Grover's Search-Inspired Quantum Reinforcement Learning for Massive MIMO User Scheduling

TL;DR

The paper tackles the challenge of efficient user scheduling in downlink mMIMO with large antenna/user counts and CSI overhead. It introduces a Grover's search-inspired Quantum Reinforcement Learning framework that integrates a gate-based quantum circuit with reinforcement learning, using amplitude amplification to explore the combinatorial scheduling space. The approach leverages a proportional fairness objective and an ergodic-rate approximation $S_t \approx \log_2\left(1+\mathbb{E}[\mathrm{SINR}_t]\right)$, with memory updates $\bar{S}_\gamma(t+1) = (1-\omega)\bar{S}_\gamma(t) + \omega \hat{S}_\gamma(t)$ guiding long-term performance. Experimental results show convergence within 500 epochs and higher average sum rates than CNN and QNN across varied $A$, $T$, and SNR, demonstrating scalability and robustness for future quantum-enabled wireless networks.

Abstract

The efficient user scheduling policy in the massive Multiple Input Multiple Output (mMIMO) system remains a significant challenge in the field of 5G and Beyond 5G (B5G) due to its high computational complexity, scalability, and Channel State Information (CSI) overhead. This paper proposes a novel Grover's search-inspired Quantum Reinforcement Learning (QRL) framework for mMIMO user scheduling. The QRL agent can explore the exponentially large scheduling space effectively by applying Grover's search to the reinforcement learning process. The model is implemented using our designed quantum-gate-based circuit, which imitates the layered architecture of reinforcement learning, where quantum operations act as policy updates and decision-making units. Moreover, the simulation results demonstrate that the proposed method achieves proper convergence and significantly outperforms classical Convolutional Neural Networks (CNN) and Quantum Deep Learning (QDL) benchmarks.

Figures (6)

• Figure 1: Grover's search-based quantum circuit architecture.
• Figure 2: Grover-based Quantum Circuit (5 qubits).
• Figure 3: Training convergence of the Grover-inspired QRL agent showing steady reward growth and stabilization.
• Figure 4: Comparison of sum-rate performance versus number of users for QNN, CNN, and QRL under A=32, SNR=20 dB, showing the scalability of QRL in user number dimension.
• Figure 5: Sum-rate versus number of BS antennas for QNN, CNN, and QRL under $T=6$, SNR=20 dB, showing QRL’s scalability in antenna number dimension.