Digital Signal Processing (DSP)

The DSP API provides an architecture agnostic way for signal processing. Currently, the API will work on any architecture but will likely not be optimized. The status of the various architectures can be found below:

Architecture

Status

ARC ARM ARM64 MIPS NIOS2 POSIX RISCV RISCV64 SPARC X86 XTENSA

Unoptimized Optimized Optimized Unoptimized Unoptimized Unoptimized Unoptimized Unoptimized Unoptimized Unoptimized Unoptimized

Using zDSP

zDSP provides various backend options which are selected automatically for the application. By default, including the CMSIS module will enable all architectures to use the zDSP APIs. This can be done by setting:

CONFIG_CMSIS_DSP=y

If your application requires some additional customization, it’s possible to enable CONFIG_DSP_BACKEND_CUSTOM which means that the application is responsible for providing the implementation of the zDSP library.

Optimizing for your architecture

If your architecture is showing as Unoptimized, it’s possible to add a new zDSP backend to better support it. To do that, a new Kconfig option should be added to subsys/dsp/Kconfig along with the required dependencies and the default set for DSP_BACKEND Kconfig choice.

Next, the implementation should be added at subsys/dsp/<backend>/ and linked in at subsys/dsp/CMakeLists.txt.

API Reference

group math_dsp

DSP Interface.

Typedefs

typedef int8_t q7_t

8-bit fractional data type in 1.7 format.

typedef int16_t q15_t

16-bit fractional data type in 1.15 format.

typedef int32_t q31_t

32-bit fractional data type in 1.31 format.

typedef int64_t q63_t

64-bit fractional data type in 1.63 format.

typedef __fp16 float16_t

16-bit floating point type definition.

typedef float float32_t

32-bit floating-point type definition.

typedef double float64_t

64-bit floating-point type definition.