Research And Performance Analysis Of Clustering Data Collection Mechanism In Wireless Sensor Networks

Wireless sensor networks, consisting of a large number of sensor nodes, is a multi-hop, Ad Hoc network system. Sensor nodes sense, collect and manipulate the data of the network, and transmit to the base station or the users. Data collection is one kind of the basic technologies for wireless sensor networks. The application of WSN is determined by collecting the appropriate data effectively.The loads of sensor nodes in conventional data collection scheme, such as baseline transmission, are unbalanced. Sensor nodes which is closed to the sink need to transmit more packets by a multicast routing scheme. Because of the heavy loads, the lifetime of the network is short. To optimize the performance of WSN, this paper proposes two model of data transmission schemes, which are used to improve the network transmission capacity and reduce the delay of the network, and reduce the energy consumption of network nodes. We describe and analyze the transmission performance of the wireless sensor network by the way of PEPA, so as to achieve the objective to optimize the data transmission:Firstly, we present an improved model of data collection with grid partition in clustered networks to concentrate on how to balance the loads of sensor nodes in the network. Due to the analysis of three transmission stages of a sensor network, which is composed of a sensor node and a sink node, we derived the formula of the capacity and delay. Unlike the conventional data collection scheme, the data collection scheme with compressive sensing can improve the throughput and reduce the delay of the network. By using compressed sensing technology, the loads of each node can obtain the balance. The sensor nodes will not die in a short time and the lifetime of WSN is extended.Secondly, to save the energy consumption of sensor nodes, we propose a data collection model of a clustered network with compressive sensing (CS-C) to reduce the transmissions, which can save node energy consumption effectively. In this model, each node needs to transmit several measurements to the sink. For large-scale WSN, the total transmissions of the network is still a large number. In the cluster network, nodes within the cluster transmit data to the cluster head in a multi-hop way. Cluster head receives the data of the member nodes in the cluster, the data will be transmitted to the neighbor cluster head in a way of compressed sensing technology. Sink receives the data of all nodes in the network. The cluster size determines the change of the transmission number. The transmissions of intra-cluster and inter-cluster is determined by the size and the number of transmissions of the cluster respectively. In CS-C model, combined with the advantages of multi-hop transmission and compressed sensing, we can reduce the energy consumption of the sensor nodes. We derive the formula of the optimal cluster size and provide the clustering algorithm.Finally, we analyze the performance of WSN by using process algebra performance evaluation (PEPA). PEPA is a high-level model specification for low-level stochastic models. This paper analyzes the performance of wireless sensor networks with LEACH protocol.By using the semantic of PEPA, we describe the process of transmission to achieve the throughput, the utilization rate and the response time varying with the number of nodes. In clustered networks, each cluster is a subsystem of the whole network system. The process of data transmissions of WSN can be described by the way of PEPA to analyze the performance qualitatively and quantitatively. We achieve how the capacity, utilization rate, response time changes with the number of nodes, which is significant to improve the model of wireless sensor networks.