| Wireless sensor networks, plastic electronics and human organs are jointly known as the three high-tech industries in future. As the emerging high-tech industries, wireless sensor network is the future direction of the development of high-tech industries, and certain progress has been achieved currently in this area, especially abroad. Wireless sensor network has a broad prospect of application, and it has played critical roles in the following fields: military, defense, industry, agriculture, environment monitoring, physiological medical and so on. Considering the fact that the node location is a prerequisite for the application of wireless sensor networks, the significance of the research on algorithm of node localization is obvious.Wireless sensor network is a kind of network with the self-organizing ability, which is deployed in certain areas, and the communication among the nodes is completed by means of wireless, multiple hops; Sensor nodes are the basic component unit of wireless sensor network, the nodes are divided into position-known nodes (beacon nodes) and position-unknown (unknown nodes). The unknown nodes take advantage of connectivity of the network, relationship between the neighbors and the distance to the beacon nodes to locate their positions. The main evaluation criterions of position are the localization accuracy, the power consumption and so on. According to different criteria, the location algorithms are mainly divided into range-based location, range-free location, centralized location, distributed location and so on.DV-Hop localization algorithm belongs to the range-free location, it is also the research point in the paper. Firstly, we analyze the traditional DV-Hop algorithm detailedly, and carry out a lot of simulation experiments; Then we focus on several of the existing algorithms, such as the DV-Hop algorithm whose beacon nodes are distributed evenly, the DV-Hop algorithm which is based on the feedback of beacon nodes, the weighted DV-Hop algorithm, the DV-Hop algorithm in which the errors of hop jumps is corrected and the D-DV-Hop algorithm with different jumps. Having analyzed their advantages and disadvantages through the simulation results, we put forward the improved ideas as follows: ①Introduce two communication radius for every beacon node: R and 0.5 R, and obtain the node number of the beacon node within one hop scope and 0.5 R range, then draw the proportion of within 0.5 R area. If the unknown node is in the 0.5 R range, the distance is 0.5 times the average hop-size; and if the unknown nodes is between the 0.5 R range and the average hop-size, the distance is the average hop-size; and when the unknown nodes is beyond one hop range, the distance between them is will be got according to the following steps: between their hops minus 1 multiply by hop-size, add the proportion(written for a) of 0.5 R range multiplied by 0.5 hop-size, finally add 1 minus a multiplying by hop-size.②Localization , if unknown nodes obtain only 3 beacon nodes information, then use the three beacon nodes to locate the position of unknown nodes, if can obtain more than 3 beacon nodes information, chooses which different combination for positioning, get different result, the average of the locating results as the final location results.③Use DV- Hop algorithm to compute the estimated position of the unknown nodes, then use Min-Max algorithm calibration the estimated position, if the positioning result of DV- Hop algorithm inside Min-Max algorithms determine the borders, retains the result, if the positioning result of DV- Hop algorithm beyond borders, is used to frame the direction to pull back.Some improvement based on above done a lot of simulation experiments, the location accuracy improved respectively 4.49%, 5.38%, 5.91%, the experiment data indicate that the algorithm can meet the, improve location accuracy. |
PDF Full Text Request