Blind Receiver Skew Compensation and Estimation for Long-Haul Non-Dispersion Managed Systems Using Adaptive Equalizer

Rios-Müller, Rafael; Renaudier, J.; Charlet, G.

doi:10.1109/jlt.2014.2377582

Cited by 77 publications

(31 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, the complexity required to equalize such signal will scale proportionally to twice the complexity of zeroforcing CD compensation, independently if the equalizer is implemented with all real-valued or all complex-valued filters. This interesting conclusion explains why in [5] and [6] four (instead of the usual two) independent CD compensation complex-valued filters are necessary to enable a low complexity adaptive equalizer to compensate the IQ skew at the receiver, and why in [4] a large number of taps is required in order to recover the signal with the adaptive equalizer. The equalizer number of taps is also the limitation of [3], which fails when the signal has a large value of accumulated CD, as indicated in [6].…”

Section: E Discussion On the Implications Of The Modelsmentioning

confidence: 97%

“…This interesting conclusion explains why in [5] and [6] four (instead of the usual two) independent CD compensation complex-valued filters are necessary to enable a low complexity adaptive equalizer to compensate the IQ skew at the receiver, and why in [4] a large number of taps is required in order to recover the signal with the adaptive equalizer. The equalizer number of taps is also the limitation of [3], which fails when the signal has a large value of accumulated CD, as indicated in [6]. In fact, we can point out the equalizer structures proposed in [3]- [6] are equivalent in the sense that, giving the necessary number of equalizer taps to solve the equalization problem, they should provide similar performance.…”

Section: E Discussion On the Implications Of The Modelsmentioning

confidence: 97%

“…The equalizer number of taps is also the limitation of [3], which fails when the signal has a large value of accumulated CD, as indicated in [6]. In fact, we can point out the equalizer structures proposed in [3]- [6] are equivalent in the sense that, giving the necessary number of equalizer taps to solve the equalization problem, they should provide similar performance.…”

Section: E Discussion On the Implications Of The Modelsmentioning

confidence: 99%

“…A hybrid approach to solve the equalization problem is also possible, as indirectly demonstrated in references [5] and [6]. There the authors propose to modify the ZF equalization of CD and to take advantage of symmetries of the CD impulse response to reduce complexity, further using a 4 × 2 complexvalued adaptive equalizer to compensate for the IQ skew.…”

Section: F On the Wl-zf Equalization Of Iq-mixing And CDmentioning

confidence: 99%

“…Recent works have proposed adaptive equalization architectures robust to receiver front-end imperfections, such as Inphase/Quadrature (IQ) imbalance and IQ skew [3]- [6], here generally referred to as IQ-mixing effects. However, none of these references have presented a detailed analysis on the connection between the receiver IQ-mixing effects and the complexity requirements for equalization algorithms at the receiver.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Widely Linear Equalization for IQ Imbalance and Skew Compensation in Optical Coherent Receivers

Silva

Zibar

2016

J. Lightwave Technol.

104

View full text Add to dashboard Cite

Abstract-In this paper, an alternative approach to design linear equalization algorithms for optical coherent receivers is introduced. Using widely linear complex analysis, a general analytical model it is shown, where In-phase/quadrature (IQ) imbalances and IQ skew at the coherent receiver front-end are naturally included in channel equalization problem. Next, the problem of chromatic dispersion (CD) compensation after imbalanced and skewed coherent detection is analyzed. Based on the analytical models obtained it is demonstrated that, under the presence of such receiver front-end imperfections, the complexity of the channel equalization filter which is able to provide optimal performance, on the minimum mean square error (MMSE) sense, will scale proportionally to twice the complexity of CD compensation, if standard zero-forcing equalization of CD is applied as first equalization procedure. For the last, it is shown that, by applying the widely linear complex analysis, one can derive a complex-valued adaptive equalizer structure which is able to compensate for linear IQ-mixing effects at the receiver front-end. By extensive numerical simulations, the performance versus complexity of the proposed equalizer is shown to match the predictions of the analytical models derived.

show abstract