Loading…

QoS constrained power minimization in the MISO broadcast channel with imperfect CSI

In this paper we consider the design of linear precoders and receivers in a Multiple-Input Single-Output (MISO) Broadcast Channel (BC). We aim to minimize the transmit power while meeting a set of per-user Quality-of-Service (QoS) constraints expressed in terms of per-user average rate requirements....

Full description

Saved in:
Bibliographic Details
Published in:Signal processing 2017-02, Vol.131, p.447-455
Main Authors: González-Coma, José P., Joham, Michael, Castro, Paula M., Castedo, Luis
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper we consider the design of linear precoders and receivers in a Multiple-Input Single-Output (MISO) Broadcast Channel (BC). We aim to minimize the transmit power while meeting a set of per-user Quality-of-Service (QoS) constraints expressed in terms of per-user average rate requirements. The Channel State Information (CSI) is assumed to be known perfectly at the receivers but only partially at the transmitter. To solve this problem we convert the QoS constraints into Minimum Mean Square Error (MMSE) constraints. We then leverage MSE duality between the BC and the Multiple Access Channel (MAC), as well as standard interference functions in the dual MAC, to perform power minimization by means of an Alternating Optimization (AO) algorithm. Problem feasibility is also studied to determine whether the QoS constraints can be met or not. Finally, we present an algorithm to balance the average rates and manage situations that may be unfeasible, or lead to an unacceptably high transmit power. •MISO BC under imperfect CSI and subject to QoS constraints is addressed.•Average MMSE-based restrictions establishes lower bounds for the average rates.•Optimum filters are found considering only statistical CSI.•Average MSE feasible region is studied to ensure convergence to the optimum.
ISSN:0165-1684
1872-7557
DOI:10.1016/j.sigpro.2016.09.007