Minimax design of sparse FIR digital filters

Jiang, Aimin; Kwan, Hon Keung; Zhu, Yanping; Liu, Xiaofeng

doi:10.1109/icassp.2012.6288670

Cited by 1 publication

(2 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The global optimal karush kuhn tucker condition for first order at point ƒ are ∆ = ʎ 1 + ʎ 2 = 0 (26) ʎ 1 − ≤ 0 ʎ 2 -+ ≤ 0 (27) where, ʎ 1 and ʎ 2 ʎ 1 ʎ 2 (28) The other karush kuhn condition are also given as below: 1. if < < , which implies that (∇ ) = 0. is the hessian matrix approximation at point k. When karush kuhn tucker condition is satisfied by ∆ = 0. In IFFA, the fireflies position is calculated by using the equation (11). The fitness function corresponding to it is given by = + × ∆ (35) where is the hessian approximation matrix , which produces disruption × ∆ in the fitness .…”

Section: Convergence Of Iffo Algorithmmentioning

confidence: 99%

“…The techniques in [10], use linear programming to obtain sparse filters while in [11] sparsity is treated as a specification and tries to minimize least square error through l 1 norm minimization. An iterative design methodology is recommended in [12] to reduce reweighted 'l" norm of the coefficients and another iterative method "second arrange cone programming" is discussed in [13] , these algorithms showed the enhancement over [10] and [11]. In [14], the orthogonal Matching Pursuit (OMP) is utilized in designing linear phase sparse FIR filters.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Sparse Finite Impulse Response Low Pass Filter Design using Improved Firefly Algorithm

Srivastava,

Dwivedi,

Nagaria

2019

IJEAT

View full text Add to dashboard Cite

In this work, optimal sparse linear phase Finite impulse response filters are designed using swarm intelligence-based Firefly optimization algorithm. Filters are designed to meet the desired specification with fixed and variable sparsity. The objective function is formulated consisting of three parameters, i. e., maximum passband ripple, maximum stopband ripple and stopband attenuation. The effectiveness of the proposed method is evaluated in two stages. In first stage, the designed filters have been compared with non- sparse in terms of deviation in their specification. The Comparative analysis depicts that the proposed approach of sparse linear phase FIR filter design method performs better than the conventional methods without significantly deviating from the desired specification. The proposed designed filter is then implemented on xilinx ISE14.7(Vertex7) design environment and their performance is compared in terms of time delay, resource utilizaion and frequency of operation. In the second stage, designed sparse FIR filters are compared with earlier state of art sparse FIR filters design techniques.

show abstract

Section: Convergence Of Iffo Algorithmmentioning

confidence: 99%