MWLP-DP: Mobile war-fighters location prediction for dark phase in Internet of Battlefield Things

In a battlefield, a base station is always interested to track the real-time location of moving war-fighters and other armaments because of its strategic importance. The base station estimates their real-time location by using the collected localization-essentials information like hop-counts matrix of the nodes. This information is collected at the base station in active phase through satellite communication among the base station and the sink nodes. In the harsh environment of the battlefield, if this satellite communication breaks accidentally or purposefully then the whole network is disconnected from the base station, and the network enters into the dark phase. In this dark phase, the base station wants to localize the mobile war-fighters and other armaments also. It motivates us to propose a novel method MWLP-DP: mobile war-fighters location prediction for dark phase in Internet of Battlefield Things to localize the moving war-fighters. It works in two stages: localization of sink nodes (LSN) and localization of blind nodes (LBN). In LSN, by using geometric Brownian motion (GBM), the mobility pattern of the sink nodes is analyzed. The stochastic nature of GBM needs repeated trials to ascertain the locations of the sink nodes. Further, computational efforts are reduced by finding the optimum counts of repeated trials of GBM using the central limit theorem for localizing the sink nodes. In LBN, the localization of remaining nodes (blind nodes) is inspired by the DV-Hop using the location of sink nodes obtained in LSN stage. Further, the erroneous distance estimations between the node pairs are optimized through linear programming to reduce localization error. The simulation shows that the proposed MWLP-DP localizes the sink nodes and blind nodes in the dark phase with 89% and 64% average accuracy, respectively.