eduzhai > Applied Sciences > Engineering >

Model-Driven Deep Learning for Massive Multiuser MIMO Constant Envelope Precoding

  • Save

... pages left unread,continue reading

Document pages: 5 pages

Abstract: Constant envelope (CE) precoding design is of great interest for massivemultiuser multi-input multi-output systems because it can significantly reducehardware cost and power consumption. However, existing CE precoding algorithmsare hindered by excessive computational overhead. In this letter, a novelmodel-driven deep learning (DL)-based network that combines DL with conjugategradient algorithm is proposed for CE precoding. Specifically, the originaliterative algorithm is unfolded and parameterized by trainable variables. Withthe proposed architecture, the variables can be learned efficiently fromtraining data through unsupervised learning approach. Thus, the proposednetwork learns to obtain the search step size and adjust the search direction.Simulation results demonstrate the superiority of the proposed network in termsof multiuser interference suppression capability and computational overhead.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×