Improved stability of QoS provisioning for 3G systems and beyond: optimum and automatic strategy selection for packet schedulers

Wireless all-IP multimedia communications systems promise a wide range of services with different quality of service (QoS) requirements and diverse ranges of supported bit rates. This implies that the nature of input traffic load at the base station is expected to vary significantly in time. When radio channels are shared by the existing mixed services (say for example real-time and non-real time services), packet scheduler and channel assignment faces a number of parallel input data pipes which are varied in terms of bit rate. The weights, assigned to different involved aspects in the packet scheduling process, outlines the strategy in terms of importance given to different aspects of packet scheduling. Schedulers with fixed weighting parameters are unable to cope with variable traffic load and radio channel conditions efficiently. The weights assigned to specific traffic or radio channel conditions, might not be suitable for others. In this paper, we propose an optimization and automation algorithm for selection and adaptation of packet scheduling weights. It is shown that the proposed algorithm is capable of improving different aspects involved in a successful QoS provisioning, such as fairness, stability, delivered delay, bit rate and throughput, simultaneously.