Distributed Learning Fractal Algorithm for Optimizing a Centralized Control Topology of Wireless Sensor Network Based...

Area of Science:

• Computer Science
• Network Engineering
• Algorithm Design

Background:

• Wireless Sensor Networks (WSNs) are crucial for smart environments, enabling remote operation of devices via the internet.
• Quality of service routing presents a significant challenge in WSNs, particularly for surveillance applications.
• Existing WSNs require efficient topology control for optimal performance and data sharing.

Purpose of the Study:

• To propose a Distributed Learning Fractal Algorithm (DFLA) for designing efficient control topologies in WSNs.
• To enhance network efficiency and reliability in WSNs through optimized node connectivity.
• To facilitate real-time monitoring of environmental parameters for smart city applications.

Main Methods:

• Generation of the Hilbert fractal using L-systems production rules to create a space-filling curve.
• Modeling the optimization of centralized WSN control topology.
• Development and application of the DFLA to identify highly reliable links between micro controller units (MCUs).

Main Results:

• The DFLA effectively identifies optimal node pairs for reliable WSN links.
• A software-defined network (SDN) with strong mobility was proposed, adaptable to varying node counts.
• Fractal routing in WSNs demonstrated effectiveness in real-time monitoring with 16 to 64 sensors.

Conclusions:

• The proposed DFLA enhances WSN efficiency and reliability for real-time monitoring.
• The fractal routing approach is suitable for large-scale WSN deployments in smart city contexts.
• This research contributes to the development of more efficient and adaptable WSNs for future smart environments.