Wireless multicast networks, especially in the domains of IoT, 5G, and e-health, are experiencing a growing adoption of random linear network coding (RLNC). Nevertheless, the exponential growth of data can lead to congestion problems. This paper presents a review on novel technique called singular value decomposition (SVD) for the identification of network congestion. When SVD combines with statistical concepts like linear regression can to effectively handle large datasets and conduct comprehensive data analytics. The svd improves network performance and reliability by proactively identifying and mitigating congestion. Additionally, it improves the efficiency of data transmission and delivery in different application scenarios.