Performance Monitoring in Sensor Networks

Abstract

The performance demands imposed on sensor networks carrying heterogeneous connections are challenging and can be satisfied via the deployment of highly dynamic statistical multiplexing protocols which honor the traffic Quality Of Service (QOS), while utilizing the network resources efficiently. While the successful design of high-performance traffic multiplexing protocols requires accurate modeling of traffic statistical characteristics and QOS, variations in the latter characteristics, topological network changes (some due to mobility and short sensor life time), and /or failures in the response of network components affect directly, and frequently dramatically, their performance characteristics. It is thus crucial that traffic and network performance monitoring techniques be deployed, first for the identification of traffic changes and network failures, and then for the subsequent adaptation of the protocol operations, the pertinent recovery of failures, and the appropriate reconfiguration of network topologies. In this paper, we describe an approach to the design and deployment of a distributed performance monitoring system.

A Core Algorithm for Object Tracking and Monitoring via Distributed Wireless Sensor Networks

“…For network traffic monitoring case, we focus on the DWSN dynamics, induced by possible mobility as well as expiration of the ENs. Referring to the discussion and Figure 1 in Section 3, the EN mobility and/or expiration induces user population changes within the clusters of the DWSN; thus, necessitating the deployment of random access (RA) algorithms for transmission to the AFNs [12,31,40,39].…”

“…When the cluster populations are time-varying due to ENs mobility and/or expirations, they transmit to their assigned AFN via a random access (RA) protocol [31,33,40,39]; the life-span of each EN is a function of the channel monitoring and the retransmissions it performs, in compliance with the rules of the deployed RA [40,39]. Operations are performed at all nodes of the backbone network: at the AFNs and the BS.…”

“…In this section, we focus on the DWSN dynamics, induced by possible mobility as well as expiration of the ENs. Referring to the discussion and Figure 1 in Section 3, the EN mobility and/or expiration induces user population changes within the clusters of the DWSN; thus, necessitating the deployment of random access (RA) algorithms for transmission to the AFNs [12,31,40,39]. Assuming starting symmetric DWSN topology, where all clusters have identical EN populations randomly distributed, the per-cluster cumulative data generating process may be modeled as homogenous Poisson, where the Poisson rates across different clusters are identical.…”

“…There has been a plethora of research efforts within the object tracking area, exemplified by the references [8,9,17,19,20,25,35,37,41,47], while related marginal issues are addressed in [20,29], and where some recent results in object identification can be found in [18]. Concurrently, transmission and routing algorithms that are appropriate for the distributed wireless sensor network environment have been proposed in [11,12,14,15,16,21,33,34,48], while some other related network issues are addressed in [44,45,46,49], and where network performance monitoring is considered in [31].…”