Smart Communication Systems

The aim of this project is to become familiar with the concept of wireless mesh networks (WMN). Afterwards, after understanding the architecture and the basics, namely routing, MAC-layer and mobility issues, the extension of the current work in the direction of multi-channel and multi-radio for the MAC layer is foreseen.

• Wireless mesh networks (WMNs): consist of wireless access point/routers (mesh routers) forming a backbone to which many other client mesh nodes connect.
• WMNs support ad hoc networking, and have the capability of self-forming, self-healing, and self-organization
• WMNs are multi-hop wireless networks, but with a wireless infrastructure/backbone provided by mesh routers
• Mesh routers have minimal mobility and perform dedicated routing and configuration, which significantly decreases the load of mesh clients and other end nodes
• Mobility of end nodes is supported easily through the wireless infrastructure
• Mesh routers integrate heterogeneous networks, including both wired and wireless. Multiple types of network access supported in WMNs
• WMNs are not stand-alone and need to be compatible an interoperable with other wireless networks

Wireless Mesh Network Interesting Properties

• WM Routers are self-configuring, self-healing links among themselves
• Mixed-mode networking supported (P2P WM client access vs infrastructure (vs wireless routers or backbone))[ref: GC 2003MM NW paper]
• Not scalable, performance strongly degrades with network size
• Throughput/capacity enhancement via: relay nodes
• Facilitation/ load reduction in routing via clustering
• Scalable routing protocols: hierarchical routing too complex to manage; geographical routing: low feasibility
• Cross-layer design: combining L2-L3 parameters && metrics; merging functions; (also L2-L4)

Conceptual

To understand Wireless Mesh Networks

To understand routing in general and in particular in WMN

To select a subset of suitable routing algorithms and analyze them

To understand scheduling, buffering, and other aspects of the MAC protocol

Construction

To build a mesh testbed based on Linux boxes so that the topology is a mesh topology and that each communication node is a multi-channel node

Simulation

To simulate in ns2 an enhancement to a MAC protocol of choice provided that the protocol itself contains an extension or enhancement to a well-known protocol today

Implementation

To implement a topology and configuration based on multi-channel communication and to configure the right routing protocols and to set the proper MAC parameters. Already protocols such as IEEE 802.11 allow setting MAC parameters e.g. for 802.11e via a web interface.

Issues to address during the project

• Multi-path routing in the mesh part
• QoS-aware routing or cross-layer enhanced routing decisions
• How to make the MAC of the mesh multi-channel? And how to benefit from this?
• How would a multi-channel MAC impact a simple randomly built mesh network: Positively: higher data rates, more efficient routing especially on multi-path? Negatively: increased complexity, necessary interfacing, etc