mmWrt

minimal raytracing for MIMO FMCW radar systems.

Intended usage:

1. educational & experimental

simple install:

1. git clone
2. pip install .

for install for developpers:

1. git clone
2. pip install .[dev]

Release Notes and Roadmap

Released

*v0.0.10:

* adding distance as a function of time in IF compute - allowing for simulation of target changing bin ranges through single chirp
* 0.0.10-pre.6: fixes #3 (numpy deprecating complex_ replacing with complex128)

v0.0.9:

* adding ULA, URA for TDM MIMO
* adding 3D points for TDM MIMO (x,y,z)

v0.0.8:

* adding TDM MIMO
* adding DDM MIMO (including Doppler desambiguation)

v0.0.7

* adding AoA
* adding sparse array initial example (feature dedicated to Amine L.)

v0.0.6:

* added micro-doppler
* added non-regression on .ipynb in docs/ folder

v0.0.5:

* moved dependancies from requirements to setup.py
* added extras [dev] for developpers (and documentation and read the docs)
* moved version checking from setup.py to test_basic.py
* added readthedocs.yaml

v0.0.4:

* adding frequency estimator
* added speed processing
* added support for radar equation (RCS, distance, ...)
* antenna gains in azimumth, elevation and freq

v0.0.3: first release

* point targets only
* 1D compute of baseband if signal for scene
* 1D FFT, CFAR, peak grouping and target position error compute
* single reflections

NEXT ()

* 2D (AoA)
* 2D FFT: range+velocity, range+AoA
* 2D peak grouping (by velocity sign)
* 3D position error compute
* 3D targets (at least spheres)
* medium attenuation
* 3D point clouds (i.e. over multiple CTI)
* multiple single reflections

Not planned yet but considered:

• reads and loads .bin from DCA1000
  • record BB signals in .bin
  • 3D targets and scene rendering with imaging side by side radar
  • Swerling's scattering

Example Code

Check on Google Colab the code:

Or Read the Docs on

On dependencies

• jupyterlab-myst: used for displaying admonition in jupyter notebook (and thus documentation on read the docs)

Release process

0. Ensure all the new .ipynb are added in docs/Hands-on.md

1. run pyroma (should be 10/10)

pyroma .

2. run flake8 runs with darglint settings for docstrings to numpy standard set in the .flake8 file should yield 0 warnings or errors

flake8

3. run pytest should yield 100% pass

pytest

4. run coverage

coverage run -m pytest

5. run coverage report (should be 100%)

coverage report

6. run tox

tox

7.run sphinx-api updates the *.rst in docs/ folder

sphinx-apidoc -f -o docs mmWrt

8. run sphinx-build (updates the read_the_docs folder)

sphinx-build -b html docs build/html

9. release the Release-Candidate to pypi-test check that init.py is "0.0.X-pre.Y" for RC numbering

python setup.py bdist_wheel

twine upload -r testpypi dist*

10. check on Google Colab (Google Colab requires py3.8 as off 2023-Jan-15)

if testing release-candidate need to add --pre -U or will install latest stable version.

!python -m pip install -i https://test.pypi.org/simple/ --pre -U mmWrt
from mmWrt import __version__
print(__version__)

11. merge dev branch with main

git checkout main git merge dev_branch_name

12. update the version to final update init.py to remove the suffix -pre.Y "0.0.X-pre.Y"

13. release on pypi (assumes your pypirc is local to the project)

twine upload -r pypi --config-file=..pypirc dist*

14. check on colab that pypi package works:

!python -m pip install mmWrt from mmWrt import version print(version)

15. check updates on read_the_docs

push to git to trigger readthedocs build: git push navigate to https://readthedocs.org/projects/mmwrt/builds/ ensure build is successful

16. (optional) add tag for release

git tag -a v0.0.X -m "version comment" git push origin v0.0.X

17. then check on google colab dev extras instals works

!python -m pip install mmWrt[dev]