PAPR Analysis of Class-III SLM Scheme Based on Variance of Correlation of Alternative OFDM Signal Sequences

• Published in: IEEE communications letters Vol. 19; no. 6; pp. 989 - 992
• Main Authors: Woo, Jun-Young, Joo, Hyun Seung, Kim, Kee-Hoon, No, Jong-Seon, Shin, Dong-Joon
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.06.2015; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Class-III selected mapping (SLM); Correlation; Fast Fourier transforms; Memory management; Modulation; orthogonal frequency division multiplexing (OFDM); Peak to average power ratio; peak-toaverage power ratio (PAPR); Simulation; orthogonal frequency division multiplexing (OFDM); peak-to-average power ratio (PAPR); correlation; Class-III selected mapping (SLM)
• ISSN: 1089-7798; 1558-2558
• DOI: 10.1109/LCOMM.2015.2422700

Selected mapping (SLM) is a well-known peak-to-average power ratio (PAPR) reduction technique for orthogonal frequency division multiplexing (OFDM) systems. Recently, a low-complexity SLM scheme, called Class-III SLM scheme, was proposed, which performs only one inverse fast Fourier transform (IFFT) to generate alternative OFDM signal sequences. By randomly selecting the cyclic shift and rotation values, Class-III SLM scheme can generate up to N 3 alternative OFDM signal sequences, where N is the IFFT size. However, all N 3 alternative OFDM signal sequences do not achieve good PAPR reduction performances. Therefore, an efficient selection method of good rotation and cyclic shift values is needed, which results in good PAPR reduction performance. In this letter, a selection method of cyclic shift values is proposed, which is optimal in terms of minimizing the variance of correlation values between alternative OFDM signal sequences. It is also shown that rotation values are useless when U ≤ N/8, where U is the number of alternative OFDM signal sequences. Also, a selection method of proper rotation values when U > N/8 is proposed. Simulation results show that the proposed method achieves the optimal PAPR reduction performance. In addition, the proposed scheme requires less memory and side information than random scheme.