In a heterogeneous wireless environment, one of the important aspects of seamless communication for ubiquitous computing is the dynamic selection of the best access network. The problem of access network selection has been addressed through various decision methods based on available network information. Available link bandwidth is one of the important information parameters, which can be used as criterion for network selection. In this paper, we consider available bandwidth as a dynamic parameter to select the network in heterogeneous environment. First, we propose a bootstrap approximation based technique to estimate available bandwidth and then utilize it for the selection of the best suitable network in the heterogeneous environment consisting of 2G and 3G standards based wireless networks. The proposed algorithm is implemented in temporal and spatial domains to check its robustness. Estimation time with varying size of files is used as the performance metric. Through numerical results, it is shown that the proposed algorithm gives improved performance as compared to the existing algorithm.

Wireless Mesh Networks (WMNs) are the evolutionary self-organizing multi-hop wireless networks to promise last mile access. Due to the emergence of stochastically varying network environments, routing in WMNs is critically affected. In this paper, we first propose a fuzzy logic based hybrid performance metric comprising of link and node parameters. This Integrated Link Cost (ILC) is computed for each link based upon throughput, delay, jitter of the link and residual energy of the node and is used to compute shortest path between a given source-terminal node pair. Further to address the optimal routing path selection, two soft computing based approaches are proposed and analyzed along with a conventional approach. Extensive simulations are performed for various architectures of WMNs with varying network conditions. It was observed that the proposed approaches are far superior in dealing with dynamic nature of WMNs as compared to Adhoc On-demand Distance Vector (AODV) algorithm.