STMicroelectronics UM2580 User Manual

UM2580

User manual

How to use the wireless multi-sensor development kit with customizable app for IoT and wearable sensor applications

Introduction

The STEVAL-MKSBOX1V1 (SensorTile.box) is a ready-to-use box kit with wireless IoT sensor platform designed to help you build apps that use motion and environmental sensors, regardless of your level of expertise.

The hardware node is a board that fits into a small plastic case (IP54) with a rechargeable battery. You can connect with your smartphone to the board via Bluetooth by using the ST BLE Sensor app (available both on Google Play and Apple Store) and immediately build your own apps through a special interface that offers beginner and expert level functionality. This multi-sensor kit therefore allows you to design wireless IoT and wearable sensor applications quickly and easily, without performing any programming.

SensorTile.box includes a firmware programming and debugging interface that allows professional developers to engage in more complex firmware customization using the STM32 Open Development Environment (STM32 ODE), which includes a sensing AI function pack with neural network libraries.

The kit board includes an embedded SPBTLE-1S Bluetooth SMART application processor that is compliant with BT specification v4.2. This transmitter module is FCC (ID:S9NSPBTLE1S) certified and IC (IC:8976-SPBTLE1S) certified.

Note:	SPBTLE-1S has been replaced by the BlueNRG-M2 module, compliant with BT specification v5.2 in latest
	production batches.
	Figure 1. STEVAL-MKSBOX1V1 (SensorTile.box) multi-sensor development kit

UM2580 - Rev 6 - April 2021	www.st.com
For further information contact your local STMicroelectronics sales office.

UM2580

How to set up the hardware

1How to set up the hardware

Important:

Before you begin, please check the insert card that comes with the SensorTile.box blister pack. If it doesn't show a procedure for battery connection similar to the steps below, then your device is supplied with the battery already connected to the board. In this case, you only need to connect the device via USB to wake it up the first time.

If the insert card has a similar procedure to the steps below, your device is supplied with the battery disconnected and you should follow this procedure to connect the battery and wake the device up.

Step 1. Remove the SensorTile.box contents from its package.

Step 2. Unscrew the shroud cover.

You should have the following items:

–An evaluation board in a plastic shroud

–A LiPo battery

Step 3. Slide the male battery connector vertically into the female connector on the board.

You will hear a light click when the connector is attached correctly.

Figure 2. STEVAL-MKSBOX1V1 battery connection

Step 4. Re-position the circuit with the battery below it and the close the shroud with one of the following types of lid:

–with flanges

–without flanges

Step 5. If necessary, charge the battery via a USB cable.

The blinking of the red LED indicates the battery charging status.

UM2580 - Rev 6	page 2/40

UM2580

How to use ST BLE Sensor app with SensorTile.box

2How to use ST BLE Sensor app with SensorTile.box

Before you begin, you need to download and install our ST BLE Sensor app on your smartphone. The app is available from the Google and Apple online stores.

Step 1. Launch the app on your smartphone.

Figure 3. ST BLE Sensor app main screen

UM2580 - Rev 6	page 3/40

UM2580

How to use ST BLE Sensor app with SensorTile.box

Step 2. Select [CREATE A NEW APP].

The Example Apps screen that follows lists the preloaded apps that you can use immediately.

Figure 4. Example Apps screen

UM2580 - Rev 6	page 4/40

UM2580

How to use ST BLE Sensor app with SensorTile.box

Step 3. Select one of the apps with the icon from the list.

After you select the app, ST BLE Sensor will scan for available SensorTile.box devices in range.

Figure 5. Board selection

Step 4. Select the appropriate SensorTile.box device from the Board screen.

A blue LED on the SensorTile.box device will flash slowly to confirm Bluetooth pairing.

A pop up message in ST BLE Sensor will prompt you to confirm loading the new app in replacement of any previously opened apps.

Step 5. Select the appropriate SensorTile.box device from the Device List.

The app will commence monitoring or logging activity and return real time feedback data to the corresponding app screen in ST BLE Sensor.

UM2580 - Rev 6	page 5/40

UM2580

Application descriptions

3Application descriptions

3.1Mode 1 example apps

The ST BLE Sensor bundles the following ready-to-use app scenarios:

•Baby Crying Detector

•Barometer

•Compass and Level

•Data Recorder

•Human Activity recognition

•In-Vehicle Baby Alarm

•Pedometer

•Sensor Fusion - Quaternion

•Vibration Monitor - Training

•Vibration Monitor - Compare

Figure 6. Apps screen

App scenarios with the icon produce immediate outputs on your smartphone in real time.

App scenarios with the icon store sample data on the internal micro SD card.

App scenarios with the icon are reserved for Expert mode.

UM2580 - Rev 6	page 6/40

UM2580

Mode 1 example apps

RELATED LINKS

2 How to use ST BLE Sensor app with SensorTile.box on page 3

3.1.1Baby crying detector

The baby crying detector (BCD) app implements the Fast Fourier Transform (FFT) and artificial intelligence processing to detect baby crying events using the Sensortile.box on-board microphone.

The analysis of the acquired audio is based on the FFT that converts a signal from its original time domain to a representation in the frequency domain.

Figure 7. FFT analysis - from time to frequency domain

The FFT of the audio signal is the result of all the contributions of each frequency and the related magnitude factor generated by the audio signal.

Figure 8. FFT analysis - principles

UM2580 - Rev 6	page 7/40

UM2580

Mode 1 example apps

The FFT feature extraction of the acquired signal is processed by the STM32 MCU which calculates the MEL FFT and the MFCC (MEL frequency cepstral coefficient) parameters sent to the implemented MCU neural network: if a baby crying event is detected, the green user LED on the Sensortile.box board lights up and a warning is sent to the smartphone via Bluetooth.

Figure 9. STBLESensor - baby crying detection process

The neural network is classified as a deep feed forward neural network and its structure is composed of 2 hidden nodes of 100 neurons each.

The tool used to develop the neural network is Keras with an open source high level library written in Python.

Optimization and loading of the neural network on the Sensortile.box has been performed using STM32CubeMX.AI.

Figure 10. STBLESensor - baby crying app neural network

UM2580 - Rev 6	page 8/40

UM2580

Mode 1 example apps

The baby crying app works with the following ST high sensitivity audio sensor and operating parameter settings:

•APP DATA INPUT: microphone audio acquisition

•SENSOR USED: MP23ABS1 MEMS microphone

•SENSOR SETTINGS: 16 KHz sample acquisition

•APP DATA OUTPUT: baby crying/not crying icon