The reference design for a battery-powered smart alarm system (SAS) employs artificial intelligence (AI)-based sensor fusion algorithm that complements acoustic event detection (AED) processing with a wake detection engine. This reference design solution unveiled by Infineon Technology comprises the following major building blocks:
- A high signal-to-noise ratio (SNR) analog MEMS microphone
- A digital pressure sensor
- An Arm Cortex M4F-based microcontroller
- Sensor fusion algorithm based on precisely-trained AI/ML models
The sensor fusion software algorithm, which runs on a CY8C62 PSoC 6 microcontroller, combines acoustic and pressure sensor data to accurately differentiate between sharp sounds and distinctive audio/pressure events. The events can be created when glass is broken or a house alarm is triggered due to a window or door opening. Moreover, the AI-based sensor fusion algorithm can eliminate other background sounds or pressure events that can generate false positives due to the similarities to alarm systems.
According to David Jones, head of marketing and business development for Intuitive Sensing Solutions at Infineon Technologies, what’s unique about this solution is that it uses AI models rather than sound frequencies to detect acoustic events.
Then there are power consumption benefits ensured by deep-sleep modes in the microcontroller. Only IM73A135V01 MEMS microphone is always-on, listening to potential acoustic events. Here, Infineon has replaced the digital MEMS microphone with an analog IM73A135V01 MEMS microphone to lower the power usage.
The DPS310 digital pressure sensor works alongside the MEMS microphone to facilitate sound classification and wake-on acoustic event detection. The PSoC 6 microcontroller controls the power-up sequence for the pressure sensor.
The entire smart alarm system design is packed into a system-in-package (SiP), which developers can embed into security use cases like thermostats. Developers can use their own wireless technology to provide connectivity for the Internet of Things (IoT) and cloud systems. “It’s a complete solution that developers can use to build low-power security and surveillance products for the marketplace,” Jones said.
The smart alarm system (SAS) reference design is available now while reference boards will be available in the third quarter of 2022.
>> This article was originally published on our sister site, EDN.
- Sensor fusion brings multiple benefits
- Sensor fusion enables sophisticated next-gen applications
- Air quality sensor embeds AI capabilities
- Smart sensor board speeds edge AI development
For more Embedded, subscribe to Embedded’s weekly email newsletter.