QoS-Aware Discrete Bit Loading for OFDMA Networks

Performance optimization of practical communication networks should be performed by considering some practical constraints such as limitation of maximum transmitted power, discrete bit allocation to carriers and users, and quality of service (QoS) requirements for different users. In this paper, we provide the optimum resource allocation solution for an OFDMA downlink in which subcarriers are loaded with discrete number of bits. Using minimum data rate with a certain BER as the criterion for the QoS, the optimization makes sure that the QoS of all users are satisfied. We analyze the complexity of this solution and propose a suboptimum solution with much less complexity. These solutions are compared with existing methods using simulation results.