Type-IV DCT, DST, and MDCT algorithms with reduced numbers of arithmetic operations

Abstract

We present algorithms for the type-IV discrete cosine transform (DCT-IV) and discrete sine transform (DST-IV), as well as for the modified discrete cosine transform (MDCT) and its inverse, that achieve a lower count of real multiplications and additions than previously published algorithms, without sacrificing numerical accuracy. Asymptotically, the operation count is reduced from 2N log 2 N + O(N ) to 17 9 N log 2 N + O(N ) for a power-of-two transform size N , and the exact count is strictly lowered for all N ≥ 8. These results are derived by considering the DCT to be a special case of a DFT of length 8N , with certain symmetries, and then pruning redundant operations from a recent improved fast Fourier transform algorithm (based on a recursive rescaling of the conjugate-pair split-radix algorithm). The improved algorithms for DST-IV and MDCT follow immediately from the improved count for the DCT-IV.

A Removal of Eye Movement and Blink Artifacts from EEG Data Using Morphological Component Analysis

“…Therefore, we used the DCT, LDCT, and DST dictionaries for validations of EEG signal. The detailed significant differences in morphology of DCT and DST were given in past studies [ 56 , 57 ], while in our analysis there were no significant differences. The right combination of redundant transforms to form overcomplete dictionary revealed the desired decomposition in principle.…”

“…UDWT contributes to separating slow and bump morphologies for EOG and EEG transient slow changes, DST is for monomorphic and polymorphic EEG components (major EEG parts), and DIRAC is for spike type activities in transient EEGs. In comparison in the simulated experiment, we used the discrete cosine transform (DCT), discrete sine transform (DST) [ 56 , 57 ], and local discrete cosine transform (LDCT) dictionaries for major EEG parts. For the verification of the hypothesis, the intracranial EEG (iEEG) data was assumed to be “true EEG” signal, which was recorded from the real brain activity, and artificial EOGs including bump and slow changes were introduced and the performance of the accurate reconstruction of the true EEGs was examined.…”

Epilepsy EEG classification using morphological component analysis

“…ζ is assumed to be negligible noise tend to zero. By using three dictionaries, undecimated wavelet transform (UDWT), local discrete cosine transform (LDCT), and Dirac (Kronecker basis) [39,45,46] in this work, coefficients are optimized as…”

A unified VLSI algorithm for a high performance systolic array implementation of type IV DCT/DST

“…The DCT-IV and DST-IV have been computed using fast software algorithms with basically two approaches: the direct approach and the indirect one. In [11], a new algorithm for the efficient software implementation of type IV DCT using the indirect approach has been proposed. It is based on a new fast FFT algorithm that allows the reduction of the arithmetic complexity.…”