Urban-X: A Self-organizing Cognitive Wireless Mesh Network for Dense City Environments

The practical deployment of Wireless Mesh Networks (WMNs) using unlicensed ISM band within dense urban scenarios is difficult due to the increasing number of wireless devices operating in those licensed exempt frequencies. For this reason, current research on WMN is directed towards novel and more flexible network paradigms which would allow the WMN to dynamically adapt to the environmental interference conditions. Here, we propose Urban-X, which is a novel cross-layer architecture for self-organizing WMNs over urban scenarios. Urban-X combines elements from classical Multi-Radio Multi-Channel (MC-MR) technology with novel Dynamic Spectrum Access (DSA) mechanisms. The self-organizing behavior is achieved through a novel distributed channel assignment scheme, an adaptive multi-path routing scheme and a flexible layer 2.5 channel and path scheduler algorithm. Based on the current interference on each channel, Urban-X performs channel allocation among the nodes of the WMN, updates the available paths towards the gateways and distributes the internal traffic among the paths/channels in order to maximize the network throughput while minimizing interference to the external networks. Simulation results demonstrate the effectiveness of our cross-layer approach in terms of increased throughput compared to traditional routing schemes for WMNs, and its adaptiveness to the variation in channel conditions and external user traffic.