UROP Proceedings 2020-21

School of Engineering Department of Electronic and Computer Engineering 159 Automated Test Equipment for Magnetic Memory and Sensors Supervisor: SHAO Qiming / ECE Student: TONG Zihan / ELEC Course: UROP3100, Summer Recent theories and experiments have been demonstrating that besides the binary-like non-volatile magnetization switching, perpendicular magnetic tunnel junctions (MTJs) based on spin-orbit torque switching may also show analogue-like behaviours, making them one of the possible candidates for the building block of the next-generation solidstate devices of neuromorphic computing. This project aims at developing a compact model for the multi-domain MTJ devices based on the Preisach model of hysteresis, which treats the ferromagnetic material as the combination of parallel and independent hysterons with respective switching thresholds or switching probability curves. Model verification based on various experiment data is also performed and discussed. Circuit Emulation of Bio-Inspired Dynamic and Topological Quantum Systems Supervisor: SHAO Qiming / ECE Student: ZHENG Yueyan / CPEG Course: UROP1000, Summer Because of the high demand for processing the temporal signal, a kind of RNN-based framework is created and whose name is reservoir computing. Reservoir computing became popular based on its lower training cost compared with the traditional RNN. This work is based on the paper- Dynamic memristor-based reservoir computing for high-efficiency temporal signal processing (Zhong et al., 2021), aimed to detect and test the performance of a dynamic memristor-based reservoir computing which can highly efficient and accurate processing temporal signals. The entire study will be divided into three tests to test the performance of this reservoir, including waveform classification, spoken-digit recognition, and time-series prediction. We hope to develop the temporal signal processing performance of dynamic memristor-based reservoir computing. Circuit Emulation of Bio-Inspired Dynamic and Topological Quantum Systems Supervisor: SHAO Qiming / ECE Student: LO Yuk Ho / PHYS Course: UROP1100, Summer Recently, it was found that electric circuits could be used to emulate bio-inspired dynamic and topological quantum systems due to various available circuit components with different characteristics. Besides, it is more convenient if people can use simulation software to emulate their desired systems. In this project, LTspice, a high-performance SPICE simulation software, is used to emulate some topological circuits proposed by the literature ”Topolectrical Circuits”. The targeted circuits are Su-Schrieffer-Heeger (SSH) circuit and the Graphene circuit, which is a higher dimensional circuit compared to SSH circuit. From the simulated circuits, some results are obtained and used to compare with those mentioned in the literature we refer to.

RkJQdWJsaXNoZXIy NDk5Njg=