striqt provides a python-based toolset for rapid prototyping of bespoke real-time signal analyzers (RTSAs).

Basic Usage

Command line

Once a striqt environment is installed and activated, the following scripts are installed into the environment PATH, so they can be run from any working directory.

• sensor-sweep: Acquire and analyze a capture sequence (sweep) according to a YAML input file specification. Output datasets are serialized to zarr with xarray. This can run locally or on a remote host (-r argument) running serve-sensor.

• serve-sensor: Serve access to local compute and radio resources for sensor-sweep on remote clients.

Detailed usage instructions for each can be discovered with the --help flag.

Python module APIs

This is alpha software. The API may still change without warning.

The API is organized into two python modules that are importable as :

striqt.analysis

• Validated routines to perform real-time signal analysis (RTSA) from complex-valued baseband
  • Spectrum and spectrogram evaluation
  • Empirical statistical distributions
  • Cellular cyclic prefix and synchronization correlators
• Packaging as xarray Dataset objects for easy-to-explore data management
• Archive datasets to zarr for local or cloud storage support
• Optimized CPU or CUDA GPU compute using full-precision float

striqt.sensor

• Acquire data from sensor hardware according to a configuration file
• Concurrent acquisition, analysis, and archival
• Automated sensor calibration based on the Y-factor technique

Installation

The following options require that the host has internet access in order to download dependencies. Installs via local package indexes may require customization that has not been tested.

Option 1: Conda Environment

Installation with radio hardware and GPU support is provided via conda environments. Several variants of a striqt environment are provided here, targeted at different host computing environments.

1. Ensure that conda is installed (or mamba/micromamba, substituted in what follows)
2. Clone this repository
3. Select a predefined environment based on use-case and hardware:
   • environments/cpu.yml: Analyze pre-recorded IQ or run remote control (cross-platform, CPU only)
   • environments/gpu-cpu.yml: Analyze pre-recorded IQ or run remote control (cross-platform, CPU or CUDA GPU)
   • environments/edge-airt.yml: Signal acquisition and analysis to run on AirT/AirStack radios
4. Create the chosen environment:

       conda env create -f <path-to-environment-here.yml>

5. Activate:
   • IDE: select the striqt virtual environment
   • Command line: conda activate striqt

NOTE: The environment operates on an editable install of the modules and command line tools. As a result, if the location of the cloned source code repository is moved, the conda environment needs to be removed and built again according to the instructions above.

Option 2: pip installation

The dependencies, APIs, and CLIs can be installed without radio hardware or GPU support (for post-analysis, plotting, testing, etc.) via pip install. In order to avoid conflicts with other projects, the recommended practice for this is to install into a python virtual environment.

1. Clone this repository
2. pip install <path-to-repository>

Documentation

In keeping with the alpha development status of this codebase, documentation is limited.

• reference for the yaml configuration files that drive edge_sensor
• Docstrings in the source code are the most up-to-date API documentation
• A few examples are located in tests and notebooks. Some of these may not be up to date.

See Also

• Validation and calibration with hardware