Movatterモバイル変換


[0]ホーム

URL:


Skip to content

Navigation Menu

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
forked fromhbe72/cdsp

Compositional DSP Library

License

NotificationsYou must be signed in to change notification settings

johnmcfarlane/cdsp

 
 

Repository files navigation

develop @ Travis-CI
(macOS & Linux)
develop @ Appveyor
(Windows)
Build StatusBuild status

CDSP is a header only DSP library written withCNL aiming for floating point andfixed-point implementation of typical DSP kernels.

Algorithm designers often start withMatlab,Scipy,R or maybeJulia and get the algorithm designed on very highlevel with a floating-point datatype. For real-time applications the finaltarget is often fixed-point DSP, FPGA or ASIC. Manual porting of the algorithmto fixed-point is needed which can lead to three separate algorithm sourcecodes: original algorithm in floating-point, fixed-point design of the same andthe final implementation on the target. Keeping them in synchronization is anightmare.

CDSP library shows how withCNL singlesource code can provide full implementation. Minimally, CDSP can provide bothfloating-point and fixed-point designs from the same source code. In the casecompliant C++ compiler exists for the target the final implementation is doneas well.

CDSP fixed-point implementations are measured against their floating pointcounterparts in terms of numerical accuracy. At this point performance withCirrus ADSP2 fixed-point datatypes have been tested and separately verifiedagainst native Cirrus ADSP2 implementation. Composing new datatypewithCNL and specializing the lowestlevel of core CDSP kernels is quite straight forward to support otherarchitectures such as Qualcomm, TI DSPs and other application specificprocessors.

Requirements

Requirements are the same as withCNL.CDSP installsCNL and CDSP side-by-sideand the only prerequisites are listed below.

Linux

Requires:

  • GCC 5.1 / Clang 3.5
  • Cmake 3.2.2

Optional:

  • Boost - for multiprecision support withCNL
  • Doxygen - generates documentation in the doc/gh-pages directory

Mac

See the instructions for Linux. Works similarly.For missing packages please useHomebrew.

Windows

  • MSBuild 15.0 (VS 2017)
  • Cmake 3.8.0

Instructions

Download

The library is hosted onGitHub

cd /some/directorygit clone https://github.com/hbe72/cdsp.git

Build

  1. Generate the build system

    mkdir buildcd buildcmake /some/directory/cdsp -DCMAKE_INSTALL_PREFIX=/directory/to/install -DCMAKE_BUILD_TYPE=Release
  2. Build tests:

    • Linux and Mac parallel with 8 cores
    cmake --build . --target Tests -- -j8
    • Windows
    cmake --build . --target test/Tests
  3. Run the tests

    ctest
  4. Install

    If you want to install the CDSP and CNL libraries to location specified byCMAKE_INSTALL_PREFIX.

    cmake --build . --target install

    By default CDSP and CNL install to /usr/local.

About

Compositional DSP Library

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • C++94.8%
  • CMake5.0%
  • Shell0.2%

[8]ページ先頭

©2009-2025 Movatter.jp