BANDWIDTH RESERVATION IN IEEE 802.16 WIRELESS MESH NETWORKS

[Total: 0    Average: 0/5]

The need for broadband wireless access systems in residential and small-to-medium sized business environments is increasing due to their requirement for higher bandwidth network access. IEEE 802.16 is a standards-based technology enabling the delivery of last mile wireless broadband access as an alternative to wired broadband like cable and DSL and includes a mesh mode operation where nodes cooperate to relay the users packets toward their destinations in a multi-hop fashion. The mesh mode does not include support to traffic flows with strict Quality of Service requirements. In this work, I propose an End-to-end Bandwidth Reservation Protocol (EBRP) in the backhaul of a Wireless Mesh Network using IEEE 802.16 mesh. Although EBRP shares the same conceptual rationale of most IP-based reservation protocols, such as RSVP, its distinctive feature is that it is carried out at the Medium Access Control (MAC) layer. EBRP exploit the control messages to solve the problem of providing traffic flows with strict real-time requirements through a end-to-end bandwidth reservation, which is integrated with the IEEE 802.16 MAC through the addition of some information elements to the MAC control messages. The messages that are used to establish traffic flows are conveyed as Information Elements (IEs) of Mesh Distributed Scheduling (MSH-DSCH) MAC control messages, which are advertised periodically by all nodes. The MSH-DSCH format has been modified so as to be still compliant with the IEEE 802.16 standard by using the fields that are reserved for future use. The framework includes a call admission control procedure which requires all the nodes to have complete knowledge of the network topology, which can be achieved by modifying existing routing protocols. EBRP allows the available resources to be allocated efficiently by exploiting technology-specific information available at the MAC. The main idea is to partition the data sub-frame into a number of groups of equal size. Each link of the network is then assigned to one of these groups in a distributed manner by all nodes so that all the links that belong to it can be activated simultaneously and any node of the network can allocate resources for one of its links independently of one another.