Identification of nonlinear dynamic systems using convex combinations of multiple adaptive radius basis function networks

Zeng, Xianlin; Zhao, Haiquan; Jin, Weidong; He, Zhengyou; Li, Tianrui

doi:10.1016/j.measurement.2012.08.022

Cited by 7 publications

(3 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“… It provides a steady equalization with convergence speed [8] is superior. This is additional showed in the works of [3,[9][10][11][12][13][14].…”

Section: Introductionmentioning

confidence: 66%

“…Following examples showsn in this part for simulation which is intended to assessment of the presentation of the proposed equalizer.Liang & Zhi, 2004 introduced a widely used channel for simulations. The 3rd order channel model having system transfer function is: = 1 -0.9 -1 + 0.385 -2 + 0.771 -3 (10) In the above equations zeros at 0.6 and 0.75 ± j0.85. We have chosen another following non linear channel i.e.…”

Section: Results and Simulationsmentioning

confidence: 99%

See 1 more Smart Citation

Modified Form of PSO and Its Variation in Channel Equalization

Panda*,

Chandran,

et al. 2020

IJRTE

View full text Add to dashboard Cite

Novelty of PSO is the techniques of parameter improvement. Using many “strategy principles” for the PSO is important for its convergence performance and the optimisation job. In the proposed work, we applied PSO and its advanced form trained with our proposed algorithm for channel equalization. Since particle swarm optimization is matured in the literature ,we apply PSO in its optimized structure trained with radial basis function Artificial Neural Network (ANN). Therefore, this work introduces most favourable design of RBFNN equalizers using OPSO. We treat equalization problem as a classification problem. We assessed a set of fitness functions of modified form of PSO and analysed with it to the presentation of existing PSO .

show abstract

“… It provides a steady equalization with convergence speed [8] is superior. This is additional showed in the works of [3,[9][10][11][12][13][14].…”

Section: Introductionmentioning

confidence: 66%

Section: Results and Simulationsmentioning

confidence: 99%

Modified Form of PSO and Its Variation in Channel Equalization

Panda*,

Chandran,

et al. 2020

IJRTE

View full text Add to dashboard Cite

show abstract

“…These algorithms are numerically less complex than Volterra filter and superior performance. Moreover, recently, to improve the convergence speed and tracking capability, and to reduce the misadjustment of the traditional algorithm, some novel nonlinear neural networks were presented based on combination scheme [25][26][27]. Nevertheless, the convergence rate was still slow during the period of convergence transition.…”

Section: Q2mentioning

confidence: 99%