Frequency sampling design of 2-D FIR filters based on DEPSO

Low power 2D finite impulse response filter design using modified artificial bee colony algorithm with experimental validation using field‐programmable gate array

“…The inclusion of HD as an objective, leads to reduction in the switching activity and hence the PC of the designed filter. The reduction in the PC for the filters designed with the aim of simultaneously minimising ripples and power (15) has been compared in Table 5 with the filters designed with the objective of minimising only ripples (14). The significant reduction is attributed to the reduction in switching activity.…”

“…All the existing works on evolutionary optimisation-based 2D FIR filter design have only concentrated on achieving the ideal frequency response characteristics. Motivated by the importance of minimising PC in CMOS-based digital systems, as outlined in the previous section, the objective function (14) has been reformulated to incorporate a component for minimising power. In this regard, the objective function is reframed as…”

“…Windowing is the simplest technique for magnitude response but it fails in linear phase for 2D filters [6]. In addition to the classical techniques, some optimisation-based approaches have also been successfully applied for 2D FIR filter design [3,[9][10][11][12][13][14][15][16][17][18].…”

“…Evolutionary algorithms (EAs) because of their ability to mimic the intelligence of natural selection and adoption process found in biological species, have been successfully applied for the solution of complex non-linear optimisation problems in different engineering fields including FIR filter design. In this regard, the optimisation techniques adopted for 2D FIR filter includes simulated annealing [9], clonal section algorithm [10], genetic algorithm (GA) [11,12], adaptive GA [13], hybrid differential evolution and particle swarm optimisation [14], harmony search [15] and artificial bee colony (ABC) algorithm [3,16,17]. Most of the above reported techniques suffer from the limitations of tuning and selection of a large number of control parameters, pre-mature convergence and large execution time.…”

The design approaches of two-dimensional digital filters based on metaheuristic optimization algorithms: a review of the literature

“…In this paper, firstly, the 1-D multiplierless FRM filter was designed using the harmony search (HS) algorithm, then the 1-D filter was transformed to the 2-D multiplierless circularly symmetric filter. Kuang et al (2012) developed an approach to refine the conventional frequency sampling method and the differential evolution particle swarm optimization algorithm (DEPSO) was used to obtain a nearly optimum minimax design. A 2-D fourfold symmetric square lowpass FIR filter was designed using the frequency sampling method and DEPSO algorithm, and the designed 2-D FIR filter was compared with the median filter for the image denoising problem.…”