arXiv subjects

Ruijin Sun

Publications and source records attributed to Ruijin Sun.

30 records · Page 2Linked to original sources

arXiv2024

Knowledge-Driven Deep Learning Paradigms for Wireless Network Optimization in 6G

In the sixth-generation (6G) networks, newly emerging diversified services of massive users in dynamic network environments are required to be satisfied by multi-dimensional heterogeneous resources. The resulting large-scale complicated network optimization problems are beyond the capability of model-based theoretical methods due to the overwhelming computational complexity and the long processing time. Although with fast online inference and universal approximation ability, data-driven deep learning (DL) heavily relies on abundant training data and lacks interpretability. To address these issues, a new paradigm called knowledge-driven DL has emerged, aiming to integrate proven domain knowledge into the construction of neural networks, thereby exploiting the strengths of both methods. This article provides a systematic review of knowledge-driven DL in wireless networks. Specifically, a holistic framework of knowledge-driven DL in wireless networks is proposed, where knowledge sources, knowledge representation, knowledge integration and knowledge application are forming as a closed loop. Then, a detailed taxonomy of knowledge integration approaches, including knowledge-assisted, knowledge-fused, and knowledge-embedded DL, is presented. Several open issues for future research are also discussed. The insights offered in this article provide a basic principle for the design of network optimization that incorporates communication-specific domain knowledge and DL, facilitating the realization of intelligent 6G networks.

cs.NI↗

arXiv2023

Knowledge-Driven Multi-Agent Reinforcement Learning for Computation Offloading in Cybertwin-Enabled Internet of Vehicles

By offloading computation-intensive tasks of vehicles to roadside units (RSUs), mobile edge computing (MEC) in the Internet of Vehicles (IoV) can relieve the onboard computation burden. However, existing model-based task offloading methods suffer from heavy computational complexity with the increase of vehicles and data-driven methods lack interpretability. To address these challenges, in this paper, we propose a knowledge-driven multi-agent reinforcement learning (KMARL) approach to reduce the latency of task offloading in cybertwin-enabled IoV. Specifically, in the considered scenario, the cybertwin serves as a communication agent for each vehicle to exchange information and make offloading decisions in the virtual space. To reduce the latency of task offloading, a KMARL approach is proposed to select the optimal offloading option for each vehicle, where graph neural networks are employed by leveraging domain knowledge concerning graph-structure communication topology and permutation invariance into neural networks. Numerical results show that our proposed KMARL yields higher rewards and demonstrates improved scalability compared with other methods, benefitting from the integration of domain knowledge.

cs.LG↗

arXiv2023

Knowledge-Driven Resource Allocation for D2D Networks: A WMMSE Unrolled Graph Neural Network Approach

This paper proposes an novel knowledge-driven approach for resource allocation in device-to-device (D2D) networks using a graph neural network (GNN) architecture. To meet the millisecond-level timeliness and scalability required for the dynamic network environment, our proposed approach incorporates the deep unrolling of the weighted minimum mean square error (WMMSE) algorithm, referred to as domain knowledge, into GNN, thereby reducing computational delay and sample complexity while adapting to various data distributions. Specifically, the aggregation and update functions in the GNN architecture are designed by utilizing the summation and power calculation components of the WMMSE algorithm, which leads to improved model generalization and interpretabiliy. Theoretical analysis of the proposed approach reveals its capability to simplify intricate end-to-end mappings and diminish the model exploration space, resulting in increased network expressiveness and enhanced optimization performance. Simulation results demonstrate the robustness, scalability, and strong performance of the proposed knowledge-driven resource allocation approach across diverse communication topologies without retraining. Our findings contribute to the development of efficient and scalable wireless resource management solutions for distributed and dynamic networks with strict latency requirements.

eess.SY↗

arXiv2023

Scalable Resource Management for Dynamic MEC: An Unsupervised Link-Output Graph Neural Network Approach

Deep learning has been successfully adopted in mobile edge computing (MEC) to optimize task offloading and resource allocation. However, the dynamics of edge networks raise two challenges in neural network (NN)-based optimization methods: low scalability and high training costs. Although conventional node-output graph neural networks (GNN) can extract features of edge nodes when the network scales, they fail to handle a new scalability issue whereas the dimension of the decision space may change as the network scales. To address the issue, in this paper, a novel link-output GNN (LOGNN)-based resource management approach is proposed to flexibly optimize the resource allocation in MEC for an arbitrary number of edge nodes with extremely low algorithm inference delay. Moreover, a label-free unsupervised method is applied to train the LOGNN efficiently, where the gradient of edge tasks processing delay with respect to the LOGNN parameters is derived explicitly. In addition, a theoretical analysis of the scalability of the node-output GNN and link-output GNN is performed. Simulation results show that the proposed LOGNN can efficiently optimize the MEC resource allocation problem in a scalable way, with an arbitrary number of servers and users. In addition, the proposed unsupervised training method has better convergence performance and speed than supervised learning and reinforcement learning-based training methods. The code is available at \url{https://github.com/UNIC-Lab/LOGNN}.

