At Embedded World, Maxim Integrated showcased the latest in analog and mixed signal solutions for automotive, healthcare, industrial, artificial intelligence (AI) and security applications. It also won the inline prize 2021. The company was honored by the expert jury for the MAX78000 low-power neural network controller. Maxim received the award during an online ceremony on March 3. Microcontrollers perform AI inference with less than 1/100 the power of software solutions, which greatly improves the runtime of battery-powered AI applications. Power improvements come without compromising latency or cost, as the MAX78000 delivers conclusions 100 times faster than software solutions running on low-power microcontrollers, at a fraction of the cost of an FPGA or GPU solution. This paves the way for new and complex use cases and enables machines to see and sense complex patterns through low-energy, localized processing done in real time. MAX78000 At the heart of the MAX78000, we can find specialized devices designed to reduce the power consumption and latency of convolutional neural networks. This device operates with minimal interference from a microcontroller, which makes the operation very smooth. Sally Ward-Foxton interviewed Chris Ardis, CEO of the Micros unit, security and software business at Maxim Integrated, during the product launch last year. You can find the full interview here. The MAX78000 rated images on the MNIST dataset at 1100 times less power and 400 times faster versus a software solution on the Cortex M4F, and was 600 times less and 200 times faster at keyword discovery than the Cortex M4F. The MAX78000 features an Arm Cortex-M4F core, an energy-optimized RISC-V core and an internally developed CNN accelerator (Image: Maxim) Intelligence towards the edge of the digital factory. The rise of IoT device connections and the exponential growth of data are generating a new phase in digital transformation. The environment in which the data is generated becomes the core of the business. Businesses face the challenge of modernizing their infrastructure to remain competitive, reaping significant performance benefits, and reducing the costs of using latency and bandwidth to upload data to the cloud. During its webinars in the Embedded World, Maxim showcased the Go-IO platform that pushes intelligence closer to the edge, enabling active monitoring and reporting of equipment health and condition information, as well as increasing productivity. The reference design also meets the PLCs’ stringent size and increased power requirements. Digital factories can adjust production line dynamically in real time based on new requirements or changes in status. To fully achieve industrial convergence, automated equipment must also have self-diagnosis and optimization capabilities. .