Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
Friend, that's depend on what kind of processor your application system to be designed. If DSP is selected than I would like to suggest DB6 wavelets. DB6 has the capability to pick up 2-nd order discontineuty ehich is good enough for you to look into quite detail signal variation. The level of detail of signal analysis using wavelets also depand on it decomposition level, for a real-time application using TI DSP, 3-level of decomposition should be good enough, if your signal frequency is not high. For implementation to real-time operation, if integer processor is selected than you should implement the wavelet filter bank using lifting scheme that makes the transform faster. You may consider other DB filters, like DB4 or DB8 to suite your application requirements. Avoid to use Haar filter, even though it is simple and fast in tern of execution and computational requirement for real-time application but lots of noise will be superimposed with your transform representation. Check with IEEE Power Magazine that are some papers on this topic.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
