Most of the work on MANET (Mobile Ad hoc Network) today has addressed flat, homogenous, single connection point or multihop networks. Applications of wireless multihop network are very specialized, and include search and rescue in the jungle, natural disaster recovery and the battlefield. These are in fact "extreme" applications of ad hoc networking. We believe that a much more common (and commercially viable) use of multihop MANET concepts will emerge in so-called "opportunistic" mesh network configurations. In these scenarios, the ad hoc network is an extension of the existing fixed telecommunications infrastructure. Moreover, there are scenarios where the fixed infrastructure may become severely degraded or even completely dysfunctional. In such cases, the opportunistic ad hoc network may take on a new role of temporary replacement of the fixed infrastructure.
In this project, we focus on the urban environment and consider the ad hoc opportunistic extension of a fixed urban infrastructure. Now, the "fixed" urban infrastructure is itself undergoing a dramatic evolution. The traditional wireless systems have mostly used cellular-phone-style radio links, with point-to-point or point-to-multipoint transmission. These wireless formats have liabilities in industrial and urban applications such as rigid structure, meticulous planning requirements, and dropped signals. In contrast, wireless mesh networks are multihop decentralized systems in which devices assist each other in transmitting packets through the network, especially in adverse conditions. These networks can be deployed into place with minimal preparation, and they provide a more robust, efficient, reliable, flexible system that can be extended to thousands of devices.
Derived from military-funded research into mobile networks, mesh networking eliminates the need for a central communication coordinator and its associated reliability and efficiency issues. Fruits of the military research are now making their way into the commercial sector, with several companies offering products for both residential and industrial use. An important application is in fact the urban Mesh network. In this project we study the extension of the urban Mesh to include car to car communications creating what may be called the "urban vehicular grid". We identify the applications of such an opportunistic ad hoc extension (from driving safety to entertainment and disaster recovery). We also identify and address the challenges of this design ranging from optimum routing protocols to minimize the number of hops while ensuring security and seamless mobility, to Quality-of-Service (QoS) MAC and network protocols to support emerging multimedia applications; to the need to guarantee efficient autonomous operation when the infrastructure fails.