Answers are grounded in this paper and cite their sources.
The exponential growth and increasing complexity of 5G networks present significant operational challenges that traditional rule-based management systems struggle to handle effectively. However, many existing approaches do not adequately capture traffic uncertainty, which limits their ability to adapt in highly dynamic network environments. To address this, The study proposes an optimized 5G Network Entropy-Guided Cyber Twin architecture to address these limitations. The methodology centers on the application of information-theoretic Shannon entropy to quantify traffic uncertainty across temporal, spatial, and service dimensions. This is executed through a tripartite framework: first, a traffic entropy model is developed to provide a mathematical foundation for uncertainty quantification; second, a Cyber Twin environment is constructed to serve as a high-fidelity virtual replica of the 5G network for safe and efficient policy testing; and third, a Soft Actor-Critic (SAC) reinforcement learning agent is designed to incorporate these entropy measures directly into its reward function. This integration allows the agent to make adaptive, data-driven resource allocation decisions that account for underlying traffic variability. Together, these components enable the joint optimization of resource allocation, power consumption, and system latency. The entropy-guided Cyber Twin framework achieves a total system throughput of 146.8 Mbps through optimized Physical Resource Block (PRB) allocation while maintaining a mean energy efficiency of 0.78 Mbits/Joule by utilizing 75.4% of maximum power capacity. Driven by entropy-augmented state representations within the agent, this predictive resource and power management successfully balances high-performance slicing demands with sustainability across dynamic network conditions.
No takes yet. Share an insight, caveat, or question.
. et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: