Figure 1: Challenges in CIM systems for neural network computation.
A cross-institutional research team led by researchers from the Department of Electrical and Electronic Engineering (EEE), under the Faculty of Engineering at The University of Hong Kong (HKU), have achieved a major breakthrough in the field of artificial intelligence (AI) hardware by developing a new type of analog-to-digital converter (ADC) that uses innovative memristor technology.
Conventional AI accelerators face challenges because the essential components that convert analog signals into digital form are often bulky and power-consuming. Led by Professor Ngai Wong, Professor Can Li and Dr Zhengwu Liu of HKU EEE, in collaboration with researchers from Xidian University and the Hong Kong University of Science and Technology, the cross-disciplinary research team developed a new type of ADC that uses innovative memristor technology. This new converter can process signals more efficiently and accurately, paving the way for faster, more energy-efficient AI chips.
The research team created an adaptive system that automatically adjusts its settings based on the data it receives, i.e. dynamically fine-tuning how signals are converted. This results in a 15.1× improvement in energy efficiency and a 12.9× reduction in circuit area compared with state-of-the-art solutions.
Beyond hardware performance, this new adaptive ADC not only improves hardware performance but also maintains high accuracy when running neural network tasks across various types of AI models. When integrated into compute-in-memory (CIM) systems, the ADC further reduces overall energy consumption and chip size by over 57% and 30%, respectively.
This breakthrough represents a significant milestone in bridging the gap between algorithmic intelligence and hardware adaptability, showcasing how memristive computing can revolutionise the design of next-generation AI chips. It also highlights HKU EEE's leadership in cross-domain research that unites device physics, circuit design, and machine learning systems.
The project received support from the Theme-based Research Scheme (TRS) project T45-701/22-R, the National Natural Science Foundation of China (62404187, 62122005), Croucher Foundation, and the General Research Fund (GRF) Project (17200925, 17203224, 17207925) of the Research Grants Council (RGC), Hong Kong SAR.
Their recent study, titled "Memristor-based adaptive analog-to-digital conversion for efficient and accurate compute-in-memory," was published in Nature Communications.
Link to the paper: https://doi.org/10.1038/s41467-025-65233-w
About Professor Ngai Wong
Ngai Wong received the B.Eng. and Ph.D. degrees in electrical and electronic engineering from HKU. He was a Visiting Scholar with Purdue University, West Lafayette, IN, USA, in 2003. He is currently an Associate Professor with the Department of Electrical and Electronic Engineering, HKU. He is the Director of the AVNET-HKU Emerging Microelectronics & Ubiquitous Systems (EMUS) Lab launched in 2025. His research interests include compact neural network design, compute-in-memory (CIM) AI chips, electronic design automation (EDA) and tensor algebra. He also serves as the project coordinator of a 5-year Hong Kong Theme-based Research Scheme (TRS) titled "ReRACE: ReRAM AI Chips on the Edge" (2022-2027) that promotes next-gen neuromorphic AI computing and applications.
