2013 | OriginalPaper | Buchkapitel
Performance Study on Complex-valued Function Approximation Problems
verfasst von : Sundaram Suresh, Narasimhan Sundararajan, Ramasamy Savitha
Erschienen in: Supervised Learning with Complex-valued Neural Networks
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
In this chapter, we evaluate the approximation performances of the fully complex valued multi-layer perceptron network and the improved fully complex-valued multi-layer perceptron network described in Chapter 2, the fully complex-valued radial basis function network and the meta-cognitive fully complex-valued radial basis function network described in Chapter 3, and the fast learning fully complex valued relaxation network described in Chapter 4. The performances of these networks are studied in comparison with existing complex-valued learning algorithms like the complex-valued extreme learning machine and the complex-valued minimal resource allocation network using two synthetic, complex-valued function approximation problems and two real-world problems. The real world problems consist of a Quadrature Amplitude Modulation (QAM) channel equalization problem with circular signals and an adaptive beam-forming problem with non-circular signals.