JOINT OPTIMIZATION OF COMPUTATION AND COMMUNICATION POWER IN MULTI-USER MASSIVE MIMO SYSTEMS.

Area of Science:

• Wireless Communication Systems
• Signal Processing
• Energy Efficiency in Telecommunications

Background:

• Massive MIMO and millimeter wave technologies are crucial for 5G, but increase baseband processing demands and power consumption.
• Conventional energy efficiency analyses often overlook computation power, leading to inaccurate assessments in massive MIMO systems.

Purpose of the Study:

• To formulate a joint optimization problem for computation and communication power in multi-user massive MIMO systems with partially-connected RF structures.
• To investigate the impact of computation power on the energy efficiency of massive MIMO systems.

Main Methods:

• Developed a joint optimization framework for computation and communication power.
• Derived an upper bound for energy efficiency considering system constraints.
• Proposed a suboptimal solution using baseband and RF precoding matrices.
• Introduced the Optimized Hybrid precOding with computation and commuNication powEr (PHONE) algorithm.

Main Results:

• Energy efficiency decreases with more antennas and RF chains when computation power is included, contrary to previous findings.
• The proposed suboptimal solution effectively approaches the derived upper bound on energy efficiency.
• The PHONE algorithm demonstrates significant improvements in energy and cost efficiencies.

Conclusions:

• Joint optimization of computation and communication power is essential for accurate energy efficiency analysis in massive MIMO.
• The PHONE algorithm offers a practical approach to enhance energy and cost efficiencies in 5G systems.
• Up to 76.59% power saving is achievable in multi-user massive MIMO systems with partially-connected structures.