Distributed State-Saturated Recursive Filtering Over Sensor Networks Under Round-Robin Protocol.

Area of Science:

• Control Systems Engineering
• Networked Systems
• Stochastic Systems

Background:

• State saturation is a practical constraint in real-world systems.
• Round-robin (RR) protocols manage network access in sensor networks to reduce load.
• Distributed filtering is crucial for processing data across multiple sensor nodes.

Purpose of the Study:

• To design a distributed recursive filter for stochastic discrete time-varying systems.
• To address challenges posed by state saturation and RR protocols in sensor networks.
• To establish an upper bound for the filtering error covariance and provide explicit filter parameters.

Main Methods:

• Utilizing matrix difference equations to derive the filtering error covariance upper bound.
• Employing a matrix simplification technique to handle sparse sensor network topologies.
• Designing filter parameters to minimize the derived upper bound.

Main Results:

• An upper bound for the filtering error covariance was established.
• Explicit expressions for the filter parameters were derived.
• The proposed filtering scheme was validated through a practical example.

Conclusions:

• The developed distributed recursive filter effectively handles state saturation and RR protocols.
• The method provides a robust approach for filtering in complex sensor networks.
• The explicit design parameters facilitate practical implementation and performance optimization.