M31’s IP Solutions of Artificial Intelligence


For artificial intelligence business opportunity, AI chips will be the role of the brain, i.e., the key component of the smart devices. Innovations in deep learning algorithms and neural network processing are driving new technology requirements for AI SoCs. Deep learning capabilities for vision, speech, context awareness, general data pattern recognition, and more are being added to SoCs across all markets including mobile, IoT, data center, automotive, and so on.

To assist customers in developing competitive AI SoCs, M31 provides IP solutions of various high speed interfaces and foundation IPs on advanced process nodes to meet the requirements of low power consumption, high energy efficiency, and high transmission performance for AI applications.

M31 IP Solutions

  • Foundation: 16FFC, 22ULP, 28HPC+ logic gate library, and GPIO
  • Memory: SRAM compiler
  • Chip-to-chip connectivity: PCIe 2.1/3.1/4.0
  • Vision: MIPI C/D PHY
  • Audio: USB 1.1/2.0/3.2
  • Display: HDMI, DP
  • Storage: SATA, PCIe 2.1/3.1/4.0, MIPI M PHY

AI Market Application Trend: Moving from Cloud to Edge

Edge AI algorithms will process data collected by the devices locally and work properly, i.e., they can process data and make decisions independently without any connection. According to the market forecast reports, the volume of worldwide edge AI hardware will grow from 610 million units in 2019 to 1.6 billion in 2024.