Efficient paths determining strategies in Mobile Crowd-sensing Networks with AI-based sensors forwarding data

Selecting a sufficient number of mobile users to collect and upload collected data to the server is a critical issue in the Mobile Crowd-sensing Networks (MCN). Previous studies have assumed that mobile users upload collected data over cellular networks, which could cause heavily burden to users. This work focus on how to forward collected data by pre-deployed wireless sensors which can fuse collected data and operate as edge nodes. Specifically, given the reward paid to each mobile user depends on the time he spends on data collection and uploading, this work investigates the problem how to select Points of Interest (PoIs) and edge nodes for participants who already have schedules with the objective of minimizing the total reward paid to all participants. We boil down this problem to the problem of determining path for each participant which connects participant’s initial location to PoI, then to an edge node and finally to participant’s destination. We formulate it as Paths determination with Cost Minimization problem. We can prove that this problem is an NP-Complete problem. Considering that the sensors acted as edge nodes which may be rechargeable or have limited energy, we design three heuristic algorithms: Minimum Cost Algorithm (MCA), Minimum Cost with Energy Consideration Algorithm (MCECA), and Energy Balance Algorithm (EBA) to address this problem. Finally, we conduct extensive simulations to validate the efficiency of the proposed algorithms. The results demonstrate that MCA finds paths for users with lower cost, while EBA effectively balances the energy consumption of edge nodes.