Search

Wireless sensor networks will be one of the fundamental technologies for realizing the future Internet of Things (IoT) environment. In IoT, the number of connected devices is expected to increase drastically and there will be a wide variety of requirements for application services, which will lead to frequent modifications or construction/destruction of topologies. In such situations, it is essential to know how power-saving, low-latency, and highly efficient IoT network topologies can be constructed. In this paper, we take inspiration from the brain's network of interconnecting neurons is known for its efficient properties. We propose a virtual IoT network construction method based on the Exponential Distance Rule (EDR) model that describes the connection structure of the areas in the cerebral cortex. Since the original EDR model deals with large-scale networks with an enormous number of neurons and generates links between nodes considering physical distance constraints, the virtual IoT network constructed by the proposed method is able to achieve high scalability, low latency, and high communication efficiency at a relatively low cost.

Average consensus algorithms have attracted increasing interest in the last decade because of their potential for use in high-density wireless sensor networks. This paper analyzes an algorithm that is based on a model of asymmetric broadcasting on a random geometric graph, in which nodes broadcast and listen only intermittently. We show that each node can easily estimate its update weight from the degree of silence in its neighborhood, and, after the weights have been assigned in this way, that the consensus algorithm converges to the true average. Both a synchronous and an asynchronous update model are analyzed.

Computer networks have become highly complicated and less flexible to handle emerging problems which often occur nowadays. In order to cope with unpredictable problems, the concept of biologically inspired networks has been introduced, which provides a high degree of robustness and adaptability to computer networks. However, the performance of the network often relies heavily on the configurable parameters assigned during the deployment process, where end nodes cannot change these parameters during runtime to achieve the desirable performance. In this paper, we introduce a new method, called attractor perturbation (AP) allowing end nodes to influence the average of an observable performance metric at runtime without directly manipulating any optimal parameters of underlying protocols. An example application in this paper is a traffic distribution over multi-path routing protocol in MANETs, where the target variable is end-to-end delay. The approach to solve for the appropriate amount of management influence and simulation results are shown in this paper.

In this paper we examine the effects of Internet traffic on the quality of service (QoS) in GPRS wireless networks. With a stochastic source traffic model describing the user behavior, we will derive subjective and objective quality of service measures in terms of WWW downloading time and the transmission bandwidths on TCP and TBF level. Comparing the obtained values with wireline network modems and ISDN yields a slowdown factor which indicates the subjective degradation that the wireless user experiences.