Adaptive Quality Of Service Call Admission Control With User Mobility Prediction For Multimedia Traffic Over Wireless Networks

Multimedia traffic is expected to be supported in the next generation wireless networks. As in wireline networks, the wireless network must also be capable of providing guaranteed quality of service (QoS) over the lifetime of mobile connections. Some challenging problems that appear in multimedia wireless networks, such as user mobility and shortage of bandwidth, influence the QoS provisioning for the users. In this thesis, we propose a new framework called Adaptive quality of service (AdQoS) to guarantee the QoS of multimedia traffic. The objectives that AdQoS framework tries to accomplish are minimum new call blocking and handoff dropping rates. The key feature of this framework is the bandwidth reallocation scheme. This scheme is developed to control the bandwidth operation of ongoing connections when the system is overloaded. The other key feature is the bandwidth reservation scheme incorporating a user mobility prediction to manage the QoS of the networks. Based on the mobility prediction, bandwidth is reserved to guarantee the uninterrupted hand off process. A comparison between existing user mobility prediction and the proposed scheme is also presented. An integrated system, which combines the Bandwidth Allocation Level technique and the user mobility prediction, is also proposed. The proposed user mobility prediction algorithm integrates the Received Signal Strength (RSS) measurements for the mobile terminal's intra-cell movement and aggregate history of mobile terminals for inter-cell movement. When compared with the conventional scheme proposed in the literature, the simulation results show that our proposed scheme reduces the new call blocking probabilities, the handoff dropping probabilities and reduces significantly the probability of terminating calls while still maintaining efficient bandwidth usage