Date of Award

4-2017

Document Type

Thesis

Degree Name

Master of Science in Electrical Engineering (MSEE)

Department

Electrical Engineering

First Advisor

Dr. Atef Amin Abdrabou

Second Advisor

Dr. Mohammad Abdul Hafez

Third Advisor

Professor Sa bah Alkass

Abstract

Communication technologies witness a wide and rapid pervasiveness of wireless machine-to-machine (M2M) communications. It is emerging to apply data transfer among devices without human intervention. Capillary M2M networks represent a candidate for providing the reliable M2M connectivity. In this thesis, we propose a wireless network architecture aims at supporting a wide range of M2M applications (either real-time or non-real time) with acceptable QoS level. The architecture uses capillary gateways to reduce the number of devices communicating directly with a cellular network such as LTE. Moreover, the proposed architecture reduces the traffic load on the cellular network by providing the capillary gateways with dual wireless interfaces. One interface is connected to the cellular network, whereas the other is proposed to communicate to the intended destination via a WiFi-based mesh backbone for cost effectiveness. We study the performance of our proposed architecture by the aide of the ns-2 simulator. An M2M capillary network is simulated in different scenarios by varying multiple factors that affect the system performance. The simulation results measure average packet delay and packet loss to evaluate the quality-of-service (QoS) of the proposed architecture. Our results reveal that the proposed architecture can satisfy the required level of QoS with low traffic load on the cellular network. It also outperforms a cellular-based capillary M2M network and WiFi-based capillary M2M network. This implies a low-cost of operation for the service provider while meeting a high-bandwidth service level agreement. In addition, we investigate how the proposed architecture behaves with different factors like the number of capillary gateways, different application traffic rate, the number of backbone routers with different routing protocols, number of destination servers, and the data rates provided by the LTE and Wi-Fi technologies. Furthermore, the simulation results show that the proposed architecture continues to be reliable in terms of packet delay and packet loss even under large number of nodes and high application traffic rates

Included in

Engineering Commons

Share

COinS