Performance evaluation of GPRS/802.11b mobile-node initiated handover based on signal strength criteria

Abstract

There has been a great concern by cellular providers about the impact Wireless Local Area Network (WLAN) systems that are based on 802.11b wireless standard pose to mobile business [24]. While cellular providers are still trying to offer high GPRS (General Packet Radio Service) data rates of more than 57kbps, 802.11b, also known as Wi-Fi, is already achieving more than 100 times 2.5G network data rates at 1 to 11Mbps. A user can enjoy high data rates using Wi-Fi in the comfort of a wireless connection to the Internet, but looses connection if he roams out of the range of the Wi-Fi Access Point (AP) covering the Mobile Node. In GPRS the user need not worry about the range shortcoming of the network, but the lower data rates of the connection to the Internet as well as the more expensive costs, which are shortcoming when compared to Wi-Fi. The apparent solution for fulltime coverage and high data rates at low cost during data communication is to integrate the two networks. This integration has proved to be a challenge for mobile operators in terms of offering mobility between 2.5G networks and 802.1 lb/a/g networks. This thesis describes research followed to realize and evaluate performance of a handover mechanism based on signal strength criteria between GPRS and 802.11b access networks. The purpose is to integrate the two access networks with one Mobile Node (MN) and to evaluate the performance of a Mobile-Node initiated handover based on signal strength. This paper also describes the developed GPRS/802.11b testbed using Mobile IP standard (RFC 2002) to achieve handover between the hybrid networks. Mobile IP is an open standard still being refined by the IETF (Internet Engineering Task Force) to allow all IP based communication devices to roam from one network to the other.