In recent years, many manufacturers have started to try to integrate AI functionality into MCU, and Ruisa Electronics is also one of the manufacturers who pay attention to MCU+AI.



Previously, AI computing was mainly done in the cloud, but now it has gradually developed towards the edge. Machine learning (ML) training is generally completed in the cloud, inference can be done in the cloud or device, and ML processing can be done at the edge. The advantage of doing so is that it can reduce the bandwidth of data uploaded from the cloud, improve the response speed of local devices, and improve the security of local data.

In the era of the Internet of Things, data is showing an explosive growth trend, and CPUs are facing enormous computational pressure. Different solutions have emerged on the market regarding how to release the computational pressure on the CPU. Some companies are starting to add accelerators in MCUs to perform ML calculations using dedicated computing power, in the hope of releasing the general computing power of CPUs.

In recent years, many manufacturers have started to try to integrate AI functionality into MCU, and Ruisa Electronics is also one of the manufacturers who pay attention to MCU+AI.

Focusing on the real-time analysis function of MCU+AI

With the gradual development of MCU computing towards the edge, it is necessary to integrate AI functions in MCU, and image and speech processing has become an important application direction of MCU+AI. In fact, there are three main scenarios for artificial intelligence/machine learning - voice, video, and real-time analysis.

Regarding this, Mr. Mohammed Dogar, Vice President of the MCU Business Development Department of Renesa Electronic Internet of Things and Infrastructure Business Headquarters, introduced: "People pay a lot of attention to voice and visual, but not enough attention to real-time analysis. We believe that this part (real-time analysis) The growth rate is also very high. We will provide application solutions in all three aspects