QLREDActive Queue Management Algorithm

The Random Early Detection (RED) algorithm has not been successful in keeping the average queue size low. In this paper, we an improved RED-based algorithm called QLRED which divides the dropping probability function of the RED algorithm into two equal segments. The first segment utilises a quadratic packet dropping function while the second segment deploys a linear packet dropping function respectively so as to distinguish between light and high traffic loads. The ns-3 simulation performance evaluations clearly showed that QLRED algorithm effectively controls the average queue size under various network conditions resulting in a low delay. Replacing/upgrading the RED algorithm in Internet routers requires minimal effort since only the packet dropping probability profile needs to be adjusted.

Distributed Event Triggering Algorithm for Multi-Agent System over a Packet Dropping Network

The availability of wireless networked control systems (WNCSs) has increased the interest in controlling multi-agent systems. Multiple feedback loops are closed over a shared communication network in such systems. An event triggering algorithm can significantly reduce network usage compared to the time triggering algorithm in WNCSs, however, the control performance is insecure in an industrial environment with a high probability of the packet dropping. This paper presents the design of a distributed event triggering algorithm in the state feedback controller for multi-agent systems, whose dynamics are subjected to the external interaction of other agents and under a random single packet drop scenario. Distributed event-based state estimation methods were applied in this work for designing a new event triggering algorithm for multi-agent systems while retaining satisfactory control performance, even in a high probability of packet drop condition. Simulation results for a multi-agent application show the main benefits and suitability of the proposed event triggering algorithm for multi-agent feedback control in WNCSs with packet drop imperfection.

Reliable and Hybridized Trust Based Algorithm to Thwart Blackhole Attacks in MANETs using Network Preponderant Determinants

<p>This Mobile adhoc networks is a perpetual and autogenous organization without framework, and the mobile nodes are coupled cordlessly. Owing to the deficiency of framework assistance, reliable data distribution is a demanding process in mobile adhoc networks and this mobile adhoc network is unguarded to many categories of attacks. A black hole attack in Mobile Adhoc networks cites to an attack by the malevolent node which strongly get hold of the path from sender to the receiver, by means of perversion of subsequence word. With regard to diminish the menace from the malevolent node, the authors encompass the notion of trust in mobile adhoc networks. In this paper, we cope with a packet dropping misconduct named Black hole Attack and we propounded a new hybrid trust based secured algorithm hinged on four new parameters to scrutinize, whether the transitional nodes are transmitting the packets correctly to the adjacent nodes. This paper presents a new hybrid trust based concept to pinpoint the malevolent node hinged on the computation of trust value. Using ns2 simulator, we analyse the performance of our proposed method and proved the detection efficiency. The investigated results shows that our proposed work can precisely diagnose the malevolent nodes and assure a good packet delivery ratio and network throughput.</p>