Efficient Adaptive Beacon Deployment Optimization for Indoor Crowd Monitoring Applications

Abstract

The indoor crowd density monitoring system using BLE beacons is one of the effective ways to prevent overcrowded indoor situations. The indoor crowd density monitoring system consists of a mobile application at the user's side and the beacon sensor network as the infrastructure. Since the performance of crowd density monitoring highly depends on how BLE beacons are placed, BLE beacon placement optimization is fundamental research work. This research proposes a beacon deployment method EABeD to incrementally place the beacons adaptively to the latest signal propagation status. Also, EABeD reduces most walking and measurement labor costs by applying Bayesian optimization and the walking distance optimization algorithm. We conducted the placement optimization experiment in the wild environment and compared the results with placements derived by the simulation-based method and people. The result shows that our proposed method can achieve 26.4% higher detection coverage than the simulation-based approach, 23.2% and 5.2% higher detection coverage than the inexperienced person's solution and the expert's solution. As for the labor cost reduction, our proposed method can reduce 90.2% of the walking distance and 74.4% of the optimization time compared with optimization by the dense data gathering method.