This paper presents new results on switching control using neural networks. Given a set of candidate controllers, a pair of neural networks is trained to identify the stability region and estimate the closed-loop performance for each controller. The neural network outputs are used in the on-line switching rule to select the controller output to be applied to the system during each control perio...

PIM-SM is the only deployed intra-domain multicast routing protocol that builds both shared and source-based trees. However, it does not provide an efficient mechanism to switch between the two modes. The PIM-SM switching mechanism proposed in this paper aims to improve both network and receivers requirements. It is a coordination-based mechanism in the sense that all concerned receivers contri...

When analyzing a nonblocking switching network, the typical problem is to find a route for a new request through the network without disturbing existing routes. By solving this problem, we can derive how many hardware components of a certain type (Banyan planes in a multi-log network, for instance) are needed for the network to be nonblocking. This scenario appears in virtually all combinations...

0140-3664/$ see front matter 2010 Elsevier B.V. A doi:10.1016/j.comcom.2010.01.022 * Corresponding author. Tel.: +39 0223993580. E-mail address: [email protected] (A. Pattav This paper considers three-stage switching networks able to support multicast traffic, i.e. connections in which one network inlet is connected to more than one network outlet at the same time. The non-blocking condi...

We propose a control strategy based on distributed adaptive leader-follower consensus algorithms for multiagent systems (MAS) affected by switching network events. The strategy allows each agent in the MAS to compute its own control input based on local information and information coming from its neighbors. In this sense, MAS distributed control laws are obtained where the coupling gain of the ...

This paper contributes a distributed packet controller which reduces queueing to a single stage in two-stage packet switches. Software and neural network based controllers are described. Simulations under a range of traffic conditions for a 1024x1024 switch size shows the simplest architecture has the best performance.

This On-Chip Permutation Network for Multiprocessor System-On-Chip presents the silicon-proven design of a novel on-chip network to support guaranteed traffic permutation in multiprocessor system-on-chip applications the routing algorithm, the deflection routing is not energy-efficient due to the extra hops needed for deflected data transfer, compared to a minimal routing. Moreover, the deflect...

A Neural Network based AC-AC voltage restorer is proposed for voltage sags and PWM type active power filter is used for voltage harmonics compensation and EMI reduction. The objective is to apply the neural network switching control technique to the AC-AC voltage restorer to reduce time delays during the switching conditions and switching losses. The aim of the IGBTs used in the AC-AC voltage r...

Conventional peak-traffic based provisioning of cellular mobile networks leads to a significant wastage of electrical energy. Therefore, we propose a novel traffic-aware twodimensional dynamic network provisioning mechanism for enhancing the energy efficiency in OFDMA-based cellular systems. Proposed scheme, named as joint dynamic sectorization and switching of base station (JDSBS), adaptively ...

Network-on-Chip (NoC) architectures represent a promising design paradigm to cope with increasing communication requirements in digital systems. Network-on-chip (NoC) has emerged as a vital factor that determines the performance and power consumption of many-core systems. A novel switching mechanism, called virtual circuit switching, is implemented with circuit switching and packet switching. A...