We consider multiarmed bandit problems with switching cost, define uniformly good allocation rules, and restrict attention to such rules. We present a lower bound on the asymptotic performance of uniformly good allocation rules and construct an allocation scheme that achieves the bound. We discover that despite the inclusion of a switching cost the proposed allocation scheme achieves the same a...

header processing, IPV4, IPV6, internet, networking, router performance, packet classification The complexity of IP header processing is commonly believed to be the main limiting factor for router performance. The purpose of this report is to explore header processing requirements for IP routers, identify different implementation alternatives and discuss their complexity. The conclusion of the ...

Contention resolution schemes are the key determinant of the network performance in optical packet switched networks. This paper presents a contention resolution scheme in wavelength and space dimension. By integrating deflection routing with wavelength conversion, this paper investigates their effects on packet-loss performance through simulation studies. In particular, we propose a load-sensi...

In this paper, we address the problem of active packet routing and forwarding in an active IP network involving both legacy IP routers and active capable routers. We first argue that any traditional routing scheme does not work well in active networks. We then give a formulation of a generalized active routing problem and propose an efficient routing scheme based on the formulation. The scheme ...

The auxiliary switch control using a neuro fuzzy technique is proposed to enhance the efficiency of bidirectional converter used in electric vehicle which performs soft switching. Continuous current data are difficult to obtain during a resonant operation due to limit of DSP ADC capacity. The auxiliary switch control has a lookup table reference of the turn on time. The ANFIS controller properl...

• NMP: Networked Music Performance • Ultra-low delay variant of conferencing • End-to-end delays of 25 rather than 150 ms • NMP is not the same as conferencing • Participants prefer to do their own mixing • The MCU should only relay data streams • The MusiNet project • Ultra-low delay audio and video coding • Each participant indicates what it wants to receive • The MCU maintains a stream routi...

In this paper we present AntNet, a novel adaptive approach to routing tables learning in packet-switched communications networks. AntNet is inspired by the stigmergy model of communication observed in ant colonies. We present compelling evidence that AntNet, when measuring performance by standard measures such as network throughput and average packet delay, outperforms the current Internet rout...

Spoken dialogue systems have strong potentiality in becoming the main stream of next generation man/machine interface. To meet this future, spoken dialogue systems must not only be portable to support various applications, but also be scalable to provide capable dialogue to handle sophisticated services. In this paper, we proposed a goal-oriented table-driven approach to design a portable and s...

In this paper, the physical-layer redundancy method is proposed for the fault-tolerant industrial network. The proposed method consists of the fault detection method and the correction method. The fault detection method uses events created by the state transition in the IEEE 802.4 MAC sublayer and the periodic status frame check for the detection of the fault. The fault correction method correc...

This paper presents a systematic theoretical approach for the analysis of bounds on power consumption in Baugh-Wooley, binary tree and Wallace tree multi-pliers. This is achieved by rst developing state transition diagrams (STDs) for the sub-circuits making up the multipliers. The STD is comprised of states and edges, with the edges representing a transition (switching activity) from one state ...