http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Statistical Quantum Federated Learning for NOMA Power Allocation
Bhaskara Narottama,Triwidyastuti Jamaluddin,Soo Young Shin 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.6
This study employs the statistical Quantum Federated Learning (sQFL) to optimize NOMA power allocation. Com-pared to the existing Federated Learning (FL), sQFL does not require other edges to perform neural network inferences. The other edge only required to transmit the statistical information to the cloud.
Statistical Quantum Federated Learning for NOMA Power Allocation
Narottama Bhaskara,Jamaluddin Triwidyastuti,Soo Young Shin 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.6
This study employs the statistical Quantum Federated Learning (sQFL) to optimize NOMA power allocation. Compared to the existing Federated Learning (FL), sQFL does not require other edges to perform neural network inferences. The other edge only required to transmit the statistical information to the cloud.
Utilizing Quantum Circuit-Based Loss Function for Quantum Machine Learning
Bhaskara Narottama,Soo Young Shin(신수용) 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.2
This study considers a quantum circuit-based loss function (QLF) for quantum machine learning (QML). Similar to the classical machine learning (cML), QML requires loss function for the training process. Finally, this study proposes a two-tier machine learning (ML) scheme that employs QLF.
Integration of Quantum Variational Circuit and SVD for Precoding Optimization
Bhaskara Narottama,Triwidyastuti Jamaluddin,Soo Young Shin 한국통신학회 2022 한국통신학회 학술대회논문집 Vol.2022 No.2
In this study, integration of quantum variational circuit and singular-value-decomposition-based precoding (QV-SVD) is presented to optimize MIMO-NOMA precoding. Considering imperfect channel information, the objective of the optimization is to maximize the achievable sum rate.
Quantum-Inspired Evolutionary Algorithms for NOMA User Pairing
Bhaskara Narottama,Denny Kusuma Hendraningrat,Soo Young Shin 한국통신학회 2022 ICT Express Vol.8 No.1
This work proposes the utilization of the quantum-inspired evolutionary algorithm (QEA) for user pairing in non-orthogonal multiple access (NOMA). By exploiting quantum concepts such as superposition, it obtained a user pairing solution that approximates the highest achievable sum rate. Moreover, elitist QEA (E-QEA) is proposed to further enhance the performance by eliminating the risk of losing the best solution of the current iteration in the next iteration. Simulation result demonstrates that E-QEA and QEA yield higher average achievable sum rates compared to random user pairing.