Botoks is a batteryless timekeeping sensor that operates intermittently harvesting ambient energy. Botoks's hardware and software components enable IoT applications that require precise and intermittency-safe timekeeping. Both hardware design and software stack are open-source.
The cascaded hierarchical remanence timekeeper (CHRT) embedded on Botoks is a novel remanence timekeeper architecture. In principle, capacitive remanence timekeepers are simple RC circuits whose energy level is converted into time. To know more about them and the CHRT, refer to our paper (to be made public) and the docs.
The sensor is based on an MSP430 ultra-low-power, FRAM-enabled microcontroller (MSP430FR5994), and currently the software stack only supports this MCU.
apps/ ## Applications' source code bin/ ## Compiled applications config/ ## Various configuration files docs/ ## Source documentation external/ ## MSP430 base drivers libs/ ## Core Botoks's libraries platforms/ ## GPIO pin mappings scripts/ ## Botoks's calibration scripts
Botoks's code can be built installing the MSP430-GCC toolchain, or using a preconfigured Docker container. Opting for the Docker container has the advantage of not having to install the toolchain, but requires you to mount the directory of the project into the container.
Executables can be uploaded and debugged using UniFlash or Code Composer Studio (CCS). The former method is easier to script, whilst the latter gives you access to a graphical IDE.
The following versions of CMake, MSP430-GCC and UniFlash were tested, but other versions might work as well.
Building with the Docker container
You can use the
build_with_docker.sh script to build all applications inside a
Docker container. Install
docker on your machine and start/enable the
docker daemon. Moreover, make sure that you can
docker commands without
root privileges (check here). The script uses this Docker
image to build applications inside a Docker container
pre-configured with CMake and the MSP430-GCC toolchain.
You can pass
-t <target> to the script to specify a target for
instance, to build all projects and install the generated executables in the
bin/ folder of this repository, run
$ ./build_with_docker.sh -t install
Building without the Docker container
First of all, install CMake and the MSP430-GCC toolchain. CMake can be
installed using your OS's package manager, though the latest version might not
be available. In that case, download the binaries using the link above. As for
the GCC toolchain, you can use the provided
install_toolchain.sh script to
download and install toolchain and support files.
To install, do
$ MSP430_GCC_OS=linux64 # can be 'linux32' or 'macos' instead $ INSTALL_PREFIX=~/ti # where to install the toolchain $ ./install_toolchain.sh
This will download toolchain and other support files from TI's website, and
install them at
$INSTALL_PREFIX/msp430-gcc. Then, assign the environment
MSP430_TOOLCHAIN_PATH the absolute path to the root directory of the
$ export MSP430_TOOLCHAIN_PATH=~/ti/msp430-gcc
Finally, to build all projects, do
$ git clone https://github.com/TUDSSL/Botoks.git $ cd Botoks $ mkdir build && cd build $ cmake .. $ make
make install to build all projects and install the generated
executables in the
bin/ folder of this repository. To build applications
First, connect Botoks to a debugger capable of debugging the MSP430 line of products. Then power Botox using the auxiliary power connector on the PCB. The voltage provided needs to be between 3.4 and 5V.
To upload an application install UniFlash. As of now, all scripts
assume UniFlash is installed at
/opt/ti/uniflash. From the project's
directory root, run:
$ ./flash.sh bin/<app-name>.out
The serial output can now be monitored using your favorite serial monitor (e.g.,
picocom) with a baudrate of 19200.
$ picocom /dev/ttyACM1 -b 19200 --imap lfcrlf
For the calibration procedure please refer to the docs. Running the
calibration procedure requires a licensed version of MATLAB and the
The Docker container comes with all the dependencies to build the documentation. Simply run
$ ./build_with_docker.sh -t docs
and find the HTML documentation in
The hardware is designed in KiCad, however, PDF schematics and gerbers are available in the hardware folder.