Table-based approach to sub-optimum peak reduction carriers

Abstract

This proposes the use of table-based approach to ease the processing burden for a genetic algorithm (GA) based search for suboptimum peak reduction carriers (PRCs). The conventional search using the same GA is required to perform a significant number of IFFT operations for PAPR evaluation at each GA iteration. This gives rise to a significant processing overhead and only a fast DSP can perform the search without giving noticeable delays in transmission. The table-based approach substitutes the computational complexity of repeated IFFT evaluation by complex additions. Time domain PRC generator patterns are stored in a table and a movable window is used to produce the various phases required of a PRC. The IFFT only needs to be performed once per symbol on the user data to yield the user data symbol and the final OFDM signal is produced by summing all the PRC signals with the data symbol in the time domain. By using the proposed approach, a better PAPR can be achieved using the same amount of processing since more GA iterations can be performed with the same overall processing delay. Alternatively the technique permits a cheaper and lower specification DSP to be used.