eess.SY↗

arXiv2023

Digital Twin-Assisted Knowledge Distillation Framework for Heterogeneous Federated Learning

In this paper, to deal with the heterogeneity in federated learning (FL) systems, a knowledge distillation (KD) driven training framework for FL is proposed, where each user can select its neural network model on demand and distill knowledge from a big teacher model using its own private dataset. To overcome the challenge of train the big teacher model in resource limited user devices, the digital twin (DT) is exploit in the way that the teacher model can be trained at DT located in the server with enough computing resources. Then, during model distillation, each user can update the parameters of its model at either the physical entity or the digital agent. The joint problem of model selection and training offloading and resource allocation for users is formulated as a mixed integer programming (MIP) problem. To solve the problem, Q-learning and optimization are jointly used, where Q-learning selects models for users and determines whether to train locally or on the server, and optimization is used to allocate resources for users based on the output of Q-learning. Simulation results show the proposed DT-assisted KD framework and joint optimization method can significantly improve the average accuracy of users while reducing the total delay.

cs.LG↗

arXiv2022

SigT: An Efficient End-to-End MIMO-OFDM Receiver Framework Based on Transformer

Multiple-input multiple-output and orthogonal frequency-division multiplexing (MIMO-OFDM) are the key technologies in 4G and subsequent wireless communication systems. Conventionally, the MIMO-OFDM receiver is performed by multiple cascaded blocks with different functions and the algorithm in each block is designed based on ideal assumptions of wireless channel distributions. However, these assumptions may fail in practical complex wireless environments. The deep learning (DL) method has the ability to capture key features from complex and huge data. In this paper, a novel end-to-end MIMO-OFDM receiver framework based on \textit{transformer}, named SigT, is proposed. By regarding the signal received from each antenna as a token of the transformer, the spatial correlation of different antennas can be learned and the critical zero-shot problem can be mitigated. Furthermore, the proposed SigT framework can work well without the inserted pilots, which improves the useful data transmission efficiency. Experiment results show that SigT achieves much higher performance in terms of signal recovery accuracy than benchmark methods, even in a low SNR environment or with a small number of training samples. Code is available at https://github.com/SigTransformer/SigT.

cs.NI↗

arXiv2022

On-Demand Resource Management for 6G Wireless Networks Using Knowledge-Assisted Dynamic Neural Networks

On-demand service provisioning is a critical yet challenging issue in 6G wireless communication networks, since emerging services have significantly diverse requirements and the network resources become increasingly heterogeneous and dynamic. In this paper, we study the on-demand wireless resource orchestration problem with the focus on the computing delay in orchestration decision-making process. Specifically, we take the decision-making delay into the optimization problem. Then, a dynamic neural network (DyNN)-based method is proposed, where the model complexity can be adjusted according to the service requirements. We further build a knowledge base representing the relationship among the service requirements, available computing resources, and the resource allocation performance. By exploiting the knowledge, the width of DyNN can be selected in a timely manner, further improving the performance of orchestration. Simulation results show that the proposed scheme significantly outperforms the traditional static neural network, and also shows sufficient flexibility in on-demand service provisioning.

eess.SY↗

arXiv2019

Crystal structures and the sign reversal Hall resistivity in iron-based superconductors Lix(C3H10N2)0.32FeSe (0.15<x<0.4)

We report the crystal structure, superconductivity and normal state properties of two iron-based materials, Li0.15(C3H10N2)0.32FeSe(P-4) and Lix(C3H10N2)0.32FeSe(P4/nmm, 0.25<x<0.4) with superconducting transition temperature from 40~46K. The determined crystal structures revealed a coupling between Li concentration and the oritation of 1,2-Diaminopropane molecules within the hyper expanded FeSe layers. Further fitting on resistivity in terms of the Lawence-Doniach model suggests the two superconductors belong to the quasi-two dimonsional system. With increasing temperature, a differences in crystal structures and doping levels. First principle calculations revealed the increase in FeSe layer diatance will restruct the Fermi surface and generate a new hole pocket around Gamma point in the Brillouin Zone. Our findings support that the increase in two dimensionalities will leads to a temperature induced Lifshitz transition in electron doped FeSe superconductors.

