Revna Acar Vural

Performance Evaluation of Evolutionary Algorithms for Optimal Filter Design

In analog filter design, component values are selected due to manufactured constant values where performing an exhaustive search on all possible combinations of preferred values for obtaining an optimized design is not feasible. The application of evolutionary algorithms (EA) in analog active filter circuit design and optimization is a promising area which is based on concepts of natural selection and survival of the fittest. In this paper, the performances of genetic algorithm, artificial bee colony optimization, and particle swarm optimization, which are nature-inspired EA techniques, are evaluated for active filter design. Each algorithm is applied to two different filter structures and performances of them are also evaluated when filter design is realized with components selected from different manufactured series.

Conic Section Function Neural Network Circuitry for Offline Signature Recognition

In this brief, conic section function neural network (CSFNN) circuitry was designed for offline signature recognition. CSFNN is a unified framework for multilayer perceptron (MLP) and radial basis function (RBF) networks to make simultaneous use of advantages of both. The CSFNN circuitry architecture was developed using a mixed mode circuit implementation. The designed circuit system is problem independent. Hence, the general purpose neural network circuit system could be applied to various pattern recognition problems with different network sizes on condition with the maximum network size of 16-16-8. In this brief, CSFNN circuitry system has been applied to two different signature recognition problems. CSFNN circuitry was trained with chip-in-the-loop learning technique in order to compensate typical analog process variations. CSFNN hardware achieved highly comparable computational performances with CSFNN software for nonlinear signature recognition problems.

Investigation of particle swarm optimization for switching characterization of inverter design

Research highlights? In this work, usage of PSO algorithm in electronic circuit design has been investigated. ? For this purpose, the performance of the algorithm has been tested on the design of an inverter considering transient performance. ? Performance criteria of inverter constitute the constraints of PSO. ? Obtained results show that theoretical design of inverter is matched with PSO based design. Having the advantage of being very simple in concept, easy to implement and computationally efficient, particle swarm optimization (PSO) algorithm is a powerful evolutionary computation tool for automated design of integrated circuits. Since circuit design deals with highly complicated nonlinear equations, obtaining optimal solution of these equations due to particular constraints in short time and acceptable error is of prime concern. Simpler structure and better result providing in case of parameter growth makes PSO an ideal candidate for optimal design of circuit topologies. In this work, usage of PSO algorithm in electronic circuit design has been investigated. For this purpose, the performance of the algorithm has been tested on the design of an inverter considering transient performance. Performance criteria of inverter constitute the constraints of PSO. Obtained results show that theoretical design of inverter is matched with PSO based design.

Swarm intelligence based sizing methodology for CMOS operational amplifier

An efficient design of optimal operational amplifier is a cornerstone of an analog design environment. This study introduces a swarm intelligence based methodology for optimal sizing of a CMOS operational amplifier. Actually, analog sizing is a constructive procedure that aims at mapping the circuit specifications into the design parameter values. Specifying design constraints, conflicting design specifications are introduced to optimization algorithm as a constrained problem and CMOS transistor area is aimed to be minimized. Simulation results demonstrate that proposed method not only meets design specifications and satisfies design constraints but also minimizes total MOS area with respect to convex optimization method.

Particle swarm optimization based inverter design considering transient performance

Evolutionary computation is an efficient tool for automated design of digital integrated circuits. Demand for electronic circuit automation has increased due to complexity growth in VLSI circuits. Since circuit design deals with highly complicated nonlinear equations, obtaining optimal solution of these equations due to particular constraints in short time and disregardable error is of prime concern. Simpler structure and better result providing in case of parameter growth makes particle swarm optimization (PSO) an ideal candidate for optimal design of circuit topologies. In this work, usage of PSO algorithm in digital electronic circuit design has been investigated. For this purpose, the performance of the algorithm has been tested on the design of an inverter considering transient performance. Performance criteria of inverter constitute the constraints of PSO. Obtained results show that theoretical design of inverter is matched with PSO based design.

Component value selection for analog active filter using particle swarm optimization

In this work, a PSO based component value selection method is proposed for analog active filter. PSO is an efficient intelligent evolutionary algorithm for solving optimization problems. The aim of this work is to minimize the total design error of a 4th order Butterworth low pass active filter that would be realized with component values which are compatible with E12 series. PSO provides a maximally flat response in the pass band and achieves smaller design error compared with previous methods.

Artificial bee Colony optimization based CMOS inverter design considering propagation delays

Artificial bee colony (ABC) algorithm is a new population based metaheuristic approach inspired by intelligent foraging behavior of honeybee swarm. Since microelectronic circuit design deals with highly complicated nonlinear equations, obtaining optimal solution of these equations due to particular constraints in short time and acceptable error is of prime concern. Simpler structure and better result providing in case of parameter growth makes ABC an ideal candidate for optimal design of circuit topologies. In this work, usage of ABC algorithm in electronic circuit design has been investigated. For this purpose, the performance of the algorithm has been tested on the design of a CMOS inverter considering transient performance. The optimizations of CMOS inverter parameters are carried out using ABC in MATLAB and the accuracy of performance prediction is verified by SPICE simulation (0.25-µm). Performance criteria of inverter constitute the constraints of ABC. Obtained results show that ABC is capable of designing the CMOS inverter in a very short time while satisfying all the design considerations with an acceptable error.

State variable filter design using Particle Swarm Optimization

Having the advantage of being very simple in concept, easy to implement and computationally efficient, Particle Swarm Optimization (PSO) algorithm is a powerful evolutionary computation tool for electronic circuit design. In this work, usage of PSO algorithm in analog active filter design is investigated. For this purpose, the performance of the algorithm has been tested on the design of a 2nd order state variable low pass active filter. PSO algorithm both minimizes the design error and estimates the component values that are compatible with either E24 or E96 series. Compared to conventional design procedure, PSO achieved smaller design error and provides a maximally flat response in the pass band.

A Mixed Mode Neural Network Circuitry for Object Recognition Application

A general purpose Conic Section Function Neural Network (CSFNN) circuitry in Very Large Scale Integration (VLSI) has been designed for an object recognition application. CSFNN is capable of making open and closed decision regions by combining the propagation rules of Radial Basis Functions (RBF) and Multilayer Perceptrons (MLP) on a single neural network with a unique propagation rule. Chip-in-the-loop learning technique was used during the training process. Utilizing mixed-mode hardware techniques, the inputs of the network and the feedforward signals are all analog while the control unit and storage of the network parameters are fully digital. CSFNN circuitry architecture is problem independent and consists of 16 inputs, 16 hidden layer neurons and 8 outputs. Inheriting the merits of CSFNN, the circuitry has good recognition performance on several objects with invariance to pose, lighting, and brightness. The designed hardware achieved a good recognition performance by means of both accuracy and computational time comparable to CSFNN software.

Optimized analog filter approximation via evolutionary algorithms

In this work, fast and efficient evolutionary algorithms, Artificial Bee Colony (ABC), Particle Swarm Optimization (PSO), Differential Evolution (DE), Harmony Search (HS) are used to optimize the denominator coefficients of the low-pass filter transfer function. Optimum selection of the coefficients will approximate the transfer function to ideal characteristic. Three different order of transfer functions are taken into consideration. Compared to conventional methods, all evolutionary algorithms obtain less approximation error in a short computation time.