Load-aware Radio Access Selection in Heterogeneous Terrestrial Wireless Networks

Terrestrial wireless network technologies such as UMTS, WiMax and WLAN are used to provide network access for both voice and data services. In big cities the densely populated areas like town centres, shopping centres and train stations which have coverage of multiple wireless networks, a large number of mobile users may be connected to the more common UMTS, thereby creating an unbalanced load across these wireless networks. This high load variation can be balanced by moving mobile users from heavily loaded networks to least loaded networks which involves execution of vertical handovers. Seamless vertical handovers across different wireless networks may be achieved using the IEEE 802.21 Media Independent Handover (MIH) specifications. Radio Access Technology (RAT) selection techniques aim to find the most suitable network that a mobile node should be connected to, for achieving seamless services and meeting the QoS requirements of the user. Traditional RAT selection algorithms are mainly based on the Always Best Connected (ABC) paradigm whereby the mobile nodes are always directed towards the available network which has the strongest and fastest link. This however could create a high variation among the load across the different co-located networks; which cause congestion on overloaded network and eventually increase the call blocking and call dropping probabilities. The unbalanced load situation in co-located networks also causes the poor radio resource utilization as some networks remain under loaded and some become over loaded. There is a need for the load balancing strategies to efficiently utilize the available radio resources and avoid the unwanted congestion situations on overloaded wireless networks. This paper proposes a novel network load aware RAT selection technique in heterogeneous terrestrial wireless networks to minimize the load variation in heterogeneous networks to a suitable level and describes how the MIH framework may be extended to make seamless vertical handover load conscious. The proposed strategy for load balancing consists of two different algorithms; first located in mobile user and the other at network entity such as RNC, BS or AP. The proposed method considers the network type, signal strength, data rate and network load as primary decision parameters for RAT selection process and tries to maintain the load equilibrium on all networks which have common or overlapped coverage areas. Different attributes like load distribution in all wireless networks, average end-to-end delays, jitters and average handover latencies have been observed to evaluate the effects of load balancing in considered scenarios.