Ambiq Micro Inc., one of a host of well-funded startups that are vying to build microprocessors to power a new generation of smart devices, today introduced the fourth generation of its Apollo processor line.

The company says it can enable some wearable, tracking and healthcare devices to run for months on a single charge.

The Apollo4 system-on-a-chip is based on the Arm Holdings Ltd. Cortex M4 processor and achieves a power efficiency of three microamperes and 10 microwatts per megahertz with low deep sleep current modes. A microampere and a microwatt are one-millionth of an ampere and a watt, respectively. At those power levels, the chip can still support clock speeds of up to 192 megahertz, 2D and 2.5D graphics acceleration to power displays up of to 640X480 resolution with 32-bit color as well as integrated Bluetooth low-energy communication at 2 megabits per second.

Ambiq claims the Apollo4’s power usage is one-tenth the industry and average and twice the efficiency of the company’s previous processor line. It delivers two to three times better power efficiency out of the box and significantly more when originally equipment manufacturers adhere to reference designs Ambiq provides to optimize power-draining features like networking and storage. One maker of smart watches saw battery life increase from three days to three weeks, said Dan Cermak, vice president of architecture and product planning at Ambiq.

Founded in 2010, Ambiq’s processors are now in use in more than 100 million devices ranging from smart watches to home health monitors. “We are particularly strong in always-on applications and those that require the processor to run all the time,” Cermak said. Core application areas include voice recognition, continuous displays and sensors.

Ambiq is one of hundreds of startups that are pursuing what is expected to be an enormous market for artificial intelligence at the edge of corporate network. Complex applications such as autonomous vehicles don’t have the luxury of passing data back to a cloud for analysis. New generations of microprocessors are being developed that are optimized for machine and deep learning with integrated wireless networking and low power consumption. Yole Développement expects the market for AI hardware will grow more than 37% annually through 2024, to $15.6 billion.

Many developers of sensors and other edge devices are attempting to implement neural networks at the edge for problems such as fusing inputs from multiple sensors or sound detection. “There are very sophisticated algorithms that have been developed on top of TensorFlow, for example,  that they want to port onto microcontrollers,” said Cermak, referring to the open-source machine learning platform originally developed by Google LLC.

The COVID-19 pandemic has charged up demand for smart devices even more, Cermak said. Among the new application areas are contact tracing cards, early warning wearables and biometric payment and access devices.

Photo: Pixabay