What is it?

BlueSense4 is an ARM Cortex M4-based extensible platform for wearable motion and sound sensing, sensor research and IoT applications.

Novelty

BlueSense4 is an evolution of BlueSense with these key advances:

• ARM Cortex M4 microcontroller (STM32L496), running at 20MHz but capable of up to 80MHz.
• The microcontroller includes a floating point unit (FPU), which makes the platform suitable for e.g. neural network inference (e.g. using ARM CMSIS NN).
• 1MB Flash, 320KB SRAM.
• Improved 9DoF intertial mesurement unit (ICM-20948) with lower power consumption. The platform provides raw IMU data as well as device orientation in quaternion (inferred in software).
• Two digital microphones for ambient sound sensing with nominal frequency up to 16KHz. The microphones are spaced to enable broadside and endfire microphone arrays tuned for an 8KHz frequency response (to be implemented in software).
• Two digital microphone inputs on the expansion connector, therefore allowing to have up to 4 independent audio sources or to design an array of 4 microphones.
• Enhanced 12-bit ADC.
• 12-bit DAC output on the expansion connector with firmware built-in signal generation capabilities.
• External EEPROM with one-time programmable area to store configuration data or unique device ID.
• Multimodal sensing firmware, allowing to simultaneously acquire motion data, sound and external ADC inputs, without any programming. Data can be streamed over USB or Bluetooth, or stored on the local SD card.

Description

The platform is 30x30mm in size - smaller than an Apple Watch - making it suitable for a wide range of wearable sensing applications, or integration in objects or furniture for ubiquitous computing applications.

BlueSense4 is designed to be functional out-of-the-box yet extensible. It’s primary purpose is to be a motion and sound sensing platform for wearable sensing and computing applications.

Besides this, the platform can be used in IoT applications or for sensor research.

Wearable motion & sound sensing

As a wearable platform, BlueSense is a tiny device which can be used to capture body movement and orientation with a 9DoF motion sensor. The platform provides access to the raw data of the 3D accelerometer, 3D gyroscope and 3D magnetometer. It also provides the device orientation in quaternions in an earth coordinate system.

The platform can acquire data from the built-in digital microphones at 8KHz or 16KHz. The microphones are spaced to enable broadside or endfire arrays with a nominal frequency of 8KHz. The actual microphone array algorithm has to be implemented by the user (e.g. post-recording).

The IMU and sound data can either be logged on an SD card or streamed over USB or Bluetooth 2.

A typical application of BlueSense would be to acquire data for motion-based or sound-based activity and context recognition. The processor is also fast enough to implement a variety of real-time algorithms.

Multimodal sensing and sensor research

BlueSense has extension ports allow to plug-in daughterbords. The extension ports offer:

• I2C
• SPI
• DAC output
• ADC inputs
• 2 sigma-delta digital microphone inputs
• Most pins can also be used as GPIO.

Out of the box, the firmware is designed to allow to acquire simultaneously motion data, sound, and external ADC inputs. Together with the extension ports, this makes the platform well suited for multimodal sensing and exploring novel sensor modalities.

IoT platform

As an IoT device it’s main appeal is a true hardware off which allows to put the device entirely to sleep (everything is powered down, including voltage regulators), yet wake up at programmed intervals thanks to a real-time clock wakeup. This allows to achieve very long battery life, as the device can wake up at desired times (once an hour, once a day, ...) to acquire and send sensor data. The extension ports allow to plug in custom sensor modalities which make this device highly versatile.

Firmware

The application firmware has been designed to be useful for a wide range of applications out of the box. It has been optimised for high-speed motion data acquisition (500Hz), sound sensing, and high-speed external ADC acquisition (1+KHz), informed by the need of wearable motion tracking, and activity/context recognition applications.

Out of the box, the firmware can stream and log the data to an SD card (motion, sound and ADC inputs).

Other features

Besides USB, Bluetooth 2 and SD card interface with a FAT32 filesystem, BlueSense has a real-time clock, a coulomb counter to measure it’s own battery level and power consumption, true hardware off, and connectors for extensions. This makes the platform also suitable for sensor research, IoT applications, or as an compact alternative to other microcontroller boards.

Hardware specifications

• STM32L496 CPU (1MB Flash, 320KB SRAM)
• LTC3553 battery charger and voltage regulator
• 3V regulated power supply
• ICM-20948 9DoF motion sensor, with software attitude and heading reference system
• USB interface for charging, device interaction
• RN41 Bluetooth 2 interface for device interaction
• MicroSD card (SDHC standard)
• MAX31343 Real-Time Clock, one of the most accurate RTC on the market (+/-5ppm over the entire temperature range, <0.6ppm at identical temperature)
• LTC2942 Coulomb counter allowing precise measurement of battery charge/discharge and characterisation of real-time current consumption by software
• Extension ports with analog inputs, digital I/O, I2C, SPI, timekeeping, etc.
• True hardware off and programmable wake-up by RTC

Why did you make it?

BlueSense addresses the needs of sensor research (it's an extensible platform) for wearable applications (it has a form factor smaller than what you would get with an Arduino or Teensy), with sound sensing, and high-speed motion sensing (up to 500Hz currently), which is useful for sports application or use in robotic setups.

What makes it special?

• It has high speed motion sensing (raw data and orientation as quaternions) up to 500Hz
• Sound sensing with 2 built-in microphones suited for broadside or endfire arrays
• Data logging and data streaming
• High accuracy real-time clock: 5ppm over the entire temperature range from -40°C to +85°C
• True hardware off
• Extensible with a variety of I/O buses, ADC, DAC, GPIO and two external digital microphone inputs.
• Open firmware
• Used in science applications