Multiradio extensions for the OLSR-Protocol

Description

Wireless Mesh Networks (WMNs) have emerged recently as a new network architecture able to extend the coverage and increase the capacity of wireless access networks. WMNs are promising solutions to provide both indoor and outdoor broadband wireless connectivity in several envi-ronments without the need for costly wired network infrastructures. The network nodes in WMNs, named mesh routers, provide access to mobile users, like access points in Wireless Lo-cal Area Networks (WLAN) or base stations in cellular systems, and they relay information hop by hop, like routers, using the wireless medium. Mesh routers are fixed and usually do not have energy constraints. Therefore, WMNs are characterized by infrequent topology changes mainly due to node failures. Due to the properties of the wireless medium, in order to successfully transmit on a channel, all the interfaces that are in the radio range of the sender and that use the same channel can't trans-mit. Implicitly, this means that an interface, while receiving data, can not transmit at the same time. Therefore, data-throughput in single-interface mesh networks is highly limited. Recent studies propose to increase throughput in mesh network by using multi-radio nodes. Deriving from those considerations, two questions arise: 1st: Since interfaces within the radio range of the sending device that use the same channel as the sending interface are prohibited to send, how does the routing have to be managed to minimise this inevitable blocking? 2nd: How do channels in ad-hoc networks have to be assigned in order to maximise throughput in an ad hoc wireless network? A focus of development of the Advanced Network Technologies Laboratory (ANTlab) of Politecnico di Milano is the development of the proactive routing protocol Optimised Link State Routing (OLSR) for multi-radio wireless nodes. The UniK OLSR implementation has been taken as a base for the development. At the moment, research is focused on finding new metrics to improve throughput and to make routes more stable. Simulations that have been carried out at the ANTlab have shown promising results. Since simulations can provide different results from real life testing, the next step is to create an implementation and collect real life data on a test-bed. The thesis should address the following:

• "Based on the ongoing development at the ANTlab, Bernd Jdkel shall develop an exten-sion of OLSR. That extension should include a new metric estimating the blocking-time of wireless devices using the same channel as the receiving device. Furthermore the sug-gestion should be implemented in the UniK OLSR implementation using the C pro-gramming language. Testing shall be conducted on a 5-node testbed. The results of the testing shall be a basis for a critical discussion of the implementation. Content of the dis-cussion should be the transmission properties concerning traffic-throughput and protocol overhead in comparison with the single-radio-interface protocol.
• "A second step is to find and compare three algorithms for the automatic assignment of channels in wireless networks. There shall be a static, a distributed and a centralised algo-rithm. Starting from that comparison, another extension to the OLSR protocol shall be suggested. That suggestion shall be implemented in the used UniK OLSR implementa-tion using the C programming language. Again, testing on a 5-nodes testbed shall be conducted. The results of this testing shall be critically discussed as well. The focus of the discussion shall be again traffic-throughput and protocol overhead this time in com-parison with the first implementation without automatic channel-assignment.

This thesis will be performed in collaboration with and within the premises of Professor Antonio Capone, Politecnico di Milano, Dipartimento di Elettronica e Informazione, Via Ponzio, 34/5, 20133 Milano, Italy