cond-mat.supr-con↗

arXiv2016

Destination-aided Wireless Power Transfer in Energy-limited Cognitive Relay Systems

This paper considers an energy-limited cognitive relay network where a secondary transmitter (ST) assists to forward the traffic from a primary transmitter (PT) to a primary receiver (PR), in exchange for serving its own secondary receiver (SR) in the same frequency. The multiple-antenna ST is assumed to be energy-constrained and powered by both information flow from source (PT) and dedicated energy streams from destinations (PR and SR), which is called destination-aided wireless power transfer (DWPT) scheme. Then, the relay processing matrix, cognitive beamforming vector and power splitter are jointly de- signed to maximize the rate of secondary users under the energy causality constraint and the constraint that the demanded rate of primary users is satisfied. For the perfect channel information state (CSI) case, by adopting semi-definite relax (SDR) technique and Charnes-Cooper transformation, the global optimal solution is given. To reduce the complexity, matrix decomposition, zero forcing (ZF) scheme, and dual method are jointly employed to derive a suboptimal solution. For the imperfect CSI case, S- procedure is used to transform the worst-case robust problem into a tractable semi-definite program (SDP). Simulation results reveal that our proposed DWPT scheme is greatly preferred for both perfect and imperfect CSI cases when ST is close to PR/SR.

cs.IT↗

arXiv2016

Transceiver Design for Cooperative Non-Orthogonal Multiple Access Systems with Wireless Energy Transfer

In this paper, an energy harvesting (EH) based cooperative non-orthogonal multiple access (NOMA) system is considered, where node S simultaneously sends independent signals to a stronger node R and a weaker node D. We focus on the scenario that the direct link between S and D is too weak to meet the quality of service (QoS) of D. Based on the NOMA principle, node R, the stronger user, has prior knowledge about the information of the weaker user, node D. To satisfy the targeted rate of D, R also serves as an EH decode-and-forward (DF) relay to forward the traffic from S to D. In the sense of equivalent cognitive radio concept, node R viewed as a secondary user assists to boost the performance of D, in exchange for receiving its own information from S. Specifically, transmitter beamforming design, power splitting ratio optimization and receiver filter design to maximize node R rate are studied with the predefined QoS constraint of D and the power constraint of S. Since the problem is non-convex, we propose an iterative approach to solve it. Moreover, to reduce the computational complexity, a zero- forcing (ZF) based solution is also presented. Simulation results demonstrate that, both two proposed schemes have better performance than the direction transmission.

cs.IT↗

arXiv2016

Two new parent compounds for FeSe-based superconducting phases

It is well established that the occurrence of superconductivity in iron pnictides is closely related to the tetragonal to orthorhombic structural and antiferromagnetic (AFM) phase transitions. It, however, has not been clear whether the same scenario is appropriate for iron chalcogenide counterparts due to the absence of parent compounds for the latter family of superconductors. Here, we report the synthesis and structure determination of two novel phases in ethylenediamine intercalated FeSe, one is tetragonal and the other orthorhombic in room temperature, which can be stabilized with neutral spacer layers. Both phases can be regarded as the parent compounds for superconductivity as they are non-superconducting (non-SC) in pristine form and superconducting (SC) with Tc up to 38K and 46K, respectively after Na doping, and the switch between SC and no-SC is reversible. Moreover, the two non-SC parent compounds show no evidence of long-range magnetic ordering down to 2K, only with dynamic spin fluctuations at low temperatures, suggesting that no competition between SC and AFM ordering. Our results reveal that undoped iron selenides are quantum paramagnetic in ground state, implying that they are distinct from their pnictide counterparts in pairing mechanism.

cond-mat.supr-con↗

arXiv2015

Transceiver Design to Maximize Sum Secrecy Rate in Full Duplex SWIPT Systems

This letter considers secrecy simultaneous wireless information and power transfer (SWIPT) in full duplex systems. In such a system, full duplex capable base station (FD-BS) is designed to transmit data to one downlink user and concurrently receive data from one uplink user, while one idle user harvests the radio-frequency (RF) signals energy to extend its lifetime. Moreover, to prevent eavesdropping, artificial noise (AN) is exploited by FD-BS to degrade the channel of the idle user, as well as to provide energy supply to the idle user. To maximize the sum of downlink secrecy rate and uplink secrecy rate, we jointly optimize the information covariance matrix, AN covariance matrix and receiver vector, under the constraints of the sum transmission power of FD-BS and the minimum harvested energy of the idle user. Since the problem is non-convex, the log-exponential reformulation and sequential parametric convex approximation (SPCA) method are used. Extensive simulation results are provided and demonstrate that our proposed full duplex scheme extremely outperforms the half duplex scheme.

cs.IT↗