On October 23, news emerged that a research team, headed by Researcher Sun Zhong from Peking University's Institute for Artificial Intelligence and in collaboration with a research group from the School of Integrated Circuits, has made a significant stride. They've successfully crafted a high-precision, scalable analog matrix computing chip that leverages resistive random access memory (RRAM). This innovative chip marks a milestone as it's the first of its kind to deliver analog computing system performance that rivals digital computing in terms of precision. When tackling critical scientific challenges like large-scale MIMO signal detection, this chip outperforms current state-of-the-art digital processors (GPUs) by a staggering margin. Its computational throughput and energy efficiency are boosted by a factor of 100 to 1000. The groundbreaking research findings were published in the esteemed journal Nature Electronics on October 13.
