Practical approximate indoor nearest neighbour locating with crowdsourced RSSIs

In the indoor space, finding nearest neighbour is of great importance in location-based services. Received Signal Strength Indication (RSSI) has received much attention due to its simplicity and compatibility with existing hardware, which been widely used for localization. Existing search methods are based on real walking distance, need ground survey labor work measure many distances. Crowdsourcing a low-cost efficient way collect RSSI space without expert surveyors designated coordinates collection points. The crowdsourced RSSIs can reflect location objects RSSI-based localization method simplistic as it needs low hardware requirements, deployment cost no distance. So we study how RSSIs. To address this problem, propose graph interval weights, called I-graph, connect represent topology space. We also construct tree index D-tree, weights efficiently. novel distance metric relationship between locate RSSIs, devise algorithms pruning strategies computing query. demonstrate efficiency effectiveness proposed solution through extensive experiments two data sets.