Combining Lyapunov Optimization With Evolutionary Transfer Optimization for Long-Term Energy Minimization in...

Area of Science:

• Wireless communication systems
• Optimization techniques
• Energy efficiency in networks

Background:

• Intelligent Reflecting Surface (IRS)-aided systems face challenges with time-varying channels and stochastic data arrivals.
• Ensuring queue stability and maximizing energy consumption in dynamic environments requires sophisticated optimization.
• Real-time decision-making is crucial for efficient operation in short time slots.

Purpose of the Study:

• To develop a method for jointly optimizing phase-shift coefficients and transmit power in IRS-aided systems.
• To maximize long-term energy consumption for mobile devices while guaranteeing queue stability.
• To address the challenges of dynamic environments and real-time decision-making.

Main Methods:

• Lyapunov optimization is employed to decouple long-term stochastic problems into deterministic, sequential time-slot problems.
• Evolutionary Transfer Optimization (ETO) is developed to solve per-time-slot optimization problems efficiently.
• A metric identifies similar past optimization problems to transfer optimal solutions, creating a high-quality initial population for current problems.

Main Results:

• The proposed LETO method effectively ensures queue stability by avoiding future information in deterministic optimization.
• LETO accelerates the search process through evolutionary transfer, enabling real-time decisions in short time slots.
• Experimental results demonstrate the effectiveness of LETO compared to other algorithms.

Conclusions:

• LETO provides an effective solution for optimizing IRS-aided communication systems under dynamic conditions.
• The combination of Lyapunov optimization and ETO offers a robust approach for balancing energy efficiency and queue stability.
• The method's ability to leverage past solutions enhances real-time decision-making capabilities.