Energy Efficiency vs Spectral Efficiency Tradeoff in MIMO Systems with Hardware Non-Linearities

Massive multiple-input multiple-output (MIMO) technology is one of the pivotal enablers for future wireless communication systems that offer high spectral efficiency (SE) and reliability due to its increased array and diversity gains. However, energy efficiency (EE) and sustainability of the system is one of the crucial parameters for next-generation wireless networks, necessitating careful consideration of the SE vs. EE tradeoff. Accurate evaluation of a system's EE requires a comprehensive end-to-end (E2E) model that accounts for practical hardware impairments, such as finite-resolution analog-to-digital converters (ADCs) / digital-to-analog converters (DACs) and non-linear power amplifiers (PAs), which degrade system performance. In this work, we develop an E2E model for MIMOorthogonal frequency division multiplexing (OFDM) system with hybrid beamforming considering both the partially connected (PC) and fully connected (FC) architectures. The behavior of finite-resolution DACs and ADCs is characterized as additive quantization noise model (AQNM), while the behavior of PAs is characterized through the Bussgang decomposition. Our analysis offers valuable insights into the influence of DAC/ADC resolution and PA input back-off (IBO) on the EE of the system, highlighting key trade-offs and performance considerations for system design.