Deep neural network-aided soft-demapping in coherent optical systems: regression versus classification

HIGHLIGHTS

SUMMARY

    Over the past few years, the "conventional" equalizers/soft-demappers have started to evolve toward the designs incorporating machine_learning techniques,, The authors call this category of NN models based on MMSE as a regression equalizer (Reg.), but the authors can term them as a dual-stage soft-demapping (i.e. the NN MMSE post-equalizer followed by the AWGN demapper). Such a device is convenient insofar as the transmitted signal in digital communications is usually generated from a discrete finite-size constellation, e_g quadrature amplitude modulation, QAM. This type of receiver model, named in . . .

     

    Logo ScioWire Beta black

    If you want to have access to all the content you need to log in!

    Thanks :)

    If you don't have an account, you can create one here.

     

Scroll to Top

Add A Knowledge Base Question !

+ = Verify Human or Spambot ?