A5101726860- Optical Network Technologies
- Advanced Photonic Communication Systems
- Photonic and Optical Devices
- Vehicle emissions and performance
- Air Quality and Health Impacts
- Advanced Fiber Laser Technologies
- Air Quality Monitoring and Forecasting
- Advancements in Solid Oxide Fuel Cells
- Semiconductor Lasers and Optical Devices
- Error Correcting Code Techniques
- Electronic and Structural Properties of Oxides
- Neural Networks and Reservoir Computing
- Advanced Optical Network Technologies
- Advanced Wireless Communication Techniques
- Polymer crystallization and properties
- Advanced Fiber Optic Sensors
- Recycling and utilization of industrial and municipal waste in materials production
- Iterative Learning Control Systems
- biodegradable polymer synthesis and properties
- Catalytic Processes in Materials Science
- Advanced materials and composites
- Catalysis and Oxidation Reactions
- Energy Efficiency and Management
- PAPR reduction in OFDM
- Concrete and Cement Materials Research
University of Electronic Science and Technology of China
2015-2025
First Affiliated Hospital Zhejiang University
2024-2025
Xi’an Jiaotong-Liverpool University
2025
University of Liverpool
2025
First Affiliated Hospital of Anhui Medical University
2024
Anhui Medical University
2024
Zhejiang University
2024
Zhejiang Lab
2024
Xijing University
2024
Joint Research Center
2024
We propose a transfer learning assisted deep neural network (DNN) method for optical-signal-to-noise ratio (OSNR) monitoring and realize fast remodel to response various system parameters changing, e.g. optical launch power, residual chromatic dispersion (CD) bit rate. By transferring the hyper-parameters of DNN at initial stage, we can channel variation with fewer training set size calculations save consumptions. For feature extraction processing, use amplitude histograms received 56-Gb/s...
A new digital control solution to motion problems is proposed. It based on a unique active disturbance rejection concept, where the, disturbances are estimated using an extended state observer (ESO) and compensated in each sampling period. The dynamic compensation reduces the system approximately double integrator which can be easily controlled nonlinear proportional-derivative (PD) controller. proposed disturbance-rejection controller (ADRC) consists of ESO PD designed without explicit...
Dam safety is considerably affected by seepage, and uplift pressure a key indicator of dam seepage. Thus, making accurate predictions trends can improve hazard forecasting. In this study, convolutional neural network, (CNN)-gated recurrent (GRU)-based prediction model was developed, which included the CNN model’s feature extractability GRU learnability for time series correlation data. Then, performance verified using as an example. The results showed that mean absolute errors (MAEs) CNN-GRU...
Using LBR-370 numerical control lathe, high speed cutting was applied to AZ31 magnesium alloy. The influence of parameters on microstructure, surface roughness and machining hardening were investigated by using the methods single factor orthogonal experiment. results show that have an important effect machine hardening. depth stress layer, present a declining tendency with increase also augment feed rate. Moreover, we established prediction model roughness, which has guidance actual process
Chromatic dispersion-enhanced signal-signal beating interference (SSBI) considerably affects the performance of intensity-modulation and direct-detection (IM/DD) fiber transmission systems. For recovering optical fields from received double sideband signals after propagating through IM/DD systems, Gerchberg-Saxton (G-S) iterative algorithms are promising, which, however, suffers slow convergence speeds local optimization problems. In this paper, we propose a multi-constraint algorithm (MCIA)...
Energy consumption of machine tool has drawn wide attention in recent years. The additional load losses tools are great importance for investigating the energy because those account 15-20% cutting power and may even be up to nearly 30% our researches. For lack adequate understanding characteristics past, coefficient, defined as ratio power, was regarded a constant while spindle speed unchanged. However, it is discovered practical measurements that not so. In this paper, proposes an model...
DSP-enhanced intensity-modulation direct-detection (IM/DD) systems can support up to 56 Gb/s over 100 km signal transmissions at C-band. To achieve higher data rates and longer transmission distances, we propose data-aided iterative algorithm (DIA) decision-directed DIA (DD-DIA) digitally mitigate signal-signal beating interference (SSBI) without requiring any modifications physical layer structures. utilizes pilot symbols with uniformly spaced insertions relax the modified Gerchberg-Saxton...
Biomass-derived materials can help develop efficient, environmentally friendly and cost-effective catalysts, thereby improving the sustainability of hydrogen production. Herein, we propose a simple method to produce nickel molybdenum composites decorated spent coffee grounds (SCG) as an efficient catalyst, SCG(200)@NiMo, for electrocatalytic The porous carbon supporter derived form SCG provided larger surface, prevented aggregation during high temperature pyrolysis, optimized electronic...
Supported by advanced digital signal processing algorithms and application specific integrated circuits, coherent receivers in elastic optical networks will be capable of measuring link impairments real time. Specifically, can work as soft performance monitors. Optical spectra usually contain rich information about links have been exploited to assist failure detection identification. However, acquiring needs the deployment numerous spectrum analyzers. Instead, received signals are easy...
Recent tightening of particulate matter (PM) emission standards for heavy-duty engines has spurred the widespread adoption diesel filters (DPFs), which need to be regenerated periodically remove trapped PM. The total impact DPFs therefore depends not only on their filtering efficiency during normal operation, but also emissions and frequency regeneration events. We performed active (parked driving) passive regenerations two vehicles (HDDVs), report chemical composition these events, as well...
We have proposed and demonstrated a transfer learning (TL)-assisted deep network (DNN) for nonlinear distortion compensation in optical side-band PAM-4 modulation direct-detection transmission. Since there exists partial correlation of distortions, we can the parameters trained DNN to target model speed up remodeling reduce complexity. conduct experiments demonstrate effectiveness scheme Nyquist transmissions. The required iterations or train size with TL be less than half that retraining...
We propose and experimentally demonstrate a method of optical signal-to-noise ratio (OSNR) monitoring modulation format identification (MFI) using binarized convolutional neural network (B-CNN) in coherent receiver. The proposed technique automatically extracts OSNR dependent features from the signals' ring constellation maps. A group schemes including nine quadrature amplitude (QAM) formats are selected as transmission signals. experimental results show that MFI accuracy can reach 100%...
We propose a cascaded neural network (NN) to simultaneously identify the modulation formats and monitor optical-signal-to-noise ratio (OSNR). In second-level network, it is single deep NN (DNN) rather than multiple sub-networks, which makes architecture more compact can save resource for real implementation. However, since data set constituted from all formats, universality be guaranteed but not accuracy complexity. To accelerate estimation process improve accuracy, we introduce transfer...
The severe band-limited effect resulted from the low-cost optical transceiver increases channel memory length and number of taps equalizers. Besides, interaction fiber dispersion square-law detection introduce nonlinear distortions in intensity modulation direct-detection (IM/DD) transmission systems. serious degrade performance bring challenges to current equalizers for low-complexity implementation. In this paper, we propose a trellis-compression maximum likelihood sequence estimation...
Deep brain stimulation (DBS) targeting the lateral habenula (LHb) is a promising therapy for treatment-resistant depression (TRD) but its clinical effect has been variable, which can be improved by adaptive DBS (aDBS) guided neural biomarker of symptoms. Existing biomarkers, however, cannot simultaneously track slow and fast symptom dynamics, do not sufficiently respond to parameters, lack neurobiological interpretability, hinder their use in developing aDBS. We conducted study on one TRD...