Related Experiment Video
Updated: Apr 18, 2026

Assembly and Characterization of Biomolecular Memristors Consisting of Ion Channel-doped Lipid Membranes
Published on: March 9, 2019
WS2 Optoelectronic Memristive Reservoir Enabling Ultra-Low-Power, Multi-Task, and Environmentally Stable Neuromorphic
Dayanand Kumar1, Hanrui Li1, Divyanshu Divyanshu2
1Smart, Advanced Memory Devices and Applications (SAMA) Laboratory, Electrical and Computer Engineering, Computer Electrical Mathematical Science and Engineering Division, King Abdullah University of Science and Technology (KAUST), Thuwal, Kingdom of Saudi Arabia.
We developed an ultra-low-power optoelectronic reservoir computing (RC) system using WS2/Zinc-Tin-Oxide (ZTO) memristors. This system achieves high accuracy in visual and speech tasks while consuming minimal energy, paving the way for efficient edge intelligence.
Area of Science:
- Materials Science
- Neuromorphic Engineering
- Energy-Efficient Computing
Background:
- Conventional silicon chips are power-hungry for edge AI.
- Developing energy-efficient processing for visual and speech tasks is crucial.
- Optoelectronic devices offer potential for integrated sensing and computation.
Purpose of the Study:
- To create an energy-efficient optoelectronic reservoir computing (RC) system.
- To integrate sensing, memory, and computation in a single WS2/ZTO memristive device.
- To demonstrate the system's capability for biological function emulation and edge intelligence.
Main Methods:
- Fabrication of a WS2/Zinc-Tin-Oxide (ZTO) memristive device for reservoir computing.
- Conversion of raw visual (N-MNIST, motion) and speech (FSDD) data into spike trains.
- Utilizing the memristor's volatile and nonlinear dynamics for temporal input decoding.
- Testing the system's performance on N-MNIST, motion perception, and speech recognition tasks.
Main Results:
- Achieved high accuracy: ~94% on N-MNIST, ~93% on motion perception, ~89% on speech recognition.
- Demonstrated ultra-low energy consumption of ~25.5 fJ/spike.
- Showcased excellent environmental stability under 95% relative humidity.
- Validated 95% accuracy on N-MNIST with minimal training energy and over 1.5 million cycles endurance.
Conclusions:
- The WS2/ZTO memristive RC system offers a humidity-resilient, ultra-low-power platform for in-sensor neuromorphic processing.
- This technology advances edge intelligence by enabling efficient visual and speech processing.
- The device exhibits synaptic functionalities, supporting future development of advanced AI hardware.
Related Concept Videos
System of Memory
MOS Capacitor
The metal gate is typically made from highly conductive materials such as aluminum or polysilicon. Beneath the metal gate lies a thin layer of...
Understanding Memory
Electrochemical Systems
Semiconductors
Metals such as copper (Cu), zinc (Zn), or lead (Pb) have low resistivity and feature conduction bands that are either not fully occupied or overlap with the valence band, making a bandgap non-existent. This allows electrons in the highest energy levels of the valence band to easily transition to the conduction band upon gaining...
Higher Mental Functions of Brain: Learning and Memory

