engleski

On Evolutionary Metaheuristic Optimization Approaches in Data-Driven Signal Processing Techniques

Data-driven signal processing techniques are used for many practical applications in various fields [1]. Due to their local-adaptive properties, these techniques enable processing noisy signals, regardless of knowledge of the signal models and characteristics [2]. The performance of the data-driven methods depends on the proper selection of algorithm parameters [3, 4]. Besides the extensive grid search in the parameter space, the advanced evolutionary metaheuristic optimization techniques, such as particle swarm optimization (PSO)-based algorithms and genetic algorithms (GA) [5], can also be an efficient solution for parameter optimization. This work considers some aspects of their implementation.

signal processing ; data-driven algorithms ; evolutionary metaheuristic optimization ; particle swarm optimization ; genetic algorithm

nije evidentirano