QoS guaranteed radio resource allocation scheme using Genetic Algorithm for OFDMA

Abstract

OFDMA (Orthogonal Frequency-Division Multiple Access) is one of the most important technologies applied in Long Term Evolution (LTE) of wireless radio system [1]. In OFDMA system, dynamic resource block (RB) allocation is utilized as the effective way to achieve multi-user diversity gain, and also the factor which will greatly affect the QoS (Quality of Service) performance of each UE (User Equipment). Along with the development of future wireless communication systems, the QoS requirement has been led to more complex situations and will be affected by various aspects, e.g. network architecture and services with multiple traffics. The traditional resource allocation schemes based on PF (Proportional Fair) or MAXCI (Maximal C/I) perform the greedy searching, and will encounter problems to effectively fulfill the complicated QoS request. In this paper, aiming at various GBR (Guaranteed Bit Rate) requests of each UE, the Genetic Algorithm (GA) based resource allocation algorithm is proposed. The algorithm gains the merit of heuristics, and introduces efficient crossover rules and fitness function to rapidly approach to the optimal resource allocation regarding to different GBR requests of UEs. The system evaluation shows that GA based algorithm owns visibly superior performance at satisfying UEs' QoS request and average system throughput than other contrast schemes.