A5103208814- Advanced Bandit Algorithms Research
- Mobile Crowdsensing and Crowdsourcing
- Video Surveillance and Tracking Methods
- Solar Radiation and Photovoltaics
- Energy Load and Power Forecasting
- Auction Theory and Applications
- Privacy-Preserving Technologies in Data
- Software-Defined Networks and 5G
- Advanced MIMO Systems Optimization
- Image and Video Quality Assessment
- Visual Attention and Saliency Detection
- Quantum-Dot Cellular Automata
- Cavitation Phenomena in Pumps
- Mobile Ad Hoc Networks
- Building Energy and Comfort Optimization
- Refrigeration and Air Conditioning Technologies
- Engineering and Agricultural Innovations
- Vehicular Ad Hoc Networks (VANETs)
- Wireless Networks and Protocols
- Opportunistic and Delay-Tolerant Networks
- Image Enhancement Techniques
- Atmospheric aerosols and clouds
- Advanced Wireless Network Optimization
- Digital Games and Media
- Advanced Neural Network Applications
Nanjing University
2020-2024
Zhengzhou University of Aeronautics
2023
In-band Network Telemetry (INT) is proposed to detect networks via injecting specific probes collect the hop-by-hop metadata within programmable switches. But there exist multiple challenges conducting INT at Computing Power Network, such as control decisions of different frequencies, and unforeseeable query workloads. In this study, we formulate an online non-linear time-varying integer programming problem that aims maximize overall quality service through both frequency selection workload...
Rapid growth of air conditioning (AC) has caused large effect on safe and economical operation power grid especially in summer. The research the AC load monitoring is great significance for improving stability grid. Non-intrusive (NILM) gaining attention due to many attractive features, such as low cost, low-complexity easy promotion. A NILM model using sampling rate smart meter data proposed this paper. Firstly, preliminary recognition performed detect all possible events. Secondly, several...
A variety of modern AI products essentially require raw user data for training diverse machine learning models. With the increasing concern on privacy, federated learning, a decentralized framework, enables privacy-preserving models by iteratively aggregating model updates from participants, instead data. Since all i.e., mobile devices, need to transfer their local concurrently and over edge networks, network is easily overloaded, leading high risk transmission failures. Although previous...
Many crowdsourcing platforms are emerging, leveraging the resources of recruited workers to execute various outsourcing tasks, mainly for those computing-intensive video analytics with high quality requirements. Although profit each platform is strongly related feedback, due uncertainty on diverse performance and conflicts interest over platforms, it non-trivial determine dispatch tasks maximum benefits. In this paper, we design a decentralized mechanism Crowd Crowdsourcing denoted as <sup...
Deploying deep convolutional neural network (CNN) to perform video analytics at edge poses a substantial system challenge, as running CNN inference incurs prohibitive cost in computational resources. Model partitioning, promising approach, splits CNNs and distributes them multiple devices closer proximity each other for serial inferences, however, it causes considerable cross-edge delay transmitting intermediate feature maps. To overcome this we present ResMap, new framework that...
A classified irradiance forecast approach for solar PV prediction is proposed based on wavelet decomposition. The Daubechies chose to decompose the series measured in plant into approximate component and detailed component. trend variability of are estimated respectively two components. Then all available data labeled according features extracted from In end, multiple models built trained adapt different labels. simulation results show effectiveness approach.
For intro-hour solar power forecasting approaches with sky images, the computing for cloud motion is very necessary. In this paper, a displacement estimation method images based on phase correlation theory proposed. By Fourier transforming, image matrix in frequency domain obtained and then cross-power spectrum of two adjacent can be calculated. According to inverse transform spectrum, achieved. To improve practical application ability approach, noise signal analyzed filtering exclude...
The detection of objects via neural networks plays a key role in various video analytics, but consumes huge resources. Due to the limited computing capability at edges, such real-time detections should be precisely used for that need most attention. Unfortunately, as target keep moving, existing systems fail consider both distribution and object movements over regions, tracking mechanisms are easily affected by background content. Therefore, we propose spatial temporal with attention...
Network visualization is essential for network operators to diagnose ongoing failures and understand the quality of network. In-Band Telemetry (INT) supports by inserting P4 switch state information (e.g., queue length, hop latency, link utilization) into specific INT packets. In order achieve network-wide coverage, paths packets need be delicately designed ensure non-overlapping, high performance, low overhead. However, existing path planning solutions ignore capturing dynamic status fail...
In order to determine the optimal structural parameters of a plastic centrifugal pump in framework an orthogonal-experiment approach, numerical study related flow field has been performed using CFX. The thickness S, outlet angle β2, inlet β1, wrap angle, and diameter D1 splitter blades have considered as variable factors, shaft power efficiency evaluation indices. Through parametric analysis, relative importance influence each parameter on index obtained, leading following combinations: β1...
The detection of objects via neural networks plays a key role in various video analytics, but consumes huge resources. Due to the limited on-device computing capability, such real-time detections should be precisely used for that need most attention. Unfortunately, as target keep moving, existing systems fail conduct adaptive over multiple regions video, and tracking mechanisms are easily affected by background contents. Therefore, we propose design attention-aware increase quality those...
Provisioning In-band Network Telemetry as a service for INT-based and INT follow-up applications in an online manner suffers multiple challenges, including control decisions of different frequencies, resources provisioning coupled over time, the unforeseeable query workloads. To overcome these this study formulates non-linear time-varying integer programming problem that maximizes overall quality through frequency selection workloads distribution. Specifically, learning algorithm is proposed...
This article studies the data distribution technique of unmanned aerial vehicle (UAV) multi-hop networks as communication channels in work scenario. A Game-Strategybased Distribution Protocol(GSDP) was proposed. From experimental simulation results, it can be seen that this method has shown good results reducing transmission delay and balancing energy load network nodes.
To investigate the technical guidance of high temperature compression heat pumps, fundamental physical and thermodynamic properties potential refrigerants are listed compared. Typical R152a, R1233zd (E), R1234yf, R1234ze(E), R1234ze(Z), R717, R718, R744, so on. The natural refrigerants, especially water, show large prospects in high-temperature industrial application. Meanwhile, main system parameters pump, including pressure ratio, compressor volume, discharge temperature, unit power...