Study On Underground Colliery Personnel Locating Technology Based On Wi-Fi

Underground environments of colliery are very complex. Coal miners are often threatened by accidents from roof collapse, explosion, flooding, fire and so on. Especially in the event of sudden accident, it is possible to accurately obtain location information of coal miners and give guidance to underground personnel to escape and rescue after the disaster, which reduces casualties and has practical significance to improve the safety of underground personnel. At present, experts and scholars at home and abroad have done a lot of research on underground colliery precise locating technique.Due to the fact that Android platform has the characteristic of supporting reuse and replace components, open source WebKit engine, rich development environment, high security and simple development, it can develop rapidly. Therefore, this paper designed the coal miners locating system based on Android platform, which realized information uploading, information receiving, information querying and personnel locating.This paper compares the current several common wireless locating techniques. Due to the fact that wireless network can be constructed and maintained easily and low-costly, especially that it can still work after accidents happened, locating techniques based on Wi-Fi have been widely used in the underground communication of colliery. Therefore, this paper selects wireless Wi-Fi technique for personnel precise locating, which is more suitable for large scale colliery. By comparing fingerprint locating algorithm based on RSSI and trilateration survey locating algorithm, the former’ locating accuracy is better than the latter, and is more suitable for underground personnel precise locating.This paper designs an underground locating algorithm named AftK-R convex based on the advantages of range-based and range-free algorithm. The signal strength is filtered by Kalman filter algorithm, which removes the noise and keeps from the divergence or loss of the signal. According to the shortage of NN algorithm, which cannot find the true nearest neighbor node of the unknown node, the paper puts forward set of the nearest neighbor candidates. What’s more, the paper designs for convex programming algorithm, and constructs a mathematical model with influence degree, influence factor and relative impact factor, by which we can work out the coordinates of the unknown node and locating precisely.Through a lot of experiment and simulation, it is proved that the AftK-R convex sets locating algorithm’ deviation is smaller than the fingerprint algorithm, and greatly improved personnel locating accuracy. The AftK-R convex sets algorithm can solve the problem of coal miners precise locating.