Graph Embedding-Based Wireless Link Scheduling With Few Training Samples

A GNN-Based Supervised Learning Framework for Resource Allocation in Wireless IoT Networks

“…• Graph embedding: Distance with quantization is taken as the node and edge features for graph embedding. The embedding feature of nodes is learned by a 3-layer classifier in a supervised manner as in [20]. • DNN: A 4-layer conventional supervised DNN is adopted, and the channel matrix is taken as the input.…”

“…Besides, a graph-based bipartite matching algorithm was utilized in [18] to obtain the optimal resource block allocation for training the federated learning algorithms in a distributed manner over wireless networks. The combination of the graph theory and the ML technologies has brought a lot of attention to the wireless research community as it benefits from both the graph properties and performance acceleration [19] [20]. A spatial convolution method was proposed in [19] to solve a D2D link scheduling problem, wherein the convolution operation is applied to the density grid which is quantified based on the numbers of transmitters and receivers in each grid.…”

“…The acquisition of a large training dataset in realworld wireless networks is expensive or even impractical. Accordingly, a graph embedding method with a multi-layer classifier was proposed in [20] to address this issue, where each D2D pair is represented by a low-dimensional vector with distance based features from itself and its neighbors, and only hundreds of training samples are required. However, the works in [19] and [20] only take distances into consideration and they are not compatible with channel information, which may lead to performance degradation in scenarios with small scale fading.…”

“…Accordingly, a graph embedding method with a multi-layer classifier was proposed in [20] to address this issue, where each D2D pair is represented by a low-dimensional vector with distance based features from itself and its neighbors, and only hundreds of training samples are required. However, the works in [19] and [20] only take distances into consideration and they are not compatible with channel information, which may lead to performance degradation in scenarios with small scale fading. Graph neural networks (GNNs) that have been proven to be successful in a wide range of applications including computer vision, natural language processing and chemistry [21], can effectively exploit non-Euclidean data such as channel state information (CSI).…”

“…• The proposed framework is verified using two resource allocation problems, namely the link scheduling problem in the D2D networks, and the joint channel and power allocation for D2D underlaid cellular networks. The proposed framework is compared to three benchmark schemes: the unsupervised GNN [22], the graph embedding method [20] and the conventional DNN. Simulation results demonstrate that this framework outperforms the benchmark designs and maintains a stable end performance with various system settings and network scales.…”