Optimal peak-to-average power ratio reduction in OFDM systems

Orthogonal-frequency division multiplexing (OFDM) is widely used in wireless communication systems due to its bandwidth efficiency and robustness to multi-path channel effects. The primary disadvantage of OFDM is the high time-domain peak-to-average power ratio (PAR) that severely limits the transmitter power efficiency. This dissertation demonstrates how to optimally distort the data carriers and the free carriers of an OFDM signal to achieve the globally minimum PAR. The proposed method is compatible with existing OFDM standards because no changes are required in the receiver structure or sensitivity.; The optimal use of data carrier distortion is fundamental to this work. Distorting the OFDM constellation significantly reduces the PAR and the power wasted in the free carriers. An optimal amount of distortion exists for minimizing the transmitter power consumption while maintaining a constant bit error rate at the receiver. An optimal amount of distortion also exists for maximizing the data rate while maintaining a constant power consumption at the transmitter.; The OFDM PAR minimization problem has an inherent fast Fourier transform structure that can be exploited to improve the computational efficiency of a convex optimization algorithm. When applied to 802.11a/g wireless local area network signals, the customized algorithm is over 100 times faster than a general-purpose algorithm, and is estimated to require 1.6 mm2 and 20 mW for real-time implementation in a 90 nm CMOS technology. The PAR optimization improves the power efficiency of a typical non-linear power amplifier by 45%--100%, resulting in a power savings of 300--500 mW. The technique is scalable to multiple-input multiple-output (MIMO) OFDM systems with linear channel precoding.