Electronic dispersion compensation of fiber links using sparsity induced volterra equalizers

Performance evaluation of decision feedback equalizers in fiber communication links

“…The resulted sparse representation, compared to the full terms (dense) design, is more tractable from an implementation point of view, since operators involving zero values can be omitted leading to large computational complexity savings. The sparsification task is performed using the approach introduced in [17], which employs the Sparse Learning via Iterative Minimization (SLIM) algorithm,originally been proposed in the context of radar imaging [22]. In this way, the structure of the equalizer is dynamically determined discarding coefficients that have a marginal contribution to the performance of the equalizer.…”

“…(b) and (f), noting that the cost of the DPSK equalizers is about twice of that required by the OOK counterparts. The performance of the sparsity induced electronic equalizers [17] is illustrated in Fig.1 (c) and (g). Compared to the dense counterparts, the sparsity induced electronic equalizers demonstrate almost identical performance, at a reduced implementation cost, which is about 30% of that required by the dense schemes, as it is indicated in Fig.…”

“…The electronic equalization schemes are used to mitigate either the residual dispersion resulting from incomplete optical compensation or the total dispersion resulting from uncompensated optical transmission. To facilitate real-time implementation, the equalizer structure is dynamically tuned discarding coefficients that have a marginal contribution to the performance of the equalizer leading to a significant computational complexity savings [17]. A comparative study of the performance of the DFE schemes is performed by means of computer simulations, using a typical 10Gb/s optical transmission set up.…”

FPGA implementation of a MIMO DFE IN 40 GB/S DQPSK optical links

“…As a result, linear effects of the optical channel have a non-linear impact. Equalization can be performed by maximum likelihood sequence estimation (MLSE) method [5], or by the use of standard DFE or Volterra DFE counterparts [4,6,7] noting however, that the use of Volterra equalizers still becomes cumbersome when the ISI extends more than a few symbols.…”

Improvement of three-level code division multiplexing via dispersion mapping

“…Compared to their all-optical counterparts, electronic compensation increases flexibility and gives a new impetus to transparent optical networks for adaptive and dynamic handling in cases where the total accumulated dispersion is not known in advance [7] Since 1980, dispersion compensation fibers (DCFs) have been used for mapping and managing dispersion compensation in optical communications [8]. Although new dispersion-shifted fibers and dispersion-flattened fiber have been developed and implemented in some applications, they are still costly and require new fiber cable installment, which tends to increase the system cost [8,9].…”