A5084461937- Advanced Wireless Communication Techniques
- Advanced MIMO Systems Optimization
- Wireless Communication Networks Research
- Cooperative Communication and Network Coding
- Advanced Wireless Communication Technologies
- PAPR reduction in OFDM
- Plant Stress Responses and Tolerance
- Full-Duplex Wireless Communications
- Plant Gene Expression Analysis
- Face and Expression Recognition
- Blind Source Separation Techniques
- Biofuel production and bioconversion
- Advanced Data Compression Techniques
- Neural Networks and Applications
- Millimeter-Wave Propagation and Modeling
- Energy Harvesting in Wireless Networks
- Genetic Mapping and Diversity in Plants and Animals
- Error Correcting Code Techniques
- Advanced Image and Video Retrieval Techniques
- Plant tissue culture and regeneration
- Photosynthetic Processes and Mechanisms
- Ecology and Conservation Studies
- Speech and Audio Processing
- Explainable Artificial Intelligence (XAI)
- Plant-Microbe Interactions and Immunity
Fudan University
2015-2025
Nanjing Medical University
2024-2025
Jiangsu Province Hospital
2024-2025
Yangtze University
2017-2025
Nanjing Agricultural University
2013-2024
Southwest Jiaotong University
2024
Guangdong Academy of Agricultural Sciences
2022-2024
Second Affiliated Hospital of Xi'an Jiaotong University
2022-2024
Hong Kong Baptist University
2024
Jiangsu University
2024
Modeling and learning of brain activity patterns represent a huge challenge to the brain-computer interface (BCI) based on electroencephalography (EEG). Many existing methods estimate uncorrelated instantaneous demixing EEG signals classify multiclass motor imagery (MI). However, condition uncorrelation does not hold true in practice, because regions work with partial or complete collaboration. This proposes novel method, termed as common Bayesian network (CBN), discriminate MI signals....
In multicell multiuser massive multi-input multi-output (MIMO) systems, pilot contamination degrades the uplink (UL) channel estimation performance. To mitigate effect of contamination, we propose a semiblind method that does not require cell cooperation or statistical information channels. proposed method, first sequentially estimate UL data from different users in target cell. do that, for each user, solve constrained minimization problem to obtain an extracting vector and then use it...
This letter proposes a low-complexity deep-learning-based direction-of-arrival (DOA) estimation method for hybrid massive multiple-input multiple-output (MIMO) system with uniform circular array at the base station. In proposed method, we first input received signal vector into some small deep feedforward networks that are trained offline. Based on outputs of networks, then generate set candidate angles. By selecting optimal one from all angles, finally obtain DOA estimation. Simulation...
Melanoma is considered one of the most lethal skin cancers. However, lesion classification based on deep learning diagnostic techniques a challenging task owing to insufficiency labeled images and intraclass-imbalanced datasets. It thus necessary utilize data augmentation methods generative adversarial networks (GANs) assist help dermatologists reach more accurate decisions. Moreover, insufficient samples can cause low accuracy in model by using medical diagnosis reduce classification. To...
Seed deterioration is poorly understood and remains an active area for research. Seeds of elm (Ulmus pumila L.) were aged at 37 °C above water [controlled treatment (CDT)] various lengths time to assess programmed cell death (PCD) reactive oxygen species (ROS) product in embryonic tissues during a 5 d period. The hallmarks PCD identified the seeds CDT including TUNEL experiments, DNA laddering, cytochrome c (cyt c) leakage enzymatic activities. These analyses indicated that occurred...
We propose a simple sparse channel estimation and tracking method for orthogonal frequency-division multiplexing (OFDM) systems based on dynamic parametric model, where the is parameterized by small number of distinct paths, each characterized path delay gain, all parameters are time varying. In proposed method, we adaptively choose grid estimate iteratively. To further reduce complexity, also algorithm fact that changes in delays over few adjacent OFDM symbols. After physical estimated,...
