Energy-aware resource allocation based on margin adaptive principle for OFDMA systems

The problem of dynamic resource allocation in OFDMA system has been concerned widely. An energy-aware resource allocation algorithm based on margin adaptive (MA) principle is developed to further improve the performance of system. In order to balance the fairness among users and total throughput of system, the number of required subcarriers of each user is assigned in advance according to their service priorities. Hence, the optimization problem is converted into another non-deterministic polynomial (NP) problem that is to minimize total transmission power while the number of required subcarriers for each user and target BER are given. Specifically, transiently chaotic neural network (TCNN) is applied to jointly allocate subcarriers, bits and power. Simulation results show that the optimal solution is found successfully through its rich neurodynamics (flexible chaos). Compared with the existing algorithms, the proposed algorithm achieves better BER performance by adaptive selection policy of modulation mode and higher data rata by exploiting multiuser diversity gain fully.