A5100408977- Error Correcting Code Techniques
- Advanced Memory and Neural Computing
- Coding theory and cryptography
- Mobile Ad Hoc Networks
- Opportunistic and Delay-Tolerant Networks
- Advanced Wireless Communication Techniques
- Distributed Sensor Networks and Detection Algorithms
- Energy Efficient Wireless Sensor Networks
- Cooperative Communication and Network Coding
- Advanced Neural Network Applications
- Caching and Content Delivery
- Green IT and Sustainability
- Neural Networks and Reservoir Computing
- Network Security and Intrusion Detection
- CCD and CMOS Imaging Sensors
- Neural Networks and Applications
- Cryptographic Implementations and Security
- Stochastic Gradient Optimization Techniques
- Advanced Algorithms and Applications
- Distributed Control Multi-Agent Systems
- Security in Wireless Sensor Networks
- Sparse and Compressive Sensing Techniques
- Advanced MIMO Systems Optimization
- Molecular Junctions and Nanostructures
- Target Tracking and Data Fusion in Sensor Networks
Shanghai Children's Medical Center
2025
Shanghai Jiao Tong University
2022-2025
South China Normal University
2022-2024
Xi'an University of Technology
2024
Nanjing University
2017-2024
Nanjing University of Science and Technology
2024
Inner Mongolia Electric Power (China)
2016-2024
Chongqing University of Technology
2023
Tunghai University
2021
Huawei Technologies (France)
2021
In this paper, we study the problem of jointly estimating optical flow and scene from synchronized 2D 3D data. Previous methods either employ a complex pipeline that splits joint task into independent stages, or fuse information in an "early-fusion" "late-fusion" manner. Such one-size-fits-all approaches suffer dilemma failing to fully utilize characteristic each modality maximize inter-modality complementarity. To address problem, propose novel end-to-end framework, called CamLiFlow. It...
Introduction Corticosteroids are used for toxicity management, raising concerns about whether they may affect the anti-leukemic effects of chimeric antigen receptor (CAR)-T cells. Methods and results In this study, we retrospectively analyzed patients (fined two subgroups based on disease burden. Of 75 cases in low burden (LDB) group (MRD < 5%, no extramedullary disease), there was significant difference between use steroids event-free survival (EFS) ( p = 0.21) overall (OS) 0.26),...
ADVERTISEMENT RETURN TO ISSUEPREVArticleNEXTA nanometer-scale galvanic cellWenjie Li, Jorma A. Virtanen, and Reginald M. PennerCite this: J. Phys. Chem. 1992, 96, 16, 6529–6532Publication Date (Print):August 1, 1992Publication History Published online1 May 2002Published inissue 1 August 1992https://pubs.acs.org/doi/10.1021/j100195a005https://doi.org/10.1021/j100195a005research-articleACS PublicationsRequest reuse permissionsArticle Views525Altmetric-Citations68LEARN ABOUT THESE...
Large language models (LLMs) have sparked a new revolution in the field of natural processing (NLP), and garnered tremendous attention both academic research everyday life, thanks to their unprecedented performance wide range applications. However, deployment remains significant challenge, primarily due intensive computational memory requirements. Hardware acceleration efficient quantization are promising solutions address two issues. In this paper, hardware architecture co-design is...
To meet the demand in a wide range of practical applications, precision-scalable deep neural network (DNN) accelerators are becoming an unavoidable trend. On other hand, it has been demonstrated that DNN accelerator may achieve better computation efficiency through exploiting sparsity. Therefore, with both precision scalability and sparsity exploitation expected to have performance. In this article, we propose efficient can exploit activations. The is obtained from decomposable multiplier...
The behavior of a wireless sensor network dedicated to distributed estimation tasks may be significantly altered by the presence nodes whose sensors are defective and produce erroneous measurements. This paper proposes analyzes performance two algorithms help each node in determining whether it is equipped with sensor. A first collects data from its neighborhood, processes them decide, using some generic local outlier detection test, these contain outliers broadcasts result. Then, determines...
Mobile applications (apps for short) often need to display images. However, inefficient image displaying (IID) issues are pervasive in mobile apps, and can severely impact app performance user experience. This paper presents an empirical study of 162 real-world IID collected from 243 popular open-source Android validating the presence severity issues, then sheds light on these issues' characteristics support future research effective issue detection. Based findings this study, we developed a...
In this study, a novel circuit model was proposed for the design of multifunction integrated diplexers with frequency division, selection, and power division functions. The highly multifunctional diplexer mainly comprises only two different-frequency multimode resonators isolation resistors, which results in compact size. distributed coupling technique adopted to integrate resonators. One resonator used form channel passband, other another passband. Moreover, each passband can be...
Generalized integrated interleaved (GII) codes have attracted much attention in distributed storage systems since they can reduce the overall redundancy by creating shared interleaves. The Berlekamp-Massey (BM) algorithm makes GII decoding efficient reusing preceding results at each stage. However, it also forms speed bottleneck of decoders. reformulated inversionless (riBM) is well known for its short critical path, but not generally suitable polynomials are difficult to be reused directly....
Purpose The purpose of this paper is to propose the self-presentation styles advertising influence consumer self-image, which in turn purchase intention. Design/methodology/approach Using virtual brands as stimuli a series experiments, study collects data on self-image and intention conditions different styles. Findings While mediates relationship between style intention, consumption situation (public vs private) moderates style, That is, self-enhancing promotes customers’ ideal self,...
Precision-scalable deep neural network (DNN) accelerator designs have attracted much research interest. Since the computation of most DNNs is dominated by multiply-accumulate (MAC) operations, designing efficient precision-scalable MAC (PSMAC) units central importance. This brief proposes two low-complexity PSMAC unit architectures based on well-known one, Fusion Unit (FU), which composed a few basic called Bit Bricks (BBs). We first simplify architecture BB through optimizing some redundant...
To guarantee its integrity, a wireless sensor network needs to efficiently detect faulty nodes producing erroneous measurements. This paper proposes fully distributed fault detection algorithm. A node first collects the measurements of neighborhood, processes them decide whether they contain outliers, and broadcasts result. Then, it decides autonomously about functioning status. The algorithm is proposed in two variants, depending on proportion network. theoretical analysis probability error...
Generalized integrated interleaved Bose-Chaudhuri-Hocquenghem (GII-BCH) codes show great potentials in some storage applications, such as flash memory programming, since they need less redundancy than the simple BCH for a target reliability. The coefficient vectors that are precomputed from conversion matrix play very important role GII-BCH encoding and also characterize its hardware architecture. However, require large amount of computations, including inversions multiplications both modulo...
The hot electron transport through Au∕Pd or Au∕Ag/molecular layer/n-GaAs diodes has been studied by using ballistic emission microscopy (BEEM). monolayers consisted of either octanedithiol (HS–(CH2)8–SH) hexadecanethiol (HS–(CH2)15–CH3) linear alkane chains and were deposited from ethanol solutions onto molecular beam epitaxially grown GaAs (001) surfaces. metals electron-beam evaporated the layer. Similar to previous results for Au diodes, metallization an monolayer resulted in a large...
This paper addresses the information theoretical analysis of data compression achieved by random linear network coding in wireless sensor networks. A sparse matrix is considered with columns having possibly different sparsity factors. For stationary and ergodic sources, necessary sufficient conditions are provided on number required measurements to achieve asymptotically vanishing reconstruction error. To ensure optimal ratio, factor can be arbitrary close zero absence additive noise. In...
A feature-based filtering algorithm is proposed to solve the difficult problem of indoor human body positioning. The based on multiple-input multiple-output (MIMO) frequency-modulated continuous wave (FMCW) radar, including noise floor removal in time domain, adaptive mean cancellation frequency and non-coherent accumulation for all virtual array elements. As different postures body, such as standing posture, sitting lying movement, etc., raw radar data processed remove obtain an enhanced...
Radar is capable of providing essential distance information for targets. While noise interference typically present in most indoor or outdoor scenes which makes the target range spectrum not obvious even submerged. In this work, a multi-chirps convolution technique can improve signal to average ratio (SANR) by 14.02dB on with only two chirps through experiments proposed based 60GHz frequency modulated continuous wave (FMCW) radar. The (MCCT) suppress irrelevant and enhances same signals...
Due to their inherent feature of exhibiting frequent disconnections, propagation faulty data in Delay Tolerant Networks can be a critical aspect counteract. Indeed the rare meeting events require that nodes are effective and efficient propagating correct information. Accordingly mechanisms rapidly identify possible or misbehaving should searched. Distributed fault detection has been addressed literature context sensor vehicular networks, but unfortunately these solutions suffer for long...
We consider an antenna tuning problem that is quite important to ensure a satisfying user experience in wireless networks. aim maximize the coverage ratio of large service region by properly choosing angles. In order embrace true complexity practical networks and radio channels, system level simulator required. The optimization algorithm has be performed based on stochastic numerical output simulator. proposed zeroth continuation method solve challenging black-box non-convex problem. basic...