This paper reports the first attempt to tune gain values in proportional-integral-derivative (PID) controllers using an optimizer called simulated Kalman filter (SKF) algorithm. Compared to another well-established optimizer, such as particle swarm optimization (PSO), the SKF algorithm is a relatively new optimizer and most importantly, the SKF algorithm has not been applied to parameter tuning of PID controller. A finding based on automatic voltage regulator system suggests that the SKF algorithm has promising potentials to be a good tuning method in PID controller.

An application of simulated Kalman filter optimization algorithm for parameter tuning in proportional-integral-derivative controllers for automatic voltage regulator system

In this paper, a population-based metaheuristic optimization algorithm named Opposition-based Simulated Kalman Filter (OBSKF) is proposed as an adaptive beamforming algorithm. This algorithm is based on Simulated Kalman Filter (SKF), which is a newly introduced optimization algorithm inspired by Kalman filtering. Opposition-based learning is employed in SKF to improve the performance of SKF in adaptive beamforming. Adaptive beamforming algorithm based on OBSKF is compared with SKF and the existing Adaptive Mutated Boolean Particle Swarm Optimization (AMBPSO). Experimental results show that adaptive beamforming algorithm based on OBSKF is better than SKF and AMBPSO.

An Opposition-based Simulated Kalman Filter algorithm for adaptive beamforming

Previously, an angle modulated simulated Kalman filter (AMSKF) algorithm has been implemented for feature selection in peak classification of electroencephalogram (EEG) signals. The AMSKF is an extension of simulated Kalman filter (SKF) algorithm for combinatorial optimization problems. In this paper, another extension of SKF algorithm, which is called binary SKF (BSKF) algorithm, is applied for the same feature selection problem. It is found that the BSKF algorithm performed slightly better than the AMSKF algorithm.

Feature Selection Using Binary Simulated Kalman Filter for Peak Classification of EEG Signals

This paper presents an analysis of simulated Kalman filter (SKF) optimization algorithm. The SKF algorithm is a population-based optimization algorithm and thus, requires the use of agents to perform a search process. In optimization, usually, different number of agent produces different performance in solving optimization problems. In this paper, the performance of SKF is investigated using different number of agent, from 10 up to 1000 agents. Using the same number of fitness evaluations, experimental results indicate that a surprisingly large population size offers higher performance in solving most optimization problems.

An Analysis on the Number of Agents Towards the Performance of the Simulated Kalman Filter Optimizer

Simulated Kalman Filter (SKF) is a population based optimization algorithm inspired by the Kalman filtering method. It had been successfully used for optimization of many engineering problems. In this work SKF is applied for wireless sensor networks (WSN) coverage optimization problem, where the objective is to maximize the area covered by the sensors in a region of interest. Coverage is an important issue in WSN. It is used as one of the measurement metric for a WSN’s quality of service. Many metaheuristics algorithms had been applied to solve this problem. Here, SKF is tested over several WSN and found to be able to perform better than particle swarm optimization (PSO) and genetic algorithm (GA) in improving WSN coverage.

Simulated Kalman Filter Optimization Algorithm for Maximization of Wireless Sensor Networks Coverage

Adaptive beamforming is a technique used to steer the radiation pattern towards the desired signal and cancel out any interference signal by finding the appropriate weights for every element in an array antenna, to achieve maximum signal to interference plus noise ratio (SINR). There are many methods to perform adaptive beamforming and one of the method is to use metaheuristic algorithm, to estimate the weights for individual elements in an array. Over the years, various metaheuristic algorithms have been applied to adaptive beamforming. Some of the metaheuristic algorithms have been modified from the original algorithms to improve the algorithms performance in adaptive beamforming application. A new metaheuristic algorithm named Simulated Kalman Filter (SKF), is inspired by the estimation capabilities of Kalman filter, has not been applied to adaptive beamforming application. Therefore, this research presents the first-time application of SKF algorithm to adaptive beamforming. The SKF algorithm, however, often converge prematurely at local optimum due to lack of exploration, preventing it from finding better solution. A modified version of the SKF algorithm, named Opposition-Based SKF (OBSKF), introduced by K. Zakwan, applies Opposition-Based Learning method to improve the exploration capabilities of SKF algorithm. Moreover, a new modified version of the SKF algorithm named SKF with Modified Measurement (SKFMM) is introduced to further improve the exploration capabilities of SKF algorithm by modifying the measurement-update equation. The SKF, OBSKF and SKFMM is applied to an array antenna with 10 elements arranged linearly with 0.5

Adaptive Beamforming Algorithm Based on a Simulated Kalman Filter

The Simulated Kalman Filter (SKF) algorithm is a newly introduced optimization algorithm inspired by the estimation capabilities of Kalman filter. In this paper, a population based metaheuristic algorithm named Simulated Kalman Filter with Modified Measurement (SKFMM) is proposed for adaptive beamforming application. SKFMM is compared with the existing SKF and OBSKF algorithms for adaptive beamforming. The experimental results show that the SKFMM algorithm can produce better mean Signal to Interference Plus Noise Ratio (SINR) values compared to the current SKF and OBSKF algorithms for adaptive beamforming application, producing statistically significant results.

Kelvin Lazarus

Papers

Adaptive Beamforming Algorithm Based on a Simulated Kalman Filter

An Opposition-based Simulated Kalman Filter algorithm for adaptive beamforming

An Enhanced Simulated Kalman Filter Algorithm and its Application in Adaptive Beamforming