Low hardware-complexity IQ skew tolerant finite field and time domain hybrid adaptive equalization for 400G/800G-ZR coherent transmission

Abstract

In 400G/800G-ZR transmission systems afflicted by substantial inter-symbol interference (ISI) stemming from chromatic dispersion (CD), frequency-domain adaptive equalizers (FD-AEQs) are preferred to time-domain AEQs (TD-AEQs) as the former have relatively lower computation load. However, the reliance of FD-AEQs on computationally expensive fast Fourier transform (FFT)/inverse fast Fourier transform (IFFT) operations poses a major obstacle in the development of low-power 400G/800G-ZR digital coherent transceivers. Furthermore, to mitigate the significant performance degradation due to receiver in-phase and quadrature (IQ) skew, the employment of twice the number of filters becomes imperative to enable individual adjustments of the I and Q tributaries, nearly doubling the number of necessary FFT/IFFT operations and computation load. This prohibitively high computation load poses a formidable obstacle to the realization of low-power and cost-effective transceivers. To address this issue, we introduce a novel low hardware-complexity IQ skew tolerant finite field (FF) and TD hybrid AEQ for 400G/800G-ZR coherent transmission. By optimizing the AEQ design and the Fermat number transform (FNT) based linear convolution scheme, the proposed hybrid AEQ can achieve a remarkable reduction in hardware complexity by more than 70%, compared with the existing IQ skew tolerant FD-AEQ, while preserving the same receiver sensitivity. Both numerical simulations and experiments demonstrate that the performance of the low complexity hybrid AEQ is the same as the FD-AEQ with much higher complexity. The proposed comprehensive and hardware efficient hybrid AEQ scheme is very appealing for 400G/800G-ZR coherent receivers which need to compensate for large ISI and IQ skew under stringent power constraints.