Weijie Wang

Weijie Wang

Auto-generated profile, based on publicly available information

Weijie Wang

Institute of Microelectronics, A*STAR (Agency for Science, Technology and Research), 2 Fusionopolis Way, #08-02, Innovis Tower, 138634, Singapore

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 December 2022 in ACS Applied Electronic Materials
Verified
Journal Article
Closed access
LCLi Chen
Li Chen
CLChen Liu
Chen Liu
MLMinghua Li
Minghua Li
WSWendong Song
Wendong Song
WWWeijie Wang
Weijie Wang
Yao ZhuYao Zhu
Yao Zhu

23 September 2019 in Applied Physics Letters
Verified
Journal Article
Closed access
PDF file
WSWendong Song
Wendong Song
WWWeijie Wang
Weijie Wang
HLHock Koon Lee
Hock Koon Lee
MLMinghua Li
Minghua Li
VZVictor Yi-Qian Zhuo
Victor Yi-Qian Zhuo
NSNavab Singh
Navab Singh
In this letter, we report analog switching characteristics in an analog resistive random access memory device based on a TiW/Al2O3/Ta2O5/Ta stack. For this device, both oxides were grown by using an atomic layer deposition system and the oxygen vacancies were found to exist at the interface of these oxides by using angle-resolved X-ray Photoelectron Spectroscopy. The device exhibits analog switching behaviors. Multiple states were achieved by applying 128 consecutive identical pulses of <20 μs duration and stable for at least 104 s. These characteristics show that the TiW/Al2O3/Ta2O5/Ta device is a promising candidate for synaptic applications.

14 October 2022 in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
Verified
Journal Article
Closed access
TCTiancheng Cao
Tiancheng Cao
CLChen Liu
Chen Liu
WWWeijie Wang
Weijie Wang
TZTantan Zhang
Tantan Zhang
HLHock Koon Lee
Hock Koon Lee
WGWang Ling Goh
Wang Ling Goh
In this work, a non-idealities aware software-hardware co-design framework for deep neural network (DNN) implemented on memristive crossbar is presented. The device level non-ideal factors such as device conductance variation, nonuniform quantization levels, device-to-device variation and programming failure probability are included in the model. At array level, the impact of line resistance and sneak path are considered using a new fast and accurate line resistance estimation model. The non-linearity and offset of the peripheral circuits are also considered. By incorporating these factors into a unified DNN training process, the neural network performance can be evaluated holistically. Furthermore, the proposed training process can effectively mitigate the impact of these non-idealities and reduce the inference accuracy degradations. Implemented in a hybrid fashion of Python and PyTorch, the proposed framework is evaluated with a simplified 5-layer VGG network implemented on measured 128×128 RRAM array with 3-level weight resolution. For CIFAR-10 tasks, 83% inference accuracy is achieved with less than 3% accuracy drop compared to the ideal model.

31 January 2022 in IEEE Communications Letters
Verified
Journal Article
Closed access
XZXingwei Zhong
Xingwei Zhong
KCKui Cai
Kui Cai
GSGuanghui Song
Guanghui Song
WWWeijie Wang
Weijie Wang
Yao ZhuYao Zhu
Yao Zhu
Resistive random access memory (ReRAM) is a promising emerging non-volatile memory (NVM) technology that shows high potential for both data storage and computing. However, its crossbar array architecture leads to the sneak path (SP) problem, which may severely degrade the data storage reliability of ReRAM. Due to the complicated nature of the SP-induced interference (SPI) [1] , it is difficult to derive an accurate channel model for it. The deep learning (DL)-based detection scheme (Zhong et al. , 2020) can better mitigate the SPI, at the cost of additional power consumption and read latency. In this letter, we first propose a novel constrained coding (CC) scheme which can not only reduce the SPI, but also effectively differentiate the memory arrays into two categories of SPI-free and SPI-affected arrays. For the SPI-free arrays, we can use a simple middle-point threshold detector to detect the low and high resistance cells of ReRAM. For the SPI-affected arrays, a DL detector is first trained off-line. To avoid the additional power consumption and latency introduced by the DL detector, we further propose a DL-based threshold detector, whose detection threshold can be derived based on the outputs of the DL detector. It is then utilized for the online data detection of all the identified SPI-affected arrays. Simulation results demonstrate that the above CC and DL aided threshold detection scheme can effectively mitigate the SPI of the ReRAM array and achieve better error rate performance than the prior art detection schemes, without the prior knowledge of the channel.

17 November 2022 in IEEE Electron Device Letters
Verified
Journal Article
Closed access
LCLi Chen
Li Chen
WSWendong Song
Wendong Song
WWWeijie Wang
Weijie Wang
HLHock Koon Lee
Hock Koon Lee
ZCZhixian Chen
Zhixian Chen
Yao ZhuYao Zhu
Yao Zhu
In this letter, we report the KrF excimer laser annealing for achieving high-performance ferroelectric Hf0.5Zr0.5O2(HZO). The laser anneals the target device layer effectively with a high heating/cooling rate without overheating under-layer deep structures. The impact of laser fluence, number of pulses, and pulse duration on the ferroelectricity in HZO was carefully studied through numerical simulation and experiment. The remnant polarization $2{P}_{\text {r}}$ of $22~\mu \text{C}$ /cm2, a coercive field of ~1 MV/cm, and a cycling endurance of > $10^{{8}}$ were achieved in the 8 nm ferroelectric HZO thin film annealed by the laser with an ultra-low laser fluence of ~84 mJ/cm2 and 20 pulses. Benefiting from the large beam size of KrF excimer laser, the laser annealing is practical for obtaining stable wafer-scale ferroelectric HZO within a short time.