eduzhai > Applied Sciences > Engineering >

Neural Network-Aided BCJR Algorithm for Joint Symbol Detection and Channel Decoding

  • Save

... pages left unread,continue reading

Document pages: 6 pages

Abstract: Recently, deep learning-assisted communication systems have achieved manyeye-catching results and attracted more and more researchers in this emergingfield. Instead of completely replacing the functional blocks of communicationsystems with neural networks, a hybrid manner of BCJRNet symbol detection isproposed to combine the advantages of the BCJR algorithm and neural networks.However, its separate block design not only degrades the system performance butalso results in additional hardware complexity. In this work, we propose a BCJRreceiver for joint symbol detection and channel decoding. It can simultaneouslyutilize the trellis diagram and channel state information for a more accuratecalculation of branch probability and thus achieve global optimum with 2.3 dBgain over separate block design. Furthermore, a dedicated neural network modelis proposed to replace the channel-model-based computation of the BCJRreceiver, which can avoid the requirements of perfect CSI and is more robustunder CSI uncertainty with 1.0 dB gain.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×