awr2944-dca-lab 0.1.0

Research toolkit for AWR2944EVM + DCA1000EVM raw ADC radar capture workflows.

awr2944-fmcw-radar

A Python research toolkit for TI AWR2944EVM + DCA1000EVM raw ADC radar captures.

This package provides a robust, native direct-capture pipeline that is free from mmWave Studio GUI automation and Lua scripts.

Installation

Install the base package via pip:

pip install awr2944-dca-lab

To include the optional MATLAB viewer bridge dependencies (Windows only), install with the viewer extra:

pip install "awr2944-dca-lab[viewer]"

Prerequisites

This Python package controls and orchestrates external hardware and software. The following are external prerequisites and must be installed separately:

• MATLAB (Required for the viewer component)
• TI mmWave SDK tools
• DCA1000 CLI software

Architecture

The production capture chain uses:

1. SDK Demo UART CLI for radar configuration
2. TI DCA1000 CLI utilities (external dependency) for FPGA initialization
3. Native direct UDP capture with zero-copy stream processing and metadata logging
4. Sequence/counter validation and DCA depadding
5. Canonical ADC cube extraction
6. Python DSP and standalone MATLAB viewer buildMmwsCompatibleShell.m

Historical mmWave Studio GUI automation is available as an optional legacy-mmws dependency extra for compatibility and debugging.

Reference documents

TI PDFs in reference_docs/:

• AWR2944EVM user guide (SPRUJ22C)
• DCA1000 + mmWave Studio raw capture training
• SWRA581B ADC raw data capture app report
• mmwaveSensing FMCW offline viewing deck (radar formulas)