Single-Carrier Frequency Domain Equalization for Hammerstein Communication Systems Using Complex-Valued Neural Networks

Hong, Xia; Chen, Sheng; Harris, C.J.; Khalaf, Emad

doi:10.1109/tsp.2014.2333555

Cited by 15 publications

(31 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…which is identical to the nonlinear FD equalization (NFDE) solution of [3]. For the iterations l ≥ 2, we have…”

Section: Nifddfe For Hammerstein Channelsmentioning

confidence: 99%

“…The CV B-spline neural network approach [1]- [3] offers an effective means for identifying and inverting this Hammerstein channel. We first point out that Ψ( ) meets the following conditions.…”

Section: A CV B-spline and Polynomial Models For ψ( ) 1) Cv B-splinementioning

confidence: 99%

“…We adopt the same pseudo training data approach of [2], [3], by replacing w k with its estimate w k = Ψ B (x k ) or w k = Ψ P (x k ) based on the identified HPA's nonlinearity Ψ B ( ) or Ψ P ( ).…”

Section: ) CV Polynomial Inverting Modelmentioning

confidence: 99%

“…CV B-spline neural network has widely been used as an effective means for identification and inversion of CV Hammerstein systems [1]- [3]. Compared to its conventional polynomial based counterpart, B-spline models are proven to have the optimal stability or numerical robustness [4]- [6], and achieve superior performance in challenging practical applications [1]- [3], while maintaining a similar computational complexity.…”

Section: Introductionmentioning

confidence: 99%

“…Compared to its conventional polynomial based counterpart, B-spline models are proven to have the optimal stability or numerical robustness [4]- [6], and achieve superior performance in challenging practical applications [1]- [3], while maintaining a similar computational complexity. In this paper, we review the CV B-spline neural network model as an effective means for identifying and inverting practical Hammerstein systems.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Comparative Performance of Complex-Valued B-Spline and Polynomial Models Applied to Iterative Frequency-Domain Decision Feedback Equalization of Hammerstein Channels

Chen

Hong

Khalaf

et al. 2017

IEEE Trans. Neural Netw. Learning Syst.

Self Cite

View full text Add to dashboard Cite

Abstract-Complex-valued (CV) B-spline neural network approach offers a highly effective means for identifying and inverting practical Hammerstein systems. Compared to its conventional CV polynomial based counterpart, CV B-spline neural network has superior performance in identifying and inverting CV Hammerstein systems, while imposing a similar complexity. This paper reviews the optimality of CV B-spline neural network approach. Advantages of B-spline neural network approach as compared to polynomial based modeling approach are extensively discussed, and the effectiveness of CV neural network based approach is demonstrated in a real-world application. More specifically, we evaluate the comparative performance of the CV B-spline and polynomial based approaches for the nonlinear iterative frequency-domain decision feedback equalization (NIFDDFE) of single-carrier Hammerstein channels. Our results confirm the superior performance of the CV B-spline based NIFDDFE over its CV polynomial based counterpart.Index Terms-Complex-valued (CV) B-spline neural network, CV polynomial model, identification and inversion of Hammerstein channels, nonlinear iterative frequency-domain decision feedback equalization I. INTRODUCTION In many real-world applications, the underlying system that generates complex-valued (CV) signals can be modeled by the CV Hammerstein model. The system is grey-box, as its structure is known to be consisting of an unknown static nonlinearity followed by an unknown linear dynamic model. A well-known example of CV Hammerstein systems is the single-carrier (SC) block transmission communication channel with nonlinear high power amplifier (HPA) at transmitter, whereby the CV static nonlinearity of the Hammerstein system is constituted by the nonlinear transmit HPA, and its linear dynamic subsystem is the dispersive channel which can usually be modeled as a finite-duration impulse response (FIR) filter. Effective identification and inversion of CV Hammerstein systems is therefore crucial in these practical applications.CV B-spline neural network has widely been used as an effective means for identification and inversion of CV

show abstract

“…which is identical to the nonlinear FD equalization (NFDE) solution of [3]. For the iterations l ≥ 2, we have…”

Section: Nifddfe For Hammerstein Channelsmentioning

confidence: 99%

Section: A CV B-spline and Polynomial Models For ψ( ) 1) Cv B-splinementioning

confidence: 99%