No abstract

Major benefits of wireless sensor nodes of IoT like low cost and easy deployment are advocating their usage in variety of applications. Some of them are health monitoring, agriculture, environmental and habitant monitoring, and water monitoring. These nodes are autonomous in nature. It follows that they like to operate in a dynamic and adaptive network environment. So, the communication mechanism between IoT nodes must be robust and adaptive with respect to the environmental change. Unfortunately, the traditional networking architecture supports limited and fixed network computations. These limitations inhibit flexible and robust IoT nodes communication. In addition, the energy consumption in communication nodes is high due to limited processing. To address these issues, this paper gives rebirth to the active system. The proposed active network framework brings a novel integration of the active system with recent technologies (software-defined networking and network function virtualization). As a result of integration, the active system runs as a network function virtualization under the control of software-defined networking. In our view, the amalgam of recent technologies with the active system will promote a robust and flexible IoT nodes communication along with reduced energy consumption. Moreover, various design benefits such as security, flexibility, usability, cost, and performance will be added to the system. Additionally, the proposed framework is open and generalized. It can be extended to other networks such as mobile, satellite, and vehicular networks.

In this paper we extend our recent work [3] by analyzing the behavior of mobility scenario in saturation conditions for evaluating packet delays. For the first time we have shown that the normalized measure of dispersion for service time intervals and offered load dependency holds for multi-hop IEEE 802.11 wireless networks with mobile nodes, as well. Thus this paper not only extends and validates our previous results but also establishes that wireless mobile networks have the similar dependency pattern. We further extend this work by showing that the service time intervals do not follow exponential distribution but instead follow gamma distribution. The patterns for the shape parameters are also established for accurately computing average end-to-end packet delay for mobile networks. The analytical model presented in this paper has been verified through extensive simulations.Keywords-average end-to-end packet delay; multi-hop wireless mobile networks; normalized measure of dispersion and offered load dependency; routing

Self Similarity of network traffic has been an area of significant interest in the research on performance analysis of networks. It is widely known that a number of network traffic types display Self Similar behavior and its performance characteristics are significantly different from the typical Poisson or exponentially distributed traffic characteristics. In this paper we conduct an analysis of the impact of Pareto distributed traffic on the probability of blocking of wireless LANs. In particular we consider the impact of varying the coefficient of variance of the interarrival times and packet size distributions on the probability of blocking.

No abstract