Advanced Power Allocation in OFDM Systems

I n this project, we propose a power allocation strategy by utilizing the duality gap in OFDM systems for the maximization of the sum of data rates at all subcarriers. In order to achieve the maximum capacity, the proposed power allocation strategy iteratively tunes the amount of power allocated to subcarriers according to the related parameters in the duality gap. Other suboptimal strategies for other objectives are developed including the equal gain allocation strategy, the approximate minimum bit-error rate power allocation strategy, and the minimum mean-square error power allocation strategy. These strategies aim to maximize the average signal to noise ratio (SNR) or to minimize the average bit error rate (BER). Recently, threshold-based power allocation algorithms are designed by considering the SNR thresholds. It reveals that the threshold-based algorithms offer a certain improvement in the average spectral efficiency. The target SNR should be pre-determined and also affects the performances. Without the help of sorting operations and without performing the log operation, the proposed strategy allocates power to subcarriers and has low complexity of ( ? ) . The simulation results show that the performance of the proposed power allocation strategy is very close to that of the optimum solution after a few iterations .