Combined Channel Estimation with Interference Suppression in CPSS

Area of Science:

• Cyber-Physical Systems
• Machine Learning
• Wireless Communication

Background:

• Cyber-physical systems (CPS) face complex electromagnetic environments due to integrated social characteristics.
• Low-power nodes in cyber-physical-social systems (CPSS) require efficient solutions for computational complexity, interference, and fading.

Purpose of the Study:

• To develop a machine learning-based channel estimation scheme for CPSS.
• To address simultaneous challenges of computational complexity, interference suppression, and transmission fading.

Main Methods:

• Proposed a novel channel estimation scheme integrating frequency-domain interference suppression (K-means) and time-domain noise cancellation (KNN).
• Combined K-means algorithm for frequency-domain channel impulse response (CIR) interference suppression.
• Utilized K-nearest neighbor (KNN) algorithm for time-domain CIR noise cancellation.

Main Results:

• The proposed scheme demonstrates lower computational complexity compared to traditional methods.
• Simulation results confirm superior performance over existing channel estimation schemes.
• The approach effectively meets CPSS requirements in complex electromagnetic settings.

Conclusions:

• The integrated machine learning scheme offers an efficient solution for channel estimation in CPSS.
• This method significantly improves performance and reduces complexity for resource-constrained nodes.
• The study provides a viable approach for reliable communication in challenging wireless environments.