Date of Award


Degree Name

Doctor of Philosophy


Computer Science

First Advisor

Dr. Ala Al-Fuqaha

Second Advisor

Dr. Dionysios Kountanis

Third Advisor

Dr. Ammar Rayes

Fourth Advisor

Dr. Driss Benhaddou


Quality of Service (QoS), extreme value theorem, client-side monitoring, network management


Providing an efficient Quality-of-Service (QoS) measurement model is a challenging problem in today’s mobile computing and telecommunications networks. Currently, most of QoS techniques utilize service measurements that are collected by the network elements (i.e., network-side monitoring) to evaluate the network performance. However, this process does not take into account the service performance from the clients' perspective and might contradict with the Service Level Agreement (SLA). In order to overcome the limitations of service-side QoS monitoring, a number of research studies have been conducted to present alternative architectures and algorithms for client-side QoS service assessment in computer networks. The client-side QoS approach gives the major role to the clients to evaluate the dedicated services through gathering the network measurements and reporting the necessary information. The service providers consider clients’ feedback to revise and resolve service performance issues. This model can be considered as a cooperative approach that provides a compromised service plan for both the service providers and the network clients to achieve a better network service performance.

In this research, we study the tradeoffs between network-side and client-side QoS monitoring and present a client-based architecture for the evaluation and prediction of service degradations in mobile networks. The client-side approach should be capable of utilizing multi-level service performance analysis in a scalable mobile network environment. The service performance analysis consists of three levels: service monitoring and evaluation, verifying and enhancing the network measurement accuracy (the collected performance data), and performance prediction in single and multi-hops networks. In addition, this approach should support short and long-term service evaluation scenarios. The short-term scenario gives the ability to service providers to react in a timely manner to the clients' feedback to preserve a certain level of service performance. The long-term scenario helps to clarify the service behavior by predicting the service degradation over the monitoring and evaluation sessions. Furthermore, this scenario allows the service providers to refine the degraded services and maintain the SLA.

Access Setting

Dissertation-Open Access