Open AccessOn the capacity of a class of K ‐user Gaussian broadcast channel with states known at the transmitter
Author(s) -
Ghabeli Leila
Publication year - 2018
Publication title -
iet communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.355
H-Index - 62
eISSN - 1751-8636
pISSN - 1751-8628
DOI - 10.1049/iet-com.2017.0300
Subject(s) - transmitter , channel (broadcasting) , computer science , code word , gaussian , transmission (telecommunications) , superposition principle , channel capacity , topology (electrical circuits) , channel state information , code (set theory) , block (permutation group theory) , decoding methods , state (computer science) , telecommunications , algorithm , mