Eta Compute Inc. announces the first shipment of production silicon for its ECM3532, the world’s first AI multicore processor for embedded sensor applications. This unique multicore device features the company’s patented Continuous Voltage Frequency Scaling (CVFS) and delivers power consumption of microwatts for many sensing applications.
Eta Compute’s ECM3532 is a Neural Sensor Processor (NSP) for always-on image and sensor applications. It will be on display at the 2020 tinyML Summit, February 12-13 at Samsung Electronics in San Jose, California. Eta Compute is a Gold Sponsor of tinyML and will demonstrate the ECM3532 for image recognition and other edge sensing applications. The objective of the entire tinyML community is to enable ultra-low power machine learning at the network edge.
“Our Neural Sensor Platform is a complete software and hardware platform that delivers more processing at the lowest power profiles in the industry. This essentially eliminates battery capacity as a barrier to thousands of IoT consumer and industrial applications,” said Ted Tewksbury, CEO of Eta Compute. “We are excited to see the first of many applications our customers are developing come to market later this year.”
Eta Compute’s ECM3532 family brings AI to edge devices and transforms sensor data into actionable information for voice, activity, gesture, sound, image, temperature, pressure, and bio-metrics applications, among others. The platform solves issues for the most important issues in edge computing: longer battery life, shorter response time, increased security and higher accuracy.
“We believe that power consumption, latency and data generation combined with RF transmission are all factors limiting many sensing applications,” said Jim Feldhan, president and founder at Semico Research. “It’s great seeing Eta Compute’s platform coming into the market. Their technology is orders of magnitude more power-efficient than any other technology I have seen to date and it will certainly make AI at the edge a reality.”
The company’s standalone AI platform includes a multicore processor, that includes flash memory, SRAM, I/O, peripherals and a machine learning software development platform. The patented CVFS substantially increases performance and efficiency for edge devices. The self-timed CVFS architecture automatically and continuously adjusts internal clock rate and supply voltage to maximize energy efficiency for the given workload. The ECM3532 multicore NSP combines an MCU and a DSP, both with CVFS, to optimize execution for the best efficiency making it an ideal solution for IoT sensor nodes.
- 5 x 5 mm 81 ball BGA
- As low as 100μW active power consumption in always-on applications
- Arm Cortex-M3 processor with 256KB SRAM, 512KB Flash
- 16b Dual MAC DSP with 96KB dedicated SRAM for ML acceleration
- Neural Development SDK with TensorFlow interface for seamless model integration into the ECM3532
“It’s exciting to see innovative products for low power machine learning being launched at tinyML where experts from the industry, academia, start-ups and government labs share the innovations to drive the whole ecosystem forward,” said Pete Warden, Google Researcher and General Co-chair of the tinyML organization.
“We are amazed by the ECM3532 and its efficiency for machine learning in sensing applications,” said Zach Shelby, CEO of Edge Impulse. “It is an ideal fit for our TinyML lifecycle solution that transforms developers’ abilities to deploy ML for embedded devices by gathering data, building a model that combines signal processing, neural networks and anomaly detection to understand the real world.”
“Himax Imaging HM01B0 and new HM0360 are among the industry’s lowest power image sensors with autonomous operation modes and advanced features to reduce power, latency and system overhead. Our image sensors can operate in sub-mW range and when paired with the low power multi-core processors such as Eta Compute’s ECM3532, developers can quickly deploy edge devices that perform image inference under 1mW,” said Amit Mittra, CTO of Himax Imaging.