Models for quality of service in heterogenoeous networks

Abstract: Both streaming techniques and wireless heterogeneous networks have recently become more widely deployed. As the current streaming techniques are primarily designed for homogenous wired networks, streaming multimedia applications in heterogeneous networks can perform poorly due to wireless networks conditions and vertical handover. These problems can significantly degrade the performance of streaming multimedia applications. Effects of the degradation are delay, jitter and packet loss resulting in lower multimedia quality. The use of some congestion recovery algorithms are detrimental factors to QoS. One approach used to control congestion in the network layer is Active Queue Management (AQM). This dissertation presents an evaluation and comparison of AQM mechanisms in heterogeneous networks. Based of the results of this research a new AQM algorithm, named the Adaptive AQM (AAQM) is proposed. The AAQM uses control law and link utilization in order to manage congestion. The action of the control law is to mark incoming packets in order to maintain the quotient between arriving and departing packets as low as possible. The AAQM enhances performance with respect to the queue length and packet loss as well as buffer space requirements. AAQM outperforms the other AQM algorithms in terms of multimedia packet loss and buffer delay.