Clustered Data Muling in the Internet of Things in Motion

Area of Science:

• Computer Science
• Electrical Engineering
• Network Engineering

Background:

• Unmanned Aerial Vehicles (UAVs) are increasingly used for area monitoring with Sensor Networks (SNs).
• Large-scale deployments in smart cities present challenges for efficient data collection by a single UAV.
• Visiting numerous sensors individually is energy-intensive for both the UAV and the SN.

Purpose of the Study:

• To develop an energy-efficient clustering scheme for hybrid UAV/SN systems.
• To optimize the number of clusters and reduce the UAV's traversal distance.
• To address energy consumption for both the SN and the UAV in smart city applications.

Main Methods:

• Proposed a clustering scheme optimizing SN and UAV energy usage.
• Introduced a method for computing optimal cluster numbers in dense, uniform SNs, complementing k-means.
• Developed an efficient clustering model for general networks, handling orphan nodes and multi-layer optimization.

Main Results:

• Simulations demonstrated significant energy savings for both UAV and SN.
• The proposed clustering scheme effectively reduces the number of required UAV visits.
• The model proved effective in a real-world smart city use-case (Cape Town).

Conclusions:

• The proposed clustering scheme enhances the energy efficiency of UAV-assisted SN monitoring.
• This approach is crucial for the economic viability and sustainability of smart city networks.
• The method provides a robust solution for optimizing hybrid network engineering in diverse scenarios.