输出正交频分复用(multiple-input multiple-output orthogonal frequency division multiplexing, MIMO-OFDM)系统实用化的主要障碍之一,针对这一问题提出了一种分解并行选择映射(decomposed concurrent selected mapping, D-CSLM)算法,进一步提高了算法的峰均比性能。所提算法将每根天线上OFDM 符号分解为实部和虚部,分别采用相同的相位因子,在进行逆离散傅里叶变换(inverse discrete Fourier transform, IDFT)之后进行组合,选择使所有天线具有最小平均峰均比的信号进行传输。与原有的并行选择映射算法相比,所提算法具有更大的待选信号范围,峰均比降低性能更好。同时,利用实序列固有的共轭对称特性,使算法的计算复杂度保持不变。仿真结果证明,该算法在保持计算复杂度不变的前提下,显著改善了系统的峰均比性能。

Abstract:

The high peak-to-average power ratio (PAPR) is a major drawback of multiple input multiple output orthogonal frequency division multiplexing systems (MIMOOFDM). To overcome this problem, a decomposed concurrent selected mapping (D-CSLM) algorithm is proposed. In the proposed algorithm, the candidate OFDM frames transmitted on antennas are decomposed to be real and imaginary parts, then multiplied with the same phase factors  respectively. After an inverse discrete Fourier transform (IDFT), the combined OFDM frames with minimum average PAPR is transmitted. Therefore, the proposed algorithm can provide sufficiently larger number of alternative transmit signals compared with the conventional concurrent selected mapping SLM algorithm, thus abtaining a better PAPR performance. Moreover, the computational complexity of the proposed algorithm is unchanging by using the conjugate symmetry property of real sequence. Simulation results demonstrate that the proposed algorithm can achieve a better PAPR performance with the same numerical complexity compared with the conventional concurrent SLM.