A5036831868- Advanced MIMO Systems Optimization
- Cooperative Communication and Network Coding
- Software-Defined Networks and 5G
- Opportunistic and Delay-Tolerant Networks
- Caching and Content Delivery
- Cognitive Radio Networks and Spectrum Sensing
- Complex Network Analysis Techniques
- Wireless Networks and Protocols
- Reinforcement Learning in Robotics
- Network Security and Intrusion Detection
- Distributed Control Multi-Agent Systems
- Advanced Wireless Network Optimization
- Sparse and Compressive Sensing Techniques
- Human Mobility and Location-Based Analysis
- Wireless Signal Modulation Classification
- Wireless Communication Networks Research
- Network Traffic and Congestion Control
- Advanced Fluorescence Microscopy Techniques
- Opinion Dynamics and Social Influence
- Traffic Prediction and Management Techniques
- Traffic control and management
- Privacy-Preserving Technologies in Data
- Anomaly Detection Techniques and Applications
- Photoacoustic and Ultrasonic Imaging
- Internet Traffic Analysis and Secure E-voting
Zhejiang University
2016-2025
Zhejiang Lab
2016-2025
Tsinghua University
2020-2025
Chinese Institute for Brain Research
2021-2025
McGovern Institute for Brain Research
2024
Fiberhome Technology Group (China)
2022-2023
South China Normal University
2023
China Medical University
2023
Fourth Affiliated Hospital of China Medical University
2023
Fuzhou University
2023
5G cellular networks are assumed to be the key enabler and infrastructure provider in ICT industry, by offering a variety of services with diverse requirements. The standardization is being expedited, which also implies more candidate technologies will adopted. Therefore, it worthwhile provide insight into techniques as whole examine design philosophy behind them. In this article, we try highlight one most fundamental features among revolutionary era, i.e., there emerges initial intelligence...
Time series prediction can be generalized as a process that extracts useful information from historical records and then determines future values. Learning long-range dependencies are embedded in time is often an obstacle for most algorithms, whereas LSTM solutions, specific kind of scheme deep learning, promise to effectively overcome the problem. In this article, we first give brief introduction structure forward propagation mechanism LSTM. Then, aiming at reducing considerable computing...
Network slicing is born as an emerging business to operators by allowing them sell the customized slices various tenants at different prices. In order provide better-performing and costefficient services, network involves challenging technical issues urgently looks forward intelligent innovations make resource management consistent with users' activities per slice. that regard, deep reinforcement learning (DRL), which focuses on how interact environment trying alternative actions reinforcing...
Software defined Internet of Things (SD-IoT) networks profit from centralized management and interactive resource sharing, which enhances the efficiency scalability applications. But with rapid growth in services applications, they are vulnerable to possible attacks face severe security challenges. Intrusion detection has been widely used ensure network security, but classical methods usually signature-based or explicit-behavior-based fail detect unknown intelligently, makes it hard satisfy...
Network slicing is a key technology in 5G communications system. Its purpose to dynamically and efficiently allocate resources for diversified services with distinct requirements over common underlying physical infrastructure. Therein, demand-aware resource allocation of significant importance network slicing. In this paper, we consider scenario that contains several slices radio access base stations share the same (e.g., bandwidth or slots). We leverage deep reinforcement learning (DRL)...
Computing is a critical driving force in the development of human civilization. In recent years, we have witnessed emergence intelligent computing, new computing paradigm that reshaping traditional and promoting digital revolution era big data, artificial intelligence, internet things with theories, architectures, methods, systems, applications. Intelligent has greatly broadened scope extending it from on data to increasingly diverse paradigms such as perceptual cognitive autonomous...
Abstract A fundamental challenge in fluorescence microscopy is the photon shot noise arising from inevitable stochasticity of detection. Noise increases measurement uncertainty and limits imaging resolution, speed sensitivity. To achieve high-sensitivity beyond shot-noise limit, we present DeepCAD-RT, a self-supervised deep learning method for real-time suppression. Based on our previous framework DeepCAD, reduced number network parameters by 94%, memory consumption 27-fold processing time...
Holistic understanding of physio-pathological processes requires noninvasive 3D imaging in deep tissue across multiple spatial and temporal scales to link diverse transient subcellular behaviors with long-term physiogenesis. Despite broad applications two-photon microscopy (TPM), there remains an inevitable tradeoff among spatiotemporal resolution, volumes, durations due the point-scanning scheme, accumulated phototoxicity, optical aberrations. Here, we harnessed concept synthetic aperture...
