Review on the Research Progress of Machine Learning in Network Coding

With the rapid development of the information industry, the global data traffic continues to rise, in the face of a large number of data transmission needs, the improvement of communication link throughput is particularly important. Network coding is an effective means to improve the efficiency of network transmission, and it shows the obvious advantage of throughput in multicast, unicast, broadcast, and other networks. However, the traditional network coding scheme will produce a large coding calculation and storage cost, and cannot dynamically adjust the optimal coding strategy according to environmental changes, so relevant researchers have been looking for optimization schemes. Machine learning has excelled in numerous domains, it can learn from data to enhance processing efficiency and decision-making capabilities, presenting new opportunities for optimizing network coding. This paper reviews the progress of machine learning in network coding and points out the current challenges and potential directions for future investigation to provide a basis for further exploration.