Enhancing Energy Efficiency in Buildings through DNC-Aided SCL-Flip Decoding of Polar Codes

Abstract

In pursuit of enhancing energy efficiency in buildings, this paper proposes a novel approach leveraging Deep Neural Network (DNN)-aided Successive Cancellation List (SCL) decoding of Polar Codes. Polar Codes have demonstrated exceptional error correction capabilities, making them promising candidates for reliable communication systems. However, traditional decoding algorithms such as SCL suffer from high computational complexity, limiting their practical implementation in energy-constrained environments. To address this challenge, we introduce a DNN-aided SCL-Flip decoding technique, where a neural network assists in making informed decisions during the decoding process. By exploiting the inherent structure of Polar Codes and leveraging the learning capabilities of DNNs, our proposed approach achieves significant reductions in decoding complexity while maintaining high error correction performance. This paper presents theoretical foundations, implementation details, and experimental results demonstrating the efficacy of the proposed technique in enhancing energy efficiency in building communication systems.