Summary of Learning Load Balancing with Gnn in Mptcp-enabled Heterogeneous Networks, by Han Ji et al.
Learning Load Balancing with GNN in MPTCP-Enabled Heterogeneous Networks
by Han Ji, Xiping Wu, Zhihong Zeng, Chen Chen
First submitted to arxiv on: 22 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Systems and Control (eess.SY)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel graph neural network (GNN)-based model is proposed for load balancing (LB) in hybrid light fidelity (LiFi) and wireless fidelity (WiFi) networks, also known as heterogeneous networks (HetNets). The current development of HetNets is restricted by the existing transmission control protocol (TCP), which limits user equipment (UE) to connecting one access point (AP) at a time. To overcome this limitation, the proposed GNN-based model tackles the LB problem for multipath TCP (MPTCP)-enabled HetNets with a partial mesh topology. This graph-based approach embeds channel state information and data rate requirements as node features, while LB solutions are deemed edge labels. Compared to traditional deep neural networks (DNNs), the proposed GNN-based model exhibits strengths in interpreting complex network topologies and handling various APs and UEs with a single trained model. Simulation results show that the proposed model achieves near-optimal throughput within a 11.5% gap, while reducing inference time by 4 orders of magnitude. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine having multiple internet connections at once! A new way to balance these connections is being developed for a special type of network called hybrid networks (HetNets). Right now, devices can only connect to one internet source at a time. To fix this problem, scientists have created a new model using a technique called graph neural networks. This model helps HetNets work better by considering the different paths that data takes through the network. Compared to other methods, this new approach is more efficient and effective. |
Keywords
» Artificial intelligence » Gnn » Graph neural network » Inference
