Research On Improving DV-Hop Positioning Algorithm In WSN

As a new technology of acquiring and processing information, wireless sensor network(wireless sensor network, WSN) is widely used in all aspects of human life, especially superior in the monitoring and tracking application due to the advantages of a large scale, wide distribution, low power consumption, and self-organization ability. In many wireless sensor network applications, the sensor node location information is particularly significant, which lead to the rising of the node location technology at the historic moment.As a kind of typical positioning algorithm based on the range, DV-Hop localization algorithm has the advantages of low power consumption and low cost, and the positioning accuracy can meet the needs of most applications. This paper deeply studies and analyzes the objective and subjective cause of the positioning error in the DV-Hop localization algorithm, and makes the following improvements:Aiming at the inaccurate problem while calculating the minimum Hop information, this paper proposes an improved DV-Hop algorithm of correcting least the Hop based on RSSI. Using the relationship of RSSI values and the distance to classify and elaborate the beacon node of neighbor nodes directly, improving the accuracy of the beacon node communication within the scope of the hop, which lead to directly proportional to the information between nodes and the distance between nodes. Without increasing hardware cost, the positioning precision is improved effectively.In view of the inaccurate problem of computing the average Hop distance information, an improved DV-Hop algorithm is proposed based on the correction of error weighted average jump distance. The calculation mechanism of the beacon node average jump information is improved through minimizing the mean square error of the estimated distance of beacon node, which fixed the average jumped from the information of the beacon node. The unknown node receives average jump distance of multiple new beacon nodes, taking the summation of the beacon node average jump distance based on the weighted error correction as its average distance, which makes the average jump information reflects the real distribution of the network more accurately. The positioning precision is improved to some extent without increasing hardware cost and communication overhead.With regard to the problem of sensitive and accumulation of the ranging error in the calculation method of trilateration method and maximum likelihood estimation method, this paper proposes a cuckoo difference optimization to improve DV-Hop localization algorithm, which essentially turns the positioning calculation into a group optimization problem. Using the cuckoo algorithm and differential evolution algorithm to parallel optimize with double-population, the algorithm fused the advantages of two kinds of intelligent optimization algorithm, which dynamically rectified the abandoned factor and scaled variation factor randomly at the same time. The cuckoo differential evolution algorithm’s global search ability was enhanced to maintain the population diversity, which made the estimate coordinates closer to the real value. Without any increase in communication overhead, the positioning precision was improved effectively.