The ultra-low power, analogue processing platform has been designed to overcome the power and data handling challenges associated with battery-operated, always-on sensing devices for the consumer, smart home, Internet of Things (IoT), industrial markets.
The RAMP platform incorporates machine learning into an analogue processor that can detect and classify events from background noise before the data is digitised. By directly analysing raw analogue sensor data for what’s important to the application, the RAMP platform more efficiently partitions the always-on system’s power and data resources to eliminate the higher-power processing and transmission of irrelevant data.
Compared to an older “digitise-first” architectures, where all sensor data must be continuously digitised for event analysis, the RAMP-based “analyse-first” approach brings more intelligence to the sensor edge, reducing the power required by up to 10x and the volume of data handled by up to 100x for always-on applications.
Demand for always-on voice-first devices is surging, and smart speakers and wearables/hearables are among the largest and fastest-growing market segments, with smart speakers reaching 482 million units by 2021 (according to IHS Markit) and wearables/hearables reaching 417 million hearables by 2022 (according to Juniper Research).
With device manufacturers heavily invested in the success of always-on portable sensing devices, technology developers are working to alleviate barriers to adoption. Chief among these is the short battery life that makes many always-on sensing devices unattractive to end users.
“We are at the cusp of a mass proliferation of always-listening, continuously processing devices. To reach that next level, we need to resolve the architectural issues that are deal-breakers for some applications,” said Tom Doyle, founder and CEO, Aspinity. “Voice-first devices ought to run for long periods of time without requiring battery recharge or they risk frustrating consumers. We’re committed to fixing this problem through an intelligent architectural approach. Our RAMP platform analyses the incoming sound at the microphone edge to keep the wake-word engine and other digital processors in a low-power sleep state for the 80% of the time that no voice is present. Manufacturers who can offer a voice-first TV remote that runs for a year per battery change or a smart earbud that can run for an entire day without a recharge will gain a major competitive edge in the marketplace.”
The RAMP technology replicates sophisticated digital processing tasks in compact, ultra-low power, analogue circuitry which supports event detection and classification from raw, unstructured analogue sensor data. Leveraging the nonlinear characteristics of a small number of transistors, RAMP incorporates modular, parallel and continuously operating analogue blocks that mimic the brain’s efficiency. Each of these blocks is implemented in a much smaller and more efficient programmable footprint than traditional analogue circuits.
The RAMP platform supports many applications by configuring the analogue blocks for typical digital tasks such as signal analysis and compression, as well as more complex tasks such as feature extraction, event detection, and classification.
The platform’s analogue blocks can be reprogrammed with application-specific algorithms to analyse raw analogue data from multiple types of sensors, such as accelerometers used for industrial vibration monitoring. Instead of a predictive maintenance system that continuously digitises thousands of points of data to monitor the trends in the changes of certain spectral peaks, RAMP can sample and select only the most important data points, compressing the quantity of vibration data by 100x and dramatically decreasing the amount of data collected and transmitted for analysis.
Reducing the amount of data that is handled in this type of always-on application is the key to a more easily deployable, battery-operated, wireless sensor system.
