IEEE Conference on Information and Communication Technology (CICT)

Hybrid Quantum Machine Learning Model for Power Allocation in NOMA

Tanmoy Nandi, Amit Kumar Baghel, Abhishek Kumar Pathak, Vimal Bhatia

Published Dec 6, 2024• 1 citation

Abstract

Recent advances in quantum machine learning with its inherent properties of superposition, quantum parallelism, and quantum entanglement, have opened doors of possibilities in solving complex resource allocation problems in digital communication technologies. This study investigates power allocation in non-orthogonal multiple access (NOMA) systems using hybrid quantum neural networks (QNN) and compares with the classical neural network. A dataset is prepared for training the models and analysis of performance of the model is presented. Results show that the hybrid QNN achieves similar validation accuracy, while reducing overfitting and maintaining comparable performance in maximizing the sum rate. Despite fewer number of layers and parameters, the hybrid QNN achieves competitive results, demonstrating the potential of quantum-enhanced models in NOMA power allocation tasks.

DOI: 10.1109/cict64037.2024.10899724

Related papers

Digital Precoder-Combiner and Power Allocation Optimization in MU MIMO-NOMA: A Quantum Neural Network Approach

Lia Suci Waliani, Soo Young Shin

This paper proposes a quantum neural network (QNN) to address joint optimization in wireless communication. By utilizing the advantages of quantum entanglement and superposition in quantum computing and machine learning, QNN can solve optimization problems with lower complexity than classical neural networks, due to its parallel processing capabilities. Specifically, this study applies QNN to jointly optimize digital precoder-combiner and power allocation in a multi-user multiple-input multiple-output non-orthogonal multiple access (MU MIMO-NOMA) system. The performance of the QNN is analyzed and presented.

한국통신학회논문지Dec 23, 20241 citation

Quantum Neural Network With Parallel Training for Wireless Resource Optimization

Bhaskara Narottama, Triwidyastuti Jamaluddin, Soo Young Shin

Quantum neural network with parallel training (called PS-QNN) is presented in this study to optimize wireless resource allocation. Instead of sending the whole dataset, each edge only requires to send the statistical parameters of the dataset; hence reducing the dimension of training data. As a particular case, the proposed PS-QNN is utilized to optimize transmit precoding and power allocation in non-orthogonal multiple access with multiple-input and multiple-output antennas (MIMO-NOMA). Compared to the conventional training method, analysis shows that the proposed parallel training yields a lower complexity, while achieving a comparable sum rate compared to conventional method.

IEEE Transactions on Mobile ComputingOct 2, 20239 citations

Quantum Neural Networks for Resource Allocation in Wireless Communications

Bhaskara Narottama, Soo Young Shin

This study exploits a quantum neural network (QNN) for resource allocation in wireless communications. A QNN is presented to reduce time complexity while still maintaining performance. Moreover, a reinforcement-learning- inspired QNN (RL-QNN) is presented to improve the perfor- mance. Quantum circuit design of the QNN is presented to ensure the practical implementation in noisy intermediate-scale quantum (NISQ) computers. For the QNN, the complexity and the number of required qubits are analyzed as well. As a particular use case, the QNN is utilized for user grouping in non-orthogonal multiple access. The results reveal that the QNN schemes have lower complexities and similar performance in terms of the achievable sum rate when compared with that of the classical neural network.

IEEE Transactions on Wireless CommunicationsAug 10, 202167 citations

Federated Quantum Neural Network with Quantum Teleportation for Resource Optimization in Future Wireless Communication

Bhaskara Narottama, Soo Young Shin

The following study introduces FT-QNN, a federated and quantum teleportation –based quantum neural network, utilized to optimize resource allocation for future wireless communications. The proposed FT-QNN consists of edge quantum neural networks (QNNs) and a cloud QNN, while quantum teleportation allows the cloud QNN to obtain the outputs of edge QNNs without requiring prior measurements on the output states, allowing the cloud to process the outputs directly as quantum states. As a particular case to demonstrate its applicability for wireless resource allocation, FT-QNN is then employed to optimize transmit power allocation coefficients in a power domain non-orthogonal multiple access (NOMA)-based system, aiming to maximize the achievable sum-rate. FT-QNN yields lower complexity compared to a distributed QNN scheme without quantum teleportation, while the numerical results also demonstrated that the FT-QNN is capable to achieve a similar sum-rate compared to the scheme without quantum teleportation.

IEEE Transactions on Vehicular TechnologyJan 1, 202329 citations

QNN framework based multiclass classification for downlink NOMA detectors

Hye Yeong Lee, Man Hee Lee, Soo Young Shin

Quantum neural networks (QNNs) have attracted significant attention recently, primarily because of their potential to address complex problems deemed difficult for traditional computational methods. This study explores the viability of QNN in handling multiclass classification tasks in downlink nonorthogonal multiple access (NOMA) frameworks. The investigation includes a design of QNN framework and performance evaluation of a QNN-based NOMA detector, integrating maximum likelihood (ML), successive interference cancellation (SIC), and rotated ML (RML) methods. A QNN framework was configured for all three detectors, and a comparative analysis was conducted in terms of loss, accuracy, and testing across varied signal-to-noise ratio (SNR) levels and power allocation coefficients, considering NOMA-specific characteristics. Furthermore, the computational complexity of each detector was analyzed within the proposed framework.

Journal of Communications and NetworksAug 1, 20250 citations

Hybrid Quantum-Classical Neural Networks for Downlink Beamforming Optimization

Juping Zhang, Gan Zheng, Toshiaki Koike–Akino +2 more

This paper investigates quantum machine learning to optimize the beamforming in a multiuser multiple-input single-output downlink system. We aim to combine the power of quantum neural networks and the success of classical deep neural networks to enhance the learning performance. Specifically, we propose two hybrid quantum-classical neural networks to maximize the sum rate of a downlink system. The first one proposes a quantum neural network employing parameterized quantum circuits that follows a classical convolutional neural network. The classical neural network can be jointly trained with the quantum neural network or pre-trained leading to a fine-tuning transfer learning method. The second one designs a quantum convolutional neural network to better extract features followed by a classical deep neural network. Our results demonstrate the feasibility of the proposed hybrid neural networks, and reveal that the first method can achieve similar sum rate performance compared to a benchmark classical neural network with significantly less training parameters; while the second method can achieve higher sum rate especially in presence of many users still with less training parameters. The robustness of the proposed methods is verified using both software simulators and hardware emulators considering noisy intermediate-scale quantum devices.

IEEE Transactions on Wireless CommunicationsAug 19, 202434 citations