A method for informed selection of memory-length and nonlinearity-order parameters in Volterra-Wiener systems from exponential sweep excitations.
José Luis Blanco MurilloVirginia Yagüe-JiménezPublished in: Multidimens. Syst. Signal Process. (2018)
Keyphrases
- cost function
- high accuracy
- sensitivity analysis
- fine tuning
- detection method
- selection algorithm
- data sets
- input data
- parameter space
- clustering method
- probabilistic model
- dynamic programming
- computational complexity
- preprocessing
- video sequences
- neural network
- parameter selection
- memory usage
- fixed length
- maximum likelihood estimation
- data structure
- correlation coefficient
- convergence rate
- significant improvement
- computational cost
- support vector machine
- pairwise
- parameter estimation
- model selection
- em algorithm
- artificial neural networks