• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.3A
• /
• pp.375-382
• /
• 2000

Equalization of satellite communication using Complex-Bilinear Recurrent Neural Network(C-BLRNN) is proposed in this pater. Since the BLRNN is based on the bilinear polynomial and it has been more effectively used in modeling highly nonlinear systems with time-series characteristics than multi-layer perception type neural networks(MLPNN) , it can be applied to satellite equalizer. the proposed C-BLRNN based equalizer for M-PSK with a channel model is compared with Volterra filter Equalizer, DFE, and conventional Complex MLPNN Equlizer. The results show that the proposed C-BLRNN based equalizer gives very favorable results in both of MSE and BER criteria over other equalizers.

• Journal of applied mathematics & informatics
• /
• v.25 no.1_2
• /
• pp.293-304
• /
• 2007

In this paper, we present a Krylov subspace based projection method for reduced-order modeling of large scale bilinear multi-input multi-output (MIMO) systems. The reduced-order bilinear system is constructed in such a way that it can match a desired number of moments of multi-variable transfer functions corresponding to the kernels of Volterra series representation of the original system. Numerical examples report the effectiveness of this method.

• Ocean Systems Engineering
• /
• v.4 no.2
• /
• pp.117-136
• /
• 2014

The increased interest in the design of energy efficient ships post IMO regulation on enforcing EEDI has encouraged researchers to reevaluate the numerical methods in predicting important hull design parameters. The prediction of added resistance and stability of ships in the rough sea environment dictates selection of ship hulls. A 3D panel method based on Green function is developed for vessel motion prediction. The effects of parametric instability are also investigated using the Volterra series approach to model the hydrostatic variation due to ship motions. The added resistance is calculated using the near field pressure integration method.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.1386-1388
• /
• 1996

In this paper, a recurrent neural network using polynomial is proposed for electric load forecasting. Since the proposed algorithm is based on the bilinear polynomial, it can model nonlinear systems with much more parsimony than the higher order neural networks based on the Volterra series. The proposed Bilinear Recurrent Neural Network(BLRNN) is compared with Multilayer Perceptron Type Neural Network(MLPNN) for electric load forecasting problems. The results show that the BLRNN is robust and outperforms the MLPNN in terms of forecasting accuracy.