Abstract-Structured Peer to Peer (P2P) overlay networks are becoming increasingly popular. Multi-hop systems achieve a successful lookup in O(log N) hops, whereas one-hop systems approach O(1) hops. Both approaches, but especially one-hop overlays suffer from a high number of identical messages being sent to a number of nodes on the overlay.Previous work showed that P2P networks benefit from the integration of the overlay network with the underlay network in which multi-destination multicast routing is available. This allows combining identical messages from the same source into joint multi-destination multicast messages to significantly reduce the number of messages. Our experimentation has centered around the one-hop EpiChord overlay. Here the problem is described using a Markov Model for more advanced analysis. The Markov Model is believed to be novel in two aspects: it is the first to investigate one-hop overlays and it is the first to study the performance of multi-destination multicast including the consideration of retransmissions of requests.

No abstract

No abstract

This paper addresses the trend in Wireless Sensor Network applications of increasing node density and multi-hop communications that result into extremely challenging design constraints. SSAS, an adaptive hybrid localised sink-oriented Routing/MAC protocol is proposed. It offers a self-configuration scalable solution for large random sensor networks and attempts to alleviate multi-hop congestion problems. The strength of the SSAS protocol is based on the scheduled access at the sinks and the enhancement of the number of one-hop sinks' neighbours by profiting directional antenna capabilities. By means of modelling off-the-shelf sensor node architecture with realistic radio channel assumptions and the use of network simulations, this paper shows significant improvements of SSAS (such as 10 times higher network lifetime) compared to other techniques such as pure contention-based alternative (C-MAC), B-MAC and LEACH. I. INTRODUCTIONWith cost and size of the nodes decreasing due to the technology advances, more applications consider deployment of large multi-hop wireless sensor networks (WSN). Commonly, these are nodes-to-sink (converge-casting) applications where sensed data must be delivered to designated gateway nodes (sinks) that serve as fusion centres, but also distribute control information and application requirements to the sensor nodes. This process may involve the use of multi-hop links and can lead to congestion problems (overhearing, collisions, buffer overflow and energy drain) the major impact of which occurs closer to the sinks. Sensor nodes have limited energy supply and a general approach to maximise the network lifetime is the imposition of sleep/wake cycles. Under decreasing active times, efficient contention MAC protocols attempt to maximise the use of the shared channel to achieve application requirements (throughput, delay, lifetime) but using minimal resources. However, as multi-hop is increased in larger networks congestion problems are likely to appear. More precisely, nodes that are closer to the sinks become more saturated with the forwarding task but with reduced chances of competing for access to the channel. These nodes restrict the network time operation.In this situation, the network could benefit from the use of a TDMA-based protocol that is naturally energy preserving, exempt of contention mechanisms and does not suffer from collisions. However, creating and maintaining a TDMA schedule in a multi-hop nodes-to-sink application can be extremely complex if slots need to be allocated more than one-hop away. The effect of clock drift can also be considerable with low duty-cycle MACs as can topology changes or node mobility. In such situations network design