Research On Range-free Localization Algorithm Improvements For Wireless Sensor Networks

Wireless sensor networks (WSN) are composed of plenty of sensor nodes that are deployed in the monitoring field. In many applications of wireless sensor networks, the sensor node location information is critical to the monitoring activities. On the basis of whether to measure the actual distance between nodes in the process of position, the localization algorithm can be classified into two categories, which is range-based and range-free localization algorithm. The range-free localization algorithm uses the estimated distance instead of the metrical distance to locate the unknown nodes. Compared with the range-based localization algorithm, the range-free location algorithm has the advantages of low cost, small power consumption, simple hardware and so on, so it attracts more and more attention in recent years. In this paper, we choose the DV-Hop localization algorithm, which is the most widely used range-free localization algorithm and pointed out the main reason for the error. To overcome the disadvantage we propose two improved DV-Hop algorithm, one uses a normalized weighted scheme and the other one calculates the error twice to amend localization error. In this paper, the main work and research results can be summarized as follows:(1)This dissertation provided a thorough survey of the principle, the characteristics and the taxonomy for wireless sensor networks localization system. Through extensive investigation and research, introduce several typical localization algorithms with the ideas of range-based and range-free. In the last, a comprehensive summary was brought forward for the localization environment, the error source, the advantages and disadvantages of those localization algorithms.(2)This paper introduces the DV-Hop localization algorithm, which is one of the most widely used range-free algorithms. In addition, we present the overall plan of traditional DV-Hop localization algorithm designing and discuss every function in detail. We also make a depth research in each step to find the objective and subjective factors may cause errors. Some of the main reason such as one-hop distance difficult to correctly reflect the whole network information and was inaccurate in multi-hop situation and since it uses straight distance to replace bend distance biggish error exists and so on. This research hopes to provide the foundation for the improved algorithm in the future.(3)To overcome the disadvantage we propose two improved DV-Hop algorithm. The first algorithm proposes a comprehensive consideration of all beacon nodes average one-hop distance and introducing weights corrected results, and the improved algorithm also use beacon nodes as unknown nodes calculating error and use the error optimizing accuracy. The second algorithm calculates the error twice to amend the average hop distance value and beacon node localization error region. With such improvement programs is proposed in this paper we hope to make the randomly distributed nodes have a better positioning result when using DV-Hop localization algorithm.(4)Finally, the improved algorithm proposes in paper is tested by software. Theoretical analysis and simulation results show that:compared with the traditional DV-Hop algorithm, the two improved can improve the positioning accuracy effectively without additional hardware overhead. And the proposed algorithm has better locating performance in locating precision and precision stability, achieve the expected goal.