tinyML Talks on May 12, 2021 “TinyML FPGA implementation for condition monitoring” by Altaf Khan and Martin Kellermann

We held our next tinyML Talks webcast. Altaf Khan and Martin Kellermann from Infxl and Microchip Technology presented TinyML FPGA implementation for condition monitoring on May 12, 2021.

We have reduced the size of the deep neural net inference engine by minimizing the intra-network connectivity, eliminating the need for floating-point data, and replacing the multiply-accumulate operation with just accumulation. The resultant small-footprint, low-latency deep nets are suitable for embedded applications in general. They are especially suited for processing data from IoT sensors (inertial, vibration, temperature, flow, electrical, biochemical, etc.) in battery-powered endpoint applications in wearables, robots, and automotive, particularly for predictive maintenance, real-time condition monitoring, and process automation use cases. The trained deep nets are delivered in the form of compact and simple C code that is suitable for MCU, DSP, and FPGA implementations. We present FPGA size and performance results on an IoT condition monitoring use case.

Altaf Khan is the CEO of Infxl LLC, Colleyville, TX. He started his career as an accelerometer system engineer in Silicon Valley, but simplifying neural nets has been his passion over the last three decades. He has developed fast deep nets for real-time applications, low-cost deep nets for battery-operated IoT endpoints, and small-footprint deep nets for FPGA. He has developed intelligent solutions for a major US airline and a well-known auto parts supplier. He has been the CTO of a brokerage company, CEO of two startups, consultant for software process improvement, and an industrial controls engineer. Altaf received his BSEE from Wilkes College, MSEE for the University of Pennsylvania, and PhD from the University of Warwick.

Martin Kellermann is a Marketing Manager at Microchip Technology GmbH, Munich. Earlier he was a Staff Field Application Engineer at Xilinx. He is a seasoned FPGA and SoC professional with a track record of successful customer and project engagements in the industrial, automotive, and data-center domains. He possesses a strong background in high-speed serial data transmission, signal integrity, and hardware debugging which helped numerous customers finish their designs successfully. He has also taught courses covering industrial applications and hardware concepts. Martin is a graduate of the Landshut University of Applied Sciences.


Watch on YouTube:
Altaf Khan and Martin Kellermann

Download presentation slides:
Altaf Khan and Martin Kellermann

Feel free to ask your questions on this thread and keep the conversation going!

If you want to get in contact feel free to email me on Martin.Kellermann@microchip.com.