No abstract

No abstract

Network selection in 4G networks by nature is a multi-dimensional problem and many parameters need to be considered for handover decision making. Successful classical handover methods based on Received Signal Strength (RSS) are no longer applicable for heterogeneous handovers because different network standards use different RSS and cannot be compared directly to each other. Heterogeneity does not only exist in network architectures and in their respective Quality of Service (QoS) but also in user application needs and user contexts. Network selection in heterogeneous networks has been modeled as Multiple Attribute Decision Making (MADM).In this paper we investigate how the task of network selection can be distributed across multiple network entities putting minimum possible burden on the Mobile Node (MN) which usually is the most restricted network component in terms of memory, processing power and battery life. This distribution increases the scalability of handover algorithm and will also allow the inclusion of more parameters for handover decision making. Considering more parameters for handovers will result in accurate, robust and general algorithms, capable of fulfilling a variety of user contextual, preferential and applications QoS needs. The Media Independent Handover (MIH) and its different services are used to retrieve and share information among MIH enabled nodes and for conformity among different network standards. Simulations are used to show the effectiveness of a simple handover algorithm based on Simple Additive Weighting (SAW) for network selection considering different types of user services and Service Level Agreement (SLA's). SAW belongs to the MADM classification of heterogeneous handovers and is chosen here due its simplicity and efficiency. The handover efficiency of the proposed distributed schemes is compared with an "802.11 Preferred" scheme. Only handovers between Wi-Fi and WiMAX are considered. But the handover framework is general and can be extended to consider other wireless and mobile communication networks like 3G, CDMA etc.

The widespread deployment of mobile personal communication services will lead to a tremendous increase in the number of real time updates and queries to the network databases. The centralised HLR in GSM can be a potential bottleneck. Most users have regular calling and mobility patterns, and the mobility profile of such users can be used to reduce location updates.By replicating the location of the called mobiles at the source-MSC/VLR, HLR queries can be reduced. This paper studies the cost of replicating user mobility profiles. A scheme for replicating the profiles, exploiting the regular calling patterns of users, is also presented. The replication scheme is analysed and compared with the GSM method. It is shown that considerable reduction in HLR accesses and signalling cost can be achieved at the expense of extra memory at the MSC/VLR.

No abstract