Resource Allocation For Multi-cell OFDMA Systems With Mixed Service

With the rapid development of communication technologies and the raisingrequirements of multimedia service, OFDM has been the key technique in the nextgeneration of wireless communication technology because of its superior performance.However, the existence of inter-cell interference in multi-cell OFDMA systems willreduce users’ communication quality and lead to not guarantee their quality-of-service(QoS). Therefore, it’s necessary to study resource allocation algorithms to meet variousQoS requirements in a multi-cell OFDMA system.This paper first introduced the fading characteristics of wireless channels and basicprinciples of OFDMA systems, and then described several common resource allocationalgorithms in OFDMA systems. However, these algorithms are difficult to make a goodtrade-off among various aspects of performance. Considering these shortcomings, thispaper presented a new type of resource allocation algorithm MSF according to variousQoS requirements. First, we adopt a specific spectral classification scheme to allocatesufficient sub-carriers to different users set; then, in order to achieve fairness, real-timeusers are allocated sub-carriers with compulsory resource reservation. A mathematicLagrange model is formulated with proportional fair scheduling for best-effort users.MSF algorithm considered not only the QoS requirements for multi-service, but also thefairness of overall system. Simulations results showed that MSF algorithm can providebetter fairness performance at the cost of degradation in system throughput, i.e., achievea tradeoff between system throughput and fairness. Moreover, MSF can decrease thepacket drop rate of RT users and meet the required data rates for BE users.