.png){kind=icon}
.png){kind=icon}
Return
Route Optimization Based on Quantum-Weighted Long- and Short-Term Memory Networks
.png){kind=icon}
PRE
AI
A
H
Z
M
DOI：10.23919/cje.2025.00.159.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The increasing complexity and heterogeneity of modern networks, spurred by the proliferation of diverse devices and communication protocols, have posed substantial challenges to traditional routing optimization methods. These methods often fail to adapt to the dynamic and multifaceted nature of heterogeneous networks, leading to suboptimal performance in terms of latency, throughput, and resource utilization. To address these challenges, this paper introduces a novel approach that leverages quantum-weighted long short-term memory (QW-LSTM) networks for routing optimization in heterogeneous networks. By integrating quantum computing principles with deep learning, our method enhances the ability of LSTM networks to capture complex temporal dependencies and nonlinear relationships in network traffic data. Experimental evaluations conducted on simulated and real-world heterogeneous networks demonstrate the superiority of the QW-LSTM approach over conventional routing algorithms and standard LSTM-based methods. The results show significant improvements in key performance metrics, including reduced latency, increased throughput, and enhanced adaptability to network changes. Moreover, the quantum-weighted mechanism contributes to faster convergence during training and better generalization to unseen network conditions.
Keywords:
Heterogeneous networks
Quantum-weighted long short-term memory
Heterogeneous networks
Routing optimization
Routing optimization
Quantum computing
Quantum computing
Deep learning
Deep learning
Network traffic prediction
Network traffic prediction
Journal
C
IF:
3
Papers:
62
Citations:
1.7K
Organization
Cited Papers