0 Introduction Now wireless Sensor Network has become a great potential measurement tools. It is a tiny, inexpensive, energy-constrained Sensor Composed of nodes through the wireless way Communicate Multi-hop network, the goal is within the coverage area of information collection, processing and transmission. However, the sensor nodes small, rely on Battery Power supply, and the replacement of Battery Inconvenience to the efficient use of energy, improve the life cycle of the node is the most important issue facing the sensor network. Sensor networks discussed here the data transmission, and cited several internal network data compression mechanism (within the network data transmission through the link, for pooling and compression) data transmission to reduce the amount of energy saving algorithms. 1 traditional wireless sensor network data transfer 1.1 direct transmission model Direct transmission model is the sensor node to the data collected by the larger transmit power directly hop to Sink node for centralized processing, shown in Figure 1. The disadvantage of this method is: Sink node farther from the sensor nodes need to send a great power with the sink node can achieve the purpose of communication, but communication from the sensor nodes is limited, so far from the Sink Sink node often can not reliable communication nodes, which can not be accepted. And in the larger communication distance of nodes required to complete the energy-intensive and Sink node communication, likely to cause the nodes run out of energy quickly, such sensor networks is difficult to be applied in practice. 1.2 Multi-hop transmission model This way similar to the AD-Hoc network model, shown in Figure 2. Each node does not own any of the data processing, but to adjust transmission power, low power through multi-hop to the measured data to Sink node further focus. Multi-hop transmission model of direct transfer well to improve the defects have been made more efficient energy use, which is widely used sensor network premise. The disadvantage is that: When the network size is large, there will be hot issues, namely, at the intersection of two or more nodes in the path, and hop from the Sink node node (the node it is called bottleneck ), shown in Figure 2, N1, N2, N3, N4, which in addition to its own transmission, we should also pass in a multi-hop acts as an intermediary. In this case, these nodes will soon run out of energy. To energy efficiency as a precondition for the sensor network, this is obviously not a very effective way. 2 wireless sensor network data fusion technology In large-scale wireless sensor networks, because each sensor monitoring range and reliability are limited, and in place when the sensor nodes, and sometimes the scope of monitoring to make sensor nodes overlap with each other to enhance the network The information collected robustness and accuracy. Then, in the wireless sensor network sensor data will have some spatial correlation, which is similar to the node from the data transmitted with a certain redundancy. In the traditional data transmission mode, each node will transmit all the sensor information of which will contain a lot of redundant information, that a considerable portion of the energy used for unnecessary data. The sensor networks transmit data much larger than the energy consumption for data processing. Therefore, large-scale wireless sensor networks, so that each node multi-hop transmission of sensor data to Sink node before the data fusion is necessary, data fusion technology came into being. 2.1 centralized data fusion algorithm 2.1.1 Clustering Algorithm for LEACH model Order to improve the hot issues, Wendi RabinerHeinzelman other proposed use in wireless sensor networks clustering concept for the network is divided into different levels of LEACH algorithm: the election in some way periodic random cluster head, cluster head in the wireless channel the broadcast information, the other nodes detect the signal and select the strongest signal cluster head to join, thus forming different clusters. The connection between cluster heads form the upper backbone network, communication between all the cluster were forwarded through the backbone network. Cluster member to transfer the data to the cluster head node, cluster head up a cluster head node and then transmitted, until the Sink node. Figure 3 shows the two-cluster structure. This approach reduces the node transmit power, reduce unnecessary link, reducing interference among nodes, to maintain the balance of energy consumption within the network, the purpose of extending the network lifetime. Drawback of the algorithm is: the realization of cluster and cluster head selection requires a fair amount of overhead, and too much reliance on cluster members of the cluster head for data transmission and processing, so the energy consumption of cluster head quickly. Cluster head in order to avoid depletion of energy, frequent need to select the cluster head. Meanwhile, the cluster head and cluster members of the hop-to-multipoint communication, scalability is poor, does not apply to large-scale networks. 2.1.2 PEGASIS algorithm StephanieLindsey and others based on LEACH is proposed PEGASIS algorithm. This algorithm assumes that each node in the network are isomorphic and stationary, to get through the communication nodes and other nodes in the relationship between the position. Each node using a greedy algorithm to find and connect to its nearest neighbor, thus forming a chain across the network, and set a distance from the nearest node Sink head node for the chain, it Sink to hop communication. Data is always transferred between a node and its neighbors, node multi-hop manner by transferring data to Sink Department turns. Shown in Figure 4. Obvious shortcomings of the algorithm, the first of each node must know the other nodes in the network location information. Second, the chain head node for the bottleneck node, its presence is essential, if it runs out of energy then the route will be invalid. Again, a longer chain will result in a large transmission delay. 2.2 Distributed data fusion algorithm Rules can be a sensor network topology is equivalent to an image, to obtain a wavelet transform is applied to wireless sensor networks in the distributed data fusion. Research in this area have made some initial results, the following were introduced to his. 2.2.1 Rules of the network situation Servetto first study the distributed implementation of wavelet transform, and used to solve the wireless sensor network broadcast problem. A. University of Southern California Ciancio further research in wireless sensor networks in the distributed data fusion algorithm, the introduction of lifting transform, the rules based on lifting wavelet transform network distributed data fusion algorithms ( I am an expert from steelslittingmachine.com, while we provides the quality product, such as steel bar cutting machine , metal slitting cutters, slitting and rewinding machine,and more.
Related Articles -
steel bar cutting machine, metal slitting cutters,
|