With the cellular networks becoming increasingly agile, a major challenge lies in how to support diverse services for mobile users (MUs) over common physical network infrastructure. Network slicing is promising solution tailor match such service requests. This paper considers system with radio access (RAN)-only slicing, where infrastructure split into slices providing computation and communication functionalities. A limited number of channels are auctioned across scheduling slots MUs...
Traffic learning and prediction is at the heart of evaluation performance telecommunications networks attracts a lot attention in wired broadband networks. Now, benefiting from big data cellular networks, it becomes possible to make analyses one step further into application level. In this paper, we first collect significant amount application-level traffic network operators. Afterward, with aid "big data," comprehensive study over modeling framework traffic. Our results solidly demonstrate...
Although the research on traffic prediction is an established field, most existing works have been carried out traditional wired broadband networks and rarely shed light cellular radio access (CRANs). However, with explosively growing demand for access, there urgent need to design a traffic-aware energy-efficient network architecture. In order realize such design, it becomes increasingly important model predictability theoretically discuss networking practice technically. of that...
Network slicing aims to efficiently provision diversified services with distinct requirements over the same physical infrastructure. Therein, in order allocate resources across slices, demand-aware inter-slice resource management is of significant importance. In this letter, we consider a scenario that contains several slices radio access network base stations share (e.g., bandwidth or slots). We primarily leverage advantage actor-critic (A2C), one typical deep reinforcement learning (DRL)...
Network slicing promises to provision diversified services with distinct requirements in one infrastructure. Deep reinforcement learning (e.g., deep Q-learning, DQL) is assumed be an appropriate algorithm solve the demand-aware interslice resource management issue network by regarding varying demands and allocated bandwidth as environment state action, respectively. However, allocating a finer resolution usually implies larger action space, unfortunately DQL fails quickly converge this case....
With the development of deep learning (DL), natural language processing (NLP) makes it possible for us to analyze and understand a large amount texts. Accordingly, we can achieve semantic communication in terms joint source channel coding over noisy with help NLP. However, existing method realize this goal is use fixed transformer NLP while ignoring difference information contained each sentence. To solve problem, propose new system based on Universal Transformer. Compared traditional...
Modern communications are usually designed to pursue a higher bit-level precision and fewer bits while transmitting message. This article rethinks these two major features introduces the concept advantage of semantics that characterizes new kind semantics-aware communication framework, incorporating both semantic encoding problem. After analyzing underlying defects existing techniques, we establish confidence-based distillation mechanism for joint semantics-noise coding (JSNC) problem...
Abstract Fluorescence imaging with high signal-to-noise ratios has become the foundation of accurate visualization and analysis biological phenomena. However, inevitable noise poses a formidable challenge to sensitivity. Here we provide spatial redundancy denoising transformer (SRDTrans) remove from fluorescence images in self-supervised manner. First, sampling strategy based on is proposed extract adjacent orthogonal training pairs, which eliminates dependence speed. Second, designed...
Recent works have validated the possibility of improving energy efficiency in radio access networks (RANs), achieved by dynamically turning on/off some base stations (BSs). In this paper, we extend research over BS switching operations, which should match up with traffic load variations. Instead depending on dynamic loads are still quite challenging to precisely forecast, firstly formulate variations as a Markov decision process. Afterwards, order foresightedly minimize consumption RANs,...
As an inevitable trend of future fifth generation (5G) networks, software-defined architecture has many advantages in providing centralised control and flexible resource management. However, it is also confronted with various security challenges potential threats emerging services technologies. the focus network security, intrusion detection systems (IDSs) are usually deployed separately without collaboration. They unable to detect novel attacks limited intelligent abilities, which hard meet...
Network slicing (NS) management devotes to providing various services meet distinct requirements over the same physical communication infrastructure and allocating resources on demands. Considering a dense cellular network scenario that contains several NS multiple base stations (BSs), it remains challenging design proper real-time inter-slice resource strategy, so as cope with frequent BS handover satisfy fluctuations of service requirements. In this paper, we propose formulate challenge...
The integrated satellite-terrestrial network is promising to provide global broadband communication service. However, the long propagation delay of links will lead high when users access Internet via satellites. In this paper, we investigate cooperative multilayer edge caching in reduce delay, which base station cache, satellite and gateway cache cooperatively content service for ground users. We first propose three-layer model network, based on analyze retrieving process derive hit...
Semantic communication has witnessed a great progress with the development of natural language processing (NLP) and deep learning (DL). Although existing semantic technologies can effectively reduce errors in interpretation, most these solutions adopt fixed bit length structure, along rigid transmission scheme, which is inefficient lacks scalability when faced different meanings signal-to-noise ratio (SNR) conditions. In this paper, we explore impact adaptive lengths on coding (SC) under...