Energy-Efficient Optimization for Mobile Edge Computing With Quantum Machine Learning

We investigate the joint optimization problem of stochastic computation offloading, content caching strategy, and dynamic resource allocation to maximize the energy efficiency of mobile edge computing in Internet-of-Things. Specifically, we propose a quantum deep reinforcement learning algorithm to exponentially increase the caching learning speed and content caching delivery efficiency in multi-dimensional continuous and large action spaces. Furthermore, we utilize the modified Grover's algorithm with faster computation time to improve the processing efficiency and data-content retrieval for transition quantum state probabilities. The numerical results show that our proposed quantum machine learning scheme significantly outperforms other benchmarks in terms of energy-efficiency maximization subject to transmission power, energy consumption, and transmission latency.