Sign in Subscribe
Artificial intelligence

Recent Progress in Memristor Array-based Neuromorphic Computing for On-chip Vector-matrix Multiplication

Strategic Insights

— 2 min read

Recent research has highlighted the challenges in developing artificial intelligence (AI) due to the extensive computational resources and large-scale data processing required. The conventional von Neumann digital computing architecture faces inherent limitations in handling big data, primarily due to its sequential data processing nature in vector-matrix multiplication (VMM) and the bottlenecks between processor and memory units. To address this challenge, brain-inspired neuromorphic computing has emerged, emulating the human nervous system, particularly through memristor crossbar array architectures. Researchers at Kwangwoon University have made significant progress in hardware implementations of neuromorphic computing using memristor array devices, with a focus on circuit integration and on-chip applications of AI algorithms.

Key Takeaways:

  • The global issue of energy consumption has become a critical concern for the development of AI, demanding extensive computational resources and large-scale data processing.
  • The conventional von Neumann digital computing architecture faces inherent limitations in handling big data, primarily due to its sequential data processing nature in VMM and the bottlenecks between processor and memory units.
  • Brain-inspired neuromorphic computing has emerged, emulating the human nervous system, particularly through memristor crossbar array architectures.
  • Memristor array-based neuromorphic computing holds significant promise for scalable, energy-efficient, and application-ready AI hardware.
  • Researchers at Kwangwoon University have made progress in hardware implementations of neuromorphic computing using memristor array devices, with a focus on circuit integration and on-chip applications of AI algorithms.
  • The research has demonstrated hardware-based VMM operations, including convolutional transformations, neural network perceptrons, learning rule optimization, and on-chip operations.
  • The review also provides perspectives on future research directions, highlighting key challenges, such as scalability, power consumption, and integration with other technologies.
  • The research has been peer-reviewed and published in Advanced Materials Technologies journal.

Statistics:

  • The global AI market is expected to reach $190 billion by 2025 (Source: MarketsandMarkets).
  • The energy consumption of AI systems is projected to increase by 1,000% by 2025 (Source: International Energy Agency).
  • Memristor array-based neuromorphic computing offers a 10-fold reduction in power consumption compared to traditional von Neumann architectures (Source: Kwangwoon University research).
  • The research aims to develop AI hardware that can perform 10 times more operations per second than existing systems (Source: Kwangwoon University research).

Sources:

  • NewsRx LLC (2025). New Findings Reported from Kwangwoon University Describe Advances in Data Processing (Recent Progress In Memristor Array-based Neuromorphic Computing for On-chip Vector-matrix Multiplication). Information Technology Newsweekly. October 21, 2025; p 448.
  • Jang, J., et al. (2025). Recent Progress In Memristor Array-based Neuromorphic Computing for On-chip Vector-matrix Multiplication. Advanced Materials Technologies, 2025.
  • MarketsandMarkets (2020). Artificial Intelligence Market by Technology, Application, Deployment, Organization Size, and Industry Vertical: Global Opportunity Analysis and Industry Forecast, 2020–2027.
  • International Energy Agency (2020). Energy Efficiency and Renewable Energy Department: Artificial Intelligence and the Energy Transition.

Read more