On Multi-Parameter Optimization and Proactive Reliability in 5G and Beyond Cellular Networks

Area of Science:

• Telecommunications Engineering
• Network Reliability
• Stochastic Modeling

Background:

• Ultra-dense heterogeneous cellular networks in 6G face increasing cell outage risks due to complex issues like misconfigurations and hardware failures.
• Current Autonomous Network Function (ANF)-based fault detection is reactive, identifying issues only after service quality degrades.

Purpose of the Study:

• To develop a proactive fault prediction framework for next-generation wireless networks.
• To shift network management from reactive fault detection to proactive fault anticipation and mitigation.

Main Methods:

• Introduction of a novel Discrete-Time Markov Chain (DTMC)-based stochastic framework.
• Modeling network reliability dynamics to forecast cell transitions to suboptimal states.
• Quantifying fault arrival effects and identifying sensitive parameters impacting performance.

Main Results:

• The DTMC framework accurately predicts the timing and probable causes of network faults.
• The model quantifies the fraction of time the network remains in a degraded state.
• Sensitive network parameters contributing to performance degradation are identified.

Conclusions:

• The proposed framework enables proactive fault management, significantly reducing cell outage time.
• This approach enhances the overall reliability and resilience of next-generation wireless networks.
• It provides a crucial capability for maintaining high service quality in complex network environments.