Indeterministic Data Collection in UAV-Assisted Wide and Sparse Wireless Sensor Network.

Area of Science:

• Wireless communication networks
• Robotics and control systems
• Data science and analytics

Background:

• Internet of Things (IoT) applications necessitate efficient sensor data collection.
• Unmanned aerial vehicles (UAVs) offer a viable solution for data collection in areas lacking ground communication infrastructure.
• Existing methods struggle with the challenge of indeterministic data availability in wireless sensor nodes (SNs).

Purpose of the Study:

• To maximize the amount of effective sensing data collected by UAVs in a wide and sparse wireless sensor network.
• To address the challenge of unknown effective data status at individual sensor nodes.
• To jointly optimize UAV trajectories, SN serve scheduling, and UAV-SN association.

Main Methods:

• Modeling the data collection as an optimization problem with indeterministic effective indicators.
• Introducing an effectiveness probability prediction model to handle unknown data status.
• Proposing a two-layer modified knapsack algorithm with iterative feasibility problem resolution for optimal packing strategy.

Main Results:

• The proposed scheme significantly increases the sum of effective data blocks collected.
• Completion time for collecting a specific ratio of effective data is reduced by nearly 30%.
• Demonstrated remarkable advantages in efficient data collection for UAV-assisted WSNs.

Conclusions:

• The developed optimization strategy effectively tackles the challenge of indeterministic data collection in UAV-assisted WSNs.
• The two-layer modified knapsack algorithm provides an efficient solution for complex optimization problems.
• This research offers a significant improvement in data collection efficiency and time reduction for sparse WSNs.