In orthogonal frequency division multiplexing (OFDM) systems, since virtual subcarriers are not used for transmission, approach of conventional uniformly placed pilot tones is applicable in some situations. this letter, based on nonuniform tone placement, we derive optimal sequences, which can achieve the minimum mean square error least squares estimate multiple-input multiple-output (MIMO) OFDM systems. Simulation results demonstrate effectiveness proposed approach.
This short note presents the derivation of a new priori estimate for Oldroyd-B model.Such an may provide useful information when investigating long-time behaviour macro-macro models, and stability numerical schemes.We show how this can be used as guideline to derive estimates other macroscopic like FENE-P model.
The quest for enhancing the interpretability of neural networks has become a prominent focus in recent research endeavors. Prototype-based have emerged as promising avenue imbuing models with by gauging similarity between image components and category prototypes to inform decision-making. However, these face challenges they share activations during both inference explanation processes, creating tradeoff accuracy interpretability. To address this issue ensure that network achieves high robust...
This letter is concerned with the downlink channel estimation in frequency division duplex (FDD) massive multiple-input multiple-output (MIMO) system. With number of antennas increased, acquiring state information (CSI) becomes complex, thus restricts performance communication systems. A deep learning based algorithm used to estimate CSI without feedback. In proposed method, uplink cluster firstly obtained from receiving signals. Based on data, then estimated by a neural network. Simulation...
Interpretable deep-learning models have received widespread attention in the field of image recognition. However, owing to coexistence medical-image categories and challenge identifying subtle decision-making regions, many proposed interpretable suffer from insufficient accuracy interpretability diagnosing images medical diseases. Therefore, this study a feature-driven inference network (FeaInfNet) that incorporates feature-based reasoning structure. Specifically, local feature masks (LFM)...
Uncertainty quantification in time series prediction is challenging due to the temporal dependence and distribution shift on sequential data. Conformal inference provides a pivotal flexible instrument for assessing uncertainty of machine learning models through sets. Recently, online conformal methods updated thresholds sets by performing gradient descent sequence quantile loss functions. A drawback such that they only use information revealed non-conformity scores via miscoverage indicators...
The establishment of artificial grassland is a good pathway for resolving serious social and economic problems in the Qinghai–Tibet Plateau. Some beneficial indigenous microbes may be used to improve productivity grassland. genome dark septate fungus, Exophiala tremulae CICC2537, was sequenced assembled at chromosome level using PacBio sequencing platform, with assistance Hi-C technique scaffolding, its 3D structures were investigated. size E. 51.903848 Mb, it contains eight chromosomes. A...
Satellite communication has the potential to play a key role in many applications of Internet Things (IoT). In this paper, we consider satellite-based IoT and investigate technology that can improve spectral efficiency. general, one beam satellite systems serves user. To serve multiple users, time division multiplexing or frequency is usually used. propose novel forward-link multiplexed scheme, by which signals different users be transmitted simultaneously using same band. Specifically, at...
Hemoporfin-mediated photodynamic therapy (Hemoporfin-PDT) is a safe and effective treatment modality for port-wine stain (PWS). However, there still no consensus about the influential factors efficacy of treatment. This study investigated associated with Hemoporfin-PDT.We retrospectively analyzed 321 PWS patients who underwent Hemoporfin-PDT at our center from August 2017 to July 2021. The correlation between versus sex, age, location, type PWS, numbers, lesion size were analyzed.The numbers...
Abstract Diabetic cardiomyopathy (DCM) is a common cardiovascular complication of diabetes, which may threaten the quality life and shorten expectancy in diabetic population. However, molecular mechanisms underlying diabetes are not fully elucidated. We analyzed two datasets from Gene Expression Omnibus (GEO). Differentially expressed weighted gene correlation network analysis (WGCNA) was used to screen key genes molecules. Ontology (GO), Kyoto Encyclopedia Genes Genomes (KEGG) enrichment...
Because of excellent capability description local texture, Local Binary Patterns (LBP) have been applied in many areas. In this paper, we enhance the classical LBP method from three aspects for facial expression recognition: image data, extracting features and way combining all these features. At first, adopt wavelet to decomposed images into four kinds frequency which are extracted increase original data. Then extract with a new holistic make more robust. last, order use logical, combine...