Diffusion Hammerstein Spline Adaptive Filtering Based on Orthogonal Gradient Adaptive Algorithm

In this paper, we propose a class of nonlinear diffusion filtering based on Hammerstein function with the spline adaptive filter (HSAF) implemented by normalised version orthogonal gradient (NOGA) algorithm over distributed network. Diffusion adaptation approximates variable vector help network agents using joint optimisation sum cost function. A HSAF comprises memoryless during learning interpolating polynomials respect to linear filter. We derive framework motivated from NOGA algorithm; called DHSAF-NOGA. There are two types strategies combine-then-adapt (CTA) and adapt-then-combine (ATC) that considered DHSAF-NOGA algorithm. The stability performance mean square error networks is derived. Experiment results depict proposed CTA-DHSAF-NOGA ATC-DHSAF-NOGA algorithms can learn robustly underlying model compared non-cooperative solution existing techniques.