Please use this identifier to cite or link to this item:
|Title:||Quality of Service Aware Traffic Scheduling Algorithm for Heterogeneous Wireless Networks|
|Abstract:||Recent developments in multiple technologies evolving these days provide users with broadband services of high quality and seamless mobility. Wireless networks comprising of third generation (3G) and fourth generation (4G) provide wide coverage and good mobility capabilities. All these high quality wireless networks have been deployed within one region overlapping each other, hence generating a heterogeneous network for wireless access which can be represented as 4G heterogeneous network. The architecture of the system is such that it should support bulk data transfer. Although the Worldwide Interpretability for Microwave Access (WiMax) supports huge data transfer but still there is problem with the node request and service management. It becomes a sophisticated task for the nodes to handle number of requests at the same time. This work focuses on different aspects of the WiMax network and scheduling algorithms with hybrid architecture which is a combination of the First Come First Serve (FCFS) and Round Robin (RR) Algorithm. An analytical study of the number of jobs scheduled and completed along with accuracy for FCFS, RR and Hybrid algorithm is represented. The work is preceded by designing architecture of Long Term Evolution (LTE) which is a recently evolving technology which ensures the reliable delivery of the heterogeneous traffic services with high speed data rate and lower delays through their mobile and other hand held devices. The key feature of LTE is its traffic engineering which is used for effectively managing the network resources for efficient utilization. When LTE expertise has arisen there are some prevailing problems to be taken care of i.e. load balancing and traffic scheduling. Even LTE is the debauched technology, but it is also been anguishing from these problems. The work represents a Load Balancing Strategy that has been espoused based on traffic scheduling. Various Traffic scheduling algorithms are designed earlier to assign shared resources among users to optimize the performance of LTE systems in an efficient manner. The performance of three types of scheduling algorithms is compared in this work namely FCFS, RR and Hybrid algorithm which is the combination of FCFS and RR in extremes conditions. The discussed scheduling algorithms performance is measured in terms of Symbol error rate and Signal to noise ratio. The 4G heterogeneous network overlapping LTE and WiMax has been studied and a hybrid scheduling algorithm is designed and implemented for the heterogeneous network in this research work. The three types of scheduling algorithms have been implemented on the heterogeneous network and compared. A new load balancing approach is proposed which is combination of FCFS, RR including priority scheduling. The proposed scheduling algorithm‟s performance is compared with other mentioned algorithms and measured in terms of bit error rate, signal to noise ratio and throughput analysis. LTE and WiMax both suffer from load management and congestion as algorithms like FCFS, RR and priority scheduling are not sufficient enough to cope up with high demand network traffic. The memory buffer of serving architecture is not so big that it can compensate for all the node requests at the same time. As the requests are increasing, the scheduler is unable to manage the resources and providing services to all the users. Hence, task scheduling has been implemented to overcome the increasing traffic demands. The research work proposes a parallel scheduling algorithm inspired by a parallel structure which enhances the performance of scheduling in terms of certain QoS parameters as energy consumed, throughput and packet delivery ratio (PDR).|
|Description:||Doctor of Philosophy - ECED|
|Appears in Collections:||Doctoral Theses@ECED|
Files in This Item:
|Final theis Ph D _Sonia.pdf||5.35 MB||Adobe PDF||View/Open Request a copy|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.