Search

In this paper, the problem of finite-time state estimation for delayed periodic neural networks over multiple-packet transmission is addressed. The components of measurement output are separately transmitted by multiple-packet transmission, and the randomly occurring packet dropouts of different channels are described by mutually independent Bernoulli processes. In order to improve the robustness of the estimator, a non-fragile estimator is designed. In addition, some sufficient criteria are given to ensure that the estimation error system is stochastically finite-time stable and stochastically finite-time bounded, and the gains of non-fragile estimator are then derived based on these results. Finally, simulation results are provided to illustrate the effectiveness of the proposed estimator design approach. [ABSTRACT FROM AUTHOR]

Wireless Sensor Networks nowadays find wide variety of applications especially in real time. Innovative methods of energy efficient protocols and transmission reduction techniques keep improving to enhance the lifetime of the sensor nodes as they are powered by non-rechargeable batteries. Multi hop transmission and data aggregation are major techniques to reduce the power spent by the sensor node. In this paper, we propose a new ribbon structure for the existing multi hop WSN topologies with modified media access control mechanism called co-operative MAC. The ribbon structure is proposed to reap benefits of PEGASIS and APTEEN protocols. The low power consumption as in PEGASIS is maintained but the number of data packets transmitted is reduced by half. In the proposed scheme, only one of the two nodes along the parallel path involves in data transmission alternating roles in every cycle of aggregation. However, for values sensed above threshold, the inactive node interferes with normal cycle and gets its data transmitted to the sink node. This algorithm is compared with cluster based and chain based protocols and the simulation results show significant energy savings. [ABSTRACT FROM AUTHOR]

Abstract: Optical burst switching is a promising all-optical switching technology for the next generation optical networks and signaling is one of the key components of it. According to propagation direction of control packets, the existing signaling schemes for optical burst switched networks can be roughly divided into three categories: one-way, two-way, and hybrid signaling. Since one-way signaling launches bursts without the acknowledgment of successful reservation of resources, it has small end-to-end delay but high data loss rate; two-way signaling uses a control packet to gather the state of links and needs the successful acknowledgment before sending out a burst, so it has large end-to-end delay but low data loss rate. A hybrid signaling gets a tradeoff of performance between one-way and two-way signaling. In this paper, we propose another hybrid signaling named inverse two-way signaling. By introducing the process of link state collection into one-way signaling, the inverse two-way signaling can decrease the data loss rate without increasing the end-to-end delay compared to one-way signaling. In other words, the inverse two-way signaling is an improved version of one-way signaling. [Copyright &y& Elsevier]

Abstract: There has been much interest to emulate the behavior of Output Queued switches. The early result of such attempts was reported by Prabhakar and McKeown using the CIOQ switches with speedup factor of 4. Subsequently, Stoica and Zhang and independently Chuang et al. showed that a speedup of 2 in conjunction with their scheduling schemes would be sufficient for CIOQ switches to emulate Output Queued switches. Additionally, Chuang et al. showed that in “Average Sense” a speedup of 2−1/N is necessary and sufficient for CIOQ to emulate Output Queued switch behavior. Our paper reports that in the “Strict Sense” a speedup of 2 is both necessary and sufficient. We show this requirement using examples for 2x2 and 3x3 switches. Then, with a constructed traffic pattern, it is proved that in the “Strict Sense” a speedup of 2 is necessary to emulate the behavior of an Output Queued switch for any switch size N. Combining this result with the previous scheduling schemes, we conclude that in the “Strict Sense”, a speedup of 2 is the necessary and sufficient condition to emulate the behavior of an Output Queued switch, using a CIOQ switch. Additionally, easing the assumptions and allowing the packet segmentation, it is shown that the speedup requirement to emulate the behavior of an Output Queued switch can be reduced to values even smaller than 2−1/N. For this case a lower bound of 3/2 and an upper bound of 2 is proved. [Copyright &y& Elsevier]

Abstract: In the problem of buffer management with bounded delay, packets with weights and deadlines arrive at a network switch over time, and the goal is to send those packets on the outgoing link while maximizing the total weight of the packets that are sent before their deadlines. We present a study of randomized algorithms that are competitive against an adaptive adversary. Previous studies considered only the oblivious adversary model that does not capture dependency of network traffic on the packet scheduling algorithm. We give a new analysis of a previously known algorithm, which shows that it remains -competitive even against an adaptive adversary. We complement this with a 4/3 lower bound on the competitive ratio on 2-bounded instances, in which each packet has a lifespan of one or two steps. We also study more restricted 2-uniform instances, in which every packet has a lifespan of exactly two steps. For such instances we give a 1.2 lower bound on the competitive ratio of arbitrary algorithms and 4/3 lower bound on the competitive ratio of memoryless scale-invariant algorithms. Finally, we devise a 4/3-competitive memoryless scale-invariant algorithm for 2-bounded instances, matching two of these lower bounds. [Copyright &y& Elsevier]

Abstract: This paper discusses a new idea for improving the recovery of (n, k, m) convolutional-code-based packet-level FEC. We extend the coefficients in the generator matrix to the elements on GF(28). Each byte in the redundant packets is generated independently, assuming the bytes are the elements on GF(28). The postreconstruction receiving rate is analytically derived from the necessary conditions for recovery. Moreover, the ability to recover lost packets is estimated based on analysis, and we demonstrate that the extension of coefficients improves the number of recovered packets more than the conventional method where coefficients are 0 or 1. [Copyright &y& Elsevier]