A5045947915- Advanced Memory and Neural Computing
- CCD and CMOS Imaging Sensors
- Neuroscience and Neural Engineering
- Ferroelectric and Negative Capacitance Devices
- Artificial Intelligence in Healthcare and Education
- Neural dynamics and brain function
- Advanced Neural Network Applications
- Topic Modeling
- Robotic Path Planning Algorithms
- Electric Vehicles and Infrastructure
- Autonomous Vehicle Technology and Safety
- Fault Detection and Control Systems
- Neural Networks and Reservoir Computing
- Radiomics and Machine Learning in Medical Imaging
- Robotics and Sensor-Based Localization
- Vehicle License Plate Recognition
- VLSI and Analog Circuit Testing
Xi'an Jiaotong University
2023
National University of Defense Technology
2021-2022
Shanghai Jiao Tong University
2019
Tongren Hospital
2019
Imaging examinations, such as ultrasonography, magnetic resonance imaging and computed tomography scans, play key roles in healthcare settings. To assess improve the quality of diagnosis, we need to manually find compare pre-existing reports pathology examinations which contain overlapping exam body sites from electrical medical records (EMRs). The process retrieving those is time-consuming. In this paper, propose a convolutional neural network (CNN) based method can better utilize semantic...
Memristor can store calculation results while computing, and utilizing memristor to realize in-memory computing is considered as one of the potential methods break bottleneck Von Neumann architecture. To construct high performance systems, it necessary design memristor-based units with great advantages in delay, area, integratability. This brief presents a speed <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math...
Benefiting from area and power efficiency, memristors enable the development of neural network analog-to-digital converter (ADC) to break through limitations conventional ADCs. Although some memristive ADC (mADC) architectures have been proposed recently, current research is still at an early stage, which mainly on simulation requires numerous target labels train synapse weights. In this paper, we propose a pipelined Hopfield mADC architecture experimentally demonstrate that such has...
This paper discusses the problem of local path planning for autonomous vehicles. article introduces pruning strategies and their related map construction data processing. Next, a forward strategy was introduced, universally applicable selection method provided. The value prediction driving technology demonstrated by comparing it with ordinary mobile robot algorithms. Then, optimization paths through reduced time required updating algorithm introduced in article. refers to provided two...
Abstract In‐memory computing based on memristor logic is one of the most promising approaches to break ‘memory wall’ and ‘power wall’. However, some important logics (such as AND MAJ) cannot be implemented efficiently in 1T1R arrays, which makes it difficult design low‐delay memristor‐based systems. An efficient threshold proposed by adding an auxiliary MAGIC ( N +1)‐OR gate optimising driving voltage. The can completely a array without any additional resistor, achieve MAJ directly....
With the ever-growing demands for sampling rate, conversion resolution, as well lower energy consumption, memristor-based neuromorphic analog-to-digital converters (MN-ADC) becomes one of most potential approaches to break bottleneck traditional ADCs. However, online trainable MN-ADCs are not designed be easily integrated into 1T1R crossbar array, meanwhile suffering from device non-idealities, which makes it difficult realize high-speed and accurate conversion. To overcome these issues,...
Abstract The memristive stateful logic can realize the memory and computation, thus effectively avoid huge time energy overhead caused by data moving between computation units. However, still faces reliability challenges variability of memristor, which prevents it from practical applications. A reinforcement method with low in delay, area peripheral circuits is needed to be studied urgently. This paper proposes a novel based on gates' margin. By optimizing circuit structure driving voltage...
The popularization of new energy vehicles relies on a powerful charging system. Therefore, it has become an inevitable trend to increase equipment, but today's vehicle system the following problems. First, construction speed piles cannot keep up with car iteration. Second, widespread deployment stations consumes lot human and financial costs. Third, peak-to-valley difference in demand for using is too large. Based this, we designed intelligent robot target recognition. This project uses...
Counter is one of the basic parts a digital device. Traditional counters based on Complementary Metal Oxide Semiconductor (CMOS) cannot maintain data after power down. It may require extra and time consumption to store counting results in memory. Using nonvolatile memory construct counter can solve these problems. In this paper, we designed 4-bit up-down by using Memristor-Aided LoGIC (MAGIC) auxiliary CMOS circuits. Furthermore, verify design through LTSPICE simulations. The simulation show...