CCSW‐RLNC: A Deep Learning‐Enhanced Color‐Coded Sliding Window Approach for Efficient Network Coding

ABSTRACT This paper focuses on developing an innovative Color‐Coded Sliding Window Random Linear Network Coding (CCSW‐RLNC) strategy, where the sliding window size dynamically adapts based on network conditions, including intermediate node availability, power limitations, channel noise, and multihop transmission characteristics. The system features a special two‐direction mechanism, where data windows are moved from left to right as color priority is shifted from right to left. The central color has the highest weightage to ensure data integrity. The enhancement of this approach is through a deep learning model, which optimizes the window parameters and adaptation process. The primary targets are to enhance energy efficiency and maximize throughput through the optimal synchronization of encoder and decoder and to achieve minimum latency. The color‐coding scheme serves as both a data representation system and a priority management system, enabling the more accurate reconstruction of original data packets. This new hybrid, combining RLNC, color‐coding, and deep learning adaptability addresses realistic network limitations while enhancing system performance and overall reliability.