The most promising scheduling algorithm to provide guaranteed QoS to all types of traffic in multiservice 4G wireless network

There is a growing interest of wireless operators to migrate their existing 3G networks to different 4G technologies such as WiMAX and LTE. In this heterogeneous environment of wireless networks and architectures, one of the major concerns is how to allocate network resources efficiently to diverse traffic classes with different QoS constraints. Further it has been convincingly demonstrated through numerous high quality studies that multimedia traffic found in modern wireless networks exhibits Long-Range Dependence (LRD) and self-similarity, a phenomenon which cant' be captured by traditional traffic modeling based on simplistic Poisson model. Unlike most existing studies that are primarily based on simplistic Poisson model and traditional scheduling algorithms, this research presents an analytical performance model for multiple queue systems with self-similar traffic input scheduled by a novel and promising scheduling mechanism. Our proposed model is substantiated on G/M/1 queuing system that considers multiple classes of traffic exhibiting long-range dependence and self-similar characteristics. We analyze the model on the basis of newly proposed scheduling scheme. We present closed form expressions of expected waiting times for multiple traffic classes. We develop a finite queue Markov chain for the proposed scheduling scheme. We develop a discrete event simulator to understand the behavior of multiple classes of self-similar traffic under this newly proposed scheduling mechanism. The results indicate that our proposed scheduling algorithm provides preferential treatment to real-time applications such as voice and video but not to that extent that data applications are starving for bandwidth and outperforms all other scheduling schemes that are available in the market.