The wireless mesh network with cognitive radio are formed with mesh routers with cognitive ability which are interconnected by wireless links enabling to relay the traffic and some of them act as a gateway to access the Internet. The cognitive ability is a cost efficient manner to increase available bandwidth but requires an adaptive bandwidth management mechanism to deal with dynamics of primary users’ activities.
In this thesis, we model the joint operation of channel allocation and routing of the secondary users (without license access) connections in wireless mesh networks with cognitive radio. The process of channel allocation includes the channel reuse opportunities to improve the network performance. We define an economic framework for adaptation and control of network resources that allows us to integrate the routing process with the process of channel allocation with the goal of network profit maximization. The economic framework is based on the notion of state dependent node shadow price that is derived from Markov decision theory in decomposed mode. The node shadow prices are used as routing metrics while their average values are used to allocate the channels among the different nodes. The deployment of this decomposed model has the advantage of decentralized implementation of the processes used in the economic framework enabling their real time executions.
The routing of the secondary users connections are made without damaging the transmission of primary users and without creating any interference with transmissions from other secondary users. Cognitive channels are used opportunistically and priority access of the primary user in its frequency band (its channel) is modeled by preemption. Secondary users connections are considered homogeneous and we address only the short term capacity adaptation in which the number of available channels in the network is constant we also assume that there is always sufficient number of radio interfaces when channels are reassigned to nodes.
The results are given for both ‘no channel reuse’ and ‘channel reuse’ cases and in each case the performance obtained confirms the theoretical optimization model presented in the corresponding chapters. In addition, both heuristic and analytic approaches of capacity calculation are compared showing the same performance. The simulation results illustrate the maximization of network profit and the effectiveness of the proposed channel allocation scheme that is integrated with channel reuse algorithm.
| Date | 22 Jul 2013 |
|---|
| Original language | French |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Zbigniew Dziong (Supervisor) |
|---|
Mossanen Amini, R. (Author),
Dziong (Supervisor),
22 Jul 2013Student thesis: Doctoral thesis › Doctorate in Engineering: Engineering