5G Radio Link Failure Prediction using Quantum Machine Learning

The next generation of mobile communication networks, 5G, is expected to provide ultra-high-speed, low-latency, and high-reliability wireless communication services. However, radio link failures (RLFs) in 5G networks can degrade network performance and user experience. We present a novel approach for forecasting RLFs in 5G networks using quantum machine learning (QML) in this paper. We employ quantum long short-term memory (QLSTM) with different embedding techniques like Angle embedding, amplitude Embedding, and QAOA embedding in quantum circuits to predict RLFs. We have also merged Quantum circuits to the output of classical Long Short-Term Memory (LSTM) to study any improvement over the existing LSTM model. Our approach involves training a QLSTM model using quantum circuits with different embedding techniques to capture the underlying patterns of RLFs. We evaluate our proposed approach on a real-world dataset and the experimental results indicate that our QML approach outperforms classical machine learning techniques in terms of accuracy and loss.