This library serves as the C wrapper for RadarSimCpp, offering a C interface to access its core functions, including GPU parallelization. For instructions on how to use this library, please refer to RadarSimM.
- High-fidelity radar simulation with comprehensive modeling capabilities
- C-compatible API with opaque pointer pattern for C++ interoperability
- Automatic memory management with cleanup registry for leak prevention
- GPU acceleration support for high-performance simulations
- Thread-safe operations with comprehensive error handling
- Cross-platform compatibility (Windows, Linux)
RadarSimLib includes automatic memory management with three compatibility modes:
- Lock-free implementation using STL containers
- Good balance of performance and compatibility
- Suitable for most single-threaded and multi-threaded applications
- Adds mutex locking for complete thread safety
- Use when multiple threads create/free objects simultaneously
- Enable with:
cmake .. -DRADARSIM_THREAD_SAFE=ON
- C-style arrays, no STL dependencies, no mutex
- Ideal for embedded systems or applications with strict requirements
- Enable with:
cmake .. -DRADARSIM_SIMPLE_CLEANUP=ON
Key Benefits:
- Objects created with
Create_*functions are automatically registered for cleanup - Manual cleanup with
Free_*functions is still recommended and will unregister objects - Automatic cleanup occurs when the DLL/shared library is unloaded
- Same API works across all modes - only internal implementation changes
See COMPILATION_OPTIONS.md for detailed configuration guide.
Building RadarSimLib requires to access the source code of RadarSimCpp.
- Windows CPU
build_win.bat --arch=cpu- Windows GPU
build_win.bat --arch=gpu- Linux CPU
./build_linux.sh --arch=cpu- Linux GPU
./build_linux.sh --arch=gpu