SUMMARYBandwidth management and traffic control are critical issues to guarantee the quality of service in cognitive radio networks. This paper exploits a network load refinement approach to achieve the efficient resource utilization and provide the required quality of service. A connection admission control approach is introduced in cognitive radio multimedia sensor networks to provide the data transmission reliability and decrease jitter and packet end-to-end delay. In this approach, the admission of multimedia flows is controlled based on multimedia sensors' correlation information and traffic characteristics. We propose a problem, connection admission control optimization problem, to optimize the connection admission control operation. Furthermore, using a proposed weighting scheme according to the correlation of flows issued by multimedia sensors enables us to convert the connection admission control optimization problem to a binary integerprogramming problem. This problem is a kind of a Knapsack problem that is solved by a branch and bound method. Simulation results verify the proposed admission control method's effectiveness and demonstrate the benefits of admission control and traffic management in cognitive radio multimedia sensor networks.

a b s t r a c tPerformance evaluation of transport layer protocols in cognitive radio sensor networks (CRSNs) is useful to provide quality-of-service for real-time reliable applications. This paper develops an analytical framework to model the steady-state sending rate of collecting cognitive radio (CR) sensors in rate-based generic additive-increase multiplicative-decrease (AIMD) and additive-increase additive-decrease (AIAD) congestion control schemes. Evolution process of sending rate is modeled by a discrete time Markov chain (DTMC) in the terms of queue length. We model the queue length distribution of a CR node by a semi-Markov chain (SMC) with assuming general probability density functions (PDFs) of input rate and attainable sending rate of the node. These PDFs are derived based on the parameters of MAC and physical layers and CRSN configuration. The proposed models are verified through various simulations.

No abstract

Performance guarantees for congestion control schemes in cognitive radio sensor networks (CRSNs) can be helpful in order to satisfy the quality of service (QoS) in different applications. Because of the high dynamicity of available bandwidth and network resources in CRSNs, it is more effective to use the stochastic guarantees. In this paper, the stochastic backlog and delay bounds of generic rate-based additive increase and multiplicative decrease (AIMD) congestion control scheme are modeled based on stochastic network calculus (SNC). Particularly, the probabilistic bounds are modeled through moment generating function (MGF)-based SNC with regard to the sending rate distribution of CR source sensors. The proposed stochastic bounds are verified through NS2-based simulations.

No abstract