Open Collections

PAR reduction in OFDM systems by multiple signal representation

University of British Columbia

As we are embarking on the new information age, demand for high speed wireless communication has increased substantially. One of the promising technologies that could fulfil such demands is orthogonal frequency division multiplexing (OFDM). Due to its bandwidth efficiency and robustness in wireless environments, OFDM has recently gained an increasing popularity. One major drawback of OFDM is its high peak-to-average power ratio (PAR). If not processed, the high peaks in the OFDM signal cause saturation in the power amplifier, which in turn distorts and decreases the power of the transmit signal. In order to avoid the resulting bit-error-rate performance degradation and out-of-band radiation, either expensive linear power amplifiers need to be employed, or nonlinear amplifiers must be operated with high power backoffs in power-inefficient amplification. Power inefficiency is however unacceptable to battery-powered devices. Among the numerous PAR reduction approaches available in the literature, multiple signal representation (MSR) techniques have achieved considerable attention thanks to their great PAR reduction capability and favorable performance-expense tradeoffs. Within the MSR family, partial transmit sequences (PTS) is the most popular while trellis shaping is the newest technique. This thesis makes contributions to both of them. For PTS, we restate its low-PAR signal search as a combinatorial optimization (CO) problem and propose a fair benchmark to evaluate the search algorithms. These allow us to compare the existing PTS algorithms and apply efficient heuristics from the CO literature to PTS. Surprisingly, our results show that random search, the simplest algorithm, actually outperforms all available PTS algorithms. Only our tailored versions of the two CO metaheuristics, namely simulated annealing and tabu search, attain better tradeoffs in the low-PAR region. Trellis shaping is an interesting MSR technique that does not require explicit side information. Our contributions to trellis shaping include a new metric and the use of the stack sequential decoding algorithm in shaping-sequence search. These proposals improve PAR reduction and/or reduce the complexity of the technique. Beneficial to all algorithms that involve a large number of peak-power searches, we propose using a new optimization objective to reduce their overall complexity. This is applicable to both PTS and trellis shaping. Finally, based on the trellis shaping structure, we propose a side information embedding/ detecting scheme for PTS. Our low-complexity scheme does not affect the PAR reduction capability of PTS and practically does not degrade the BER performance of the OFDM systems.

Text

2010-01-08

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

http://hdl.handle.net/2429/17797

Electrical and Computer Engineering

University of British Columbia

UBCV

DSpace