(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Implementation of wavelet based digital fir filter using dsp board

Implementation of wavelet based digital fir filter using dsp board / Then Kok Hung
Dengan evolusi yang pantas dalam teknologi bagi semikonduktor dan komputer digital, tugas pemprosesan isyarat seperti penurasan yang dilakukan oleh sistem analog tradisional telah digantikan dengan sistem DSP yang mempunyai kos yang lebih rendah, fleksibiliti dan kebolehpercayaan. Digital penuras FIR adalah yang paling popular dan biasa digunakan dalam aplikasi penurasan kerana ciri-ciri fasa linear dan kestabilannya. Tetapi, digital penuras FIR memerlukan peringkat yang lebih tinggi untuk memenuhi spesifikasi yang dikehendaki berbanding dengan digital penuras IIR. Maka, objektif bagi projek ini adalah merekabentuk penuras FIR yang berasaskan riak dalam C5510 DSK kerana kebaikan bagi sifat penempatannya. Iaitu, reputan jelmaan Penjelmaan Fourier mereka menurun dengan cepat di tindakbalas magnitud. Jadi, mereka mempunyai satu penggalan lebih tajam dan pelemah minimum jalur penahan lebih tinggi dan sesuai untuk penurasan FIR. Teknik paling mudah, kaedah peningkapan telah digunakan untuk melaksanakan riak berdasarkan FIR dalam projek ini. Riak Haar telah digunakan untuk merekabentuk penuras FIR manakala tingkap bagi segi empat tepat, Tingkap Hanning dan Blackman telah digunakan untuk merekbentuk penuras FIR yang berdasarkan tingkap untuk tujuan perbandingan. Rekabentuk dibangunkan di MATLAB sebelum pelaksanaan perkakasan penuras FIR dalam C5510 DSK yang bekerja di bawah CCS IDE. Kemudian, kod untuk penapis FIR ditulis dengan penggabungan bahasa C dan bahasa perhimpunan dalam CCS. Selepas itu, satu isyarat input yang dihasilkan dalam MATLAB digunakan untuk menilai prestasi penuras FIR berdasarkan pelemah minimum jalur penahan dan kelebaran peralihan dalam tindakbalas magnitudnya. Berdasarkan keputusan diperolehi dalam kedua-dua MATLAB and CCS, Riak Haar penuras FIR mempunyai satu peralihan yang sempit tetapi dengan pelemah minimum jalur penahan yang lebih kecil disebabkan sifat fungsi riak itu sendiri. Dalam keseluruhan,objektif projek ini telah tercapai sejak pelaksanaan penuras FIR berasaskan gelombang kecil telah berjaya. _______________________________________________________________________________________________________ With the fast evolution in semiconductor and digital computer technologies, most of the signal processing tasks like filtering that performed by traditional analogue system have been replaced by the DSP systems which have a lower cost, greater flexibility and reliability. Digital FIR filters are the most popular and commonly employed in the filtering application due to their linear phase characteristics and stability. However, higher filter order is required by digital FIR filter to meet the desired specifications compared to digital IIR filter. Hence, the objective of this project is to develop a wavelet-based FIR filter in C5510 DSK because the advantage of the localization property of wavelets. That is, their Fourier transforms decay rapidly in magnitude response and hence they have a sharper cut-off and higher stopband attenuation that suitable for FIR filtering. The simplest technique, windowing method has been used to implement the wavelet based FIR filters in this project. The classic and simplest wavelet, Haar wavelet was used to design the wavelet based FIR filter while Rectangular, Hanning and Blackman window were used to design the window based FIR filters for comparison purpose. The designs were developed in MATLAB first before the hardware implementation of the FIR filters in C5510 DSK that works under CCS IDE. Then, the coding for FIR filters was written by using a mixture of C and assembly language in CCS. After the algorithm of FIR filters is developed, an input signal that generated in MATLAB was used to evaluate the performance of the designed FIR filters based on the minimum stopband attenuation and the transition width in their magnitude response. Based on the results obtained in both MATLAB and CCS, Haar wavelet FIR filter has a narrow transition bandwidth but at the expense of smaller minimum stopband attenuation due to the characteristics of the function itself. In overall, the objective of this project has been achieved since the implementation of wavelet based FIR filter is successful.
Contributor(s):
Then Kok Hung - Author
Primary Item Type:
Final Year Project
Identifiers:
Accession Number : 875005086
Language:
English
Subject Keywords:
semiconductor; analogue; DSP
First presented to the public:
6/1/2013
Original Publication Date:
11/4/2019
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 107
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2019-11-29 15:55:29.778
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Implementation of wavelet based digital fir filter using dsp board1 2019-11-29 15:55:29.778