Distributed location management in wireless mesh networks

Description

Motivation

At an architectural level, a common belief in the research community is that, unlike in an IP (or Mobile IP) context, node identifiers and location addresses (i.e., the current location in the network of these nodes) should not be integrated into a single identifier, because it generates functional problems. In addition, a scheme designed for wireless mesh networks (WMNs) should exploit their specific features, such as the static and non-power-constrained nature of backhaul wireless mesh routers, which are not present in other wireless networks (e.g., MANETs). This is why mobility management in wireless mesh networks (WMNs) still remains a challenge for its two components, namely handoff and location management. Handoff management deals with node movement detection, re-routing of the traffic flow, and appropriate handling of such flow when a node switches its point of attachment to the network. On the other hand, location management deals with the mapping between a node identifier (node ID) and its current location address. Once the location address is obtained, packets could be eventually forwarded to the destination node.

Goal

We mainly focus on the handoff problem from a communication re-routing perspective. Therefore, this activity is closely linked with research carried out in the group on Routing in self-organized wireless mesh networks. As for distributed location management, the main challenge is how to distribute and to obtain accurate location information in a scalable and robust way. Scalability is determined by the efficiency of the scheme in terms of overhead introduced in the network and state volume in the nodes to achieve two main goals. First, a certain level of robustness must be offered. This is understood as the ability to make accessible the location of a given node even in the presence of impairments in the network. And second, accurate location information in terms of space and time must be provided.

State-of-the-art

Solutions proposed for cellular or Mobile IP networks relying on centralized servers hardly apply to self-organized WMNs. Furthermore, prior work on distributed location management has been mainly studied for mobile ad hoc networks (MANETs). Some of these approaches also integrate routing, hence also solving re-routing of traffic flows when nodes move. When ported to WMNs, these schemes do not fully exploit the additional benefits of WMNs. And those solutions labeled as location management for WMNs are not really fully distributed, hence inheriting some of the problems of centrallized solutions. Furthermore, the evaluation of these proposals is mainly done analytically or by means of simulations.

Evaluation methodology

Given the practical orientation of this task, experimental evaluation of the proposed solutions are carried out in the WMN extension of the EXTREME Testbed.

Publications

Book2
Book3