Title :
A convex interior-point method for optimal OFDM PAR reduction
Author :
Aggarwal, Alok ; Meng, Teresa H.
Author_Institution :
Dept. of Electr. Eng., Stanford Univ., CA, USA
Abstract :
The main disadvantage of OFDM is the high time-domain peak-to-average power ratio (PAR) that limits transmitter power efficiency. This paper presents a convex optimization algorithm for minimizing the PAR of an OFDM signal subject to constraints on the constellation error vector magnitude (EVM). The derivation reduces computational complexity by exploiting known features of the optimization problem, such as OFDM´s FFT structure. Simulation results are given for the 802.11 a/g WLAN standard. The customized algorithm is approximately 100 times faster than a general-purpose optimizer. The algorithm achieves PAR within 1 dB of the globally optimal solution after two iterations, where the main complexity per iteration is four FFTs plus the solution of a linear system of equations.
Keywords :
IEEE standards; OFDM modulation; convex programming; fast Fourier transforms; iterative methods; linear systems; minimisation; signal processing; telecommunication standards; wireless LAN; 802.11 a-g WLAN standard; EVM; FFT; OFDM signal; PAR reduction; constellation error vector magnitude; convex interior-point method; fast Fourier transform; iteration method; linear system equation; minimization; optimization algorithm; orthogonal frequency division multiplexing; peak-to-average power ratio; wireless local area network; Computational complexity; Computational modeling; Constellation diagram; Constraint optimization; Flexible printed circuits; OFDM; Peak to average power ratio; Time domain analysis; Transmitters; Wireless LAN;
Conference_Titel :
Communications, 2005. ICC 2005. 2005 IEEE International Conference on
Print_ISBN :
0-7803-8938-7
DOI :
10.1109/ICC.2005.1494686