An Overview on the Application of Graph Neural Networks in Wireless Networks

He, Shiwen; Xiong, Shaowen; Ou, Yang; Zhang, Jian; Wang, Jiaheng; Huang, Yongming; Zhang, Yaoxue

doi:10.1109/ojcoms.2021.3128637

Cited by 86 publications

(37 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…HW , W (l−1) , x (l−1) is permutation equivariant. Specifically, we should discuss Algorithm 1 and the beamforming vector update operation (9) whether have PE property. Let x (l) , W (l) be the outputs of…”

Section: Proposition 2 Given Matrix a Hw Inputs X And W For A Permut...mentioning

confidence: 99%

“…HW , W (l−1) , x (l−1) is permutation equivariant. While for the beamforming vector update operation (9), given the output Ξ T x (l) , take Ξ T q (l) as input, we have…”

Section: Proposition 2 Given Matrix a Hw Inputs X And W For A Permut...mentioning

confidence: 99%

“…Specifically, for given beamforming vector {w k }, the algorithm solves problem (10) to obtain {q k }. Then, the beamforming vector {w k } is updated using {q k } via (9). The aforementioned solving procedure is implemented iteratively until the objective function (7a) converges, which is illustrated in Fig.…”

Section: A Weighted Sum Rate Maximization Problemmentioning

confidence: 99%

See 2 more Smart Citations

An Unsupervised Deep Unrolling Framework for Constrained Optimization Problems in Wireless Networks

He¹,

Xiong²,

An³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

In wireless network, the optimization problems generally have complex constraints, and are usually solved via utilizing the traditional optimization methods that have high computational complexity and need to be executed repeatedly with the change of network environments. In this paper, to overcome these shortcomings, an unsupervised deep unrolling framework based on projection gradient descent, i.e., unrolled PGD network (UPGDNet), is designed to solve a family of constrained optimization problems. The set of constraints is divided into two categories according to the coupling relations among optimization variables and the convexity of constraints. One category of constraints includes convex constraints with decoupling among optimization variables, and the other category of constraints includes non-convex or convex constraints with coupling among optimization variables. Then, the first category of constraints is directly projected onto the feasible region, while the second category of constraints is projected onto the feasible region using neural network. Finally, an unrolled sum rate maximization network (USRMNet) is designed based on UPGDNet to solve the weighted SR maximization problem for the multiuser ultra-reliable low latency communication system. Numerical results show that USRMNet has a comparable performance with low computational complexity and an acceptable generalization ability in terms of the user distribution.

show abstract

Section: Proposition 2 Given Matrix a Hw Inputs X And W For A Permut...mentioning

confidence: 99%

“…HW , W (l−1) , x (l−1) is permutation equivariant. While for the beamforming vector update operation (9), given the output Ξ T x (l) , take Ξ T q (l) as input, we have…”

Section: Proposition 2 Given Matrix a Hw Inputs X And W For A Permut...mentioning

confidence: 99%

See 1 more Smart Citation

An Unsupervised Deep Unrolling Framework for Constrained Optimization Problems in Wireless Networks

He¹,

Xiong²,

An³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…After several rounds of updates, the graph embedding contains information of the neighbour network multiple-hops away from the node, and can be used to determine the optimal resource allocation policies [18], [19]. Among different GNN architectures, spatial convolutional graph neural network [20] is one of the most widely used architectures in solving the power allocation problems for wireless networks [21]. In the following, we will consider the spatial GNNs and refer them as GNN for short.…”

Section: Introductionmentioning

confidence: 99%

Distributed Graph Neural Networks for Optimizing Wireless Networks: Message Passing Over-the-Air

Gu¹,

She²,

Quan³

et al. 2022

Preprint

View full text Add to dashboard Cite

Distributed power allocation is important for interference-limited wireless networks with dense transceiver pairs. In this paper, we aim to design low signaling overhead distributed power allocation schemes by using graph neural networks (GNNs), which are scalable to the number of wireless links.We first apply the message passing neural network (MPNN), a unified framework of GNN, to solve the problem. We show that the signaling overhead grows quadratically as the network size increases.Inspired from the over-the-air computation (AirComp), we then propose an Air-MPNN framework, where the messages from neighboring nodes are represented by the transmit power of pilots and can be aggregated efficiently by evaluating the total interference power. The signaling overhead of Air-MPNN grows linearly as the network size increases, and we prove that Air-MPNN is permutation invariant. To further reduce the signaling overhead, we propose the Air message passing recurrent neural network (Air-MPRNN), where each node utilizes the graph embedding and local state in the previous frame to update the graph embedding in the current frame. Since existing communication systems send a pilot during each frame, Air-MPRNN can be integrated into the existing standards by adjusting pilot power.Simulation results validate the scalability of the proposed frameworks, and show that they outperform the existing power allocation algorithms in terms of sum-rate for various system parameters.

show abstract

“…In this work, we propose a novel solution to solve the graph topology optimization problem utilizing deep reinforcement learning (DRL), which is called Advantage Actor Critic-Graph Search (A2C-GS). DRL has been demonstrated to achieve superior performance in many scenarios, e.g., designing molecular structures [22], sharing bike scheduling [15] and IP-fiber cross-layer scheduling in network [23], and has also found applications in social recommendation [3] and wireless communication [8]. A2C-GS builds on the generalization power of DRL and consists of three key novel components, namely, a topology verifier to validate the correctness of a generated network topology, a graph neural network (GNN) to efficiently approximate the topology rating, and an RL actor based on A2C [12] to conduct topology search.…”

Section: Introductionmentioning

confidence: 99%

Network Topology Optimization via Deep Reinforcement Learning

Li¹,

Wang²,

Pan³

et al. 2022

Preprint

View full text Add to dashboard Cite

Topology impacts important network performance metrics, including link utilization, throughput and latency, and is of central importance to network operators. However, due to the combinatorial nature of network topology, it is extremely difficult to obtain an optimal solution, especially since topology planning in networks also often comes with management-specific constraints. As a result, local optimization with hand-tuned heuristic methods from human experts are often adopted in practice. Yet, heuristic methods cannot cover the global topology design space while taking into account constraints, and cannot guarantee to find good solutions.In this paper, we propose a novel deep reinforcement learning (DRL) algorithm, called Advantage Actor Critic-Graph Searching (A2C-GS), for network topology optimization. A2C-GS consists of three novel components, including a verifier to validate the correctness of a generated network topology, a graph neural network (GNN) to efficiently approximate topology rating, and a DRL actor layer to conduct a topology search. A2C-GS can efficiently search over large topology space and output topology with satisfying performance. We conduct a case study based on a real network scenario, and our experimental results demonstrate the superior performance of A2C-GS in terms of both efficiency and performance.

show abstract

An Overview on the Application of Graph Neural Networks in Wireless Networks

Cited by 86 publications

References 68 publications

An Unsupervised Deep Unrolling Framework for Constrained Optimization Problems in Wireless Networks

An Unsupervised Deep Unrolling Framework for Constrained Optimization Problems in Wireless Networks

Distributed Graph Neural Networks for Optimizing Wireless Networks: Message Passing Over-the-Air

Network Topology Optimization via Deep Reinforcement Learning

Contact Info

Product

Resources

About