Thesis (M.Sc.) - Cairo University - Faculty of Engineering - Department of Electronics and Communication

In this research, the concept of Machine Learning (ML) is utilized to adaptively provide the scheduler with some information about the User Equipment (UE), such as traffic patterns, demands, quality of service (QoS) requirements, and other network conditions. The proposed adaptive scheduler targets multiple objective scheduling strategies, where the different objectives{u2019} weights are adjusted based on the UEs{u2019}demand pattern to optimize the radio resources allocation per transmission. In addition, it overcomes the trade-off problem of the traditional scheduling methods. This technique can be used as a generic solution with any scheduling strategy. In this thesis, Genetic Algorithm (GA)-based multi-objective scheduler is adopted in order to illustrate the efficiency of the proposed technique. Moreover, the time complexity issue of the GA is addressed. Results show that using the combination of clustering and classification algorithms along with the GA optimizes the GA- based scheduler functionality and reduces its computational complexity by a multiplier factor

Issued also as CD