Power and rate allocation for heterogeneous transmission in wireless multiuser OFDM systems

Elektrotechnik Diplomarbeiten (Mai 2009 - Oktober 2009) ofdm

Betreuer

Anke Schmeink, Chunhui Liu,

Abstract

Throughout the thesis, we investigate resource allocation problems for heterogeneous services in multiuser orthogonal frequency division multiplex (OFDM) systems. Given the total transmission power limit, our goal is to maximize the data rate of rate-adaptive services while satisfying the rate requirements for margin adaptive services. Fairness among rate-adaptive services are additionally considered. Towards this end, we firstly go over the necessary background knowledge from wireless communications and optimization theory that is related to our tasks. Then optimization problems are formulated mathematically. Multiuser resource allocation problems have been proved as NP-hard. Therefore we formulate Lagrange dual function to calculate the dual optimum of the optimization problem. Study shows that the duality gap can be neglected, when number of subcarriers is sufficiently large. However, since the complexity of dual method is still too high to apply in practical systems, heuristic methods are proposed to balance the computational efficiency and the performance. In the proposed heuristic method, the resources, composing of subcarriers and transmit power, are adaptively allocated to the users according to their service requirements and channel conditions over two sequential steps. First, fulfill the rate requirements for margin-adaptive and rate-adaptive services with minimized resource usage. Then, allocate remaining resource to the rate-adaptive services to maximize their total data rate. The performance of the proposed resource allocation algorithm is evaluated in a frequency selective fading channel. Numerical results show that the proposed heuristic method has much lower computational complexity with slight performance loss compared to the dual method.