Our wireless nodes are utilized for the ambient and pest monitoring of wood. Ambient monitoring is performed by measuring the relative ambient humidity and temperature to compute the equilibrium moisture content of the wood (EMC). Pests are detected using LEDs and light sensors which detect reflection variations when an insect such as a termite, ant, cockroach, etc. crosses the detector��s selleck chem inhibitor field. Most of the energy requirements of these nodes are invested in pest detection.The installed implementation of the monitoring system is based on a star configuration, where nodes send their information to a sink. When a large number of nodes are required, the star configuration is unsuitable as it does not scale well; this is the case, say, if we need to implement such a system in a historical building, whose structure does not lend itself to this configuration type.
For example, the star configuration is appropriate for an altarpiece, but not for an entire Romanesque church built from stone.The implementation of large WSNs, as required in this environment, requires the use of multi-hop approaches, dealing with several issues, such as routing topology control, etc. A lot of approaches have been proposed [1]. As an option, node clustering has been addressed by many researchers as a new technique that will allow for simpler topology management and improved network lifetime [1]. Previous studies have shown that organization of nodes into clusters provides greater energy efficiency [2]. Furthermore, several applications of wireless sensor networks require only an aggregate value to be reported to the operator [3].
In this case, the data gathered from each node is processed locally and aggregated at a coordinator node named cluster head (CH) and the redundant data (if any) is removed to provide more accurate reports about the local region being monitored, reducing the communication overhead.There are different approaches for clustering algorithms. In homogeneous networks, where CHs are just regular sensor nodes, clustering algorithms must be distributed without coordination. In a few approaches, a centralized authority (the Sink) partitions the nodes offline and controls the cluster sizes according to the number of members, especially in heterogeneous networks where CHs are rich in resources.
The Low Energy Adaptive Clustering Hierarchy (LEACH) [4] is a clustering protocol that utilizes a random selection and frequent rotation of CHs for distribution of the total load across all Carfilzomib nodes. The clustering process involves one iteration, after which a node decides whether to become a CH or not, with nodes alternately acting as CHs. Data communication in LEACH is based on single-hop Paclitaxel communication model. There are two variants of LEACH, which are referred to as LEACH-C (LEACH-centralized), and LEACH-F (LEACH with Fixed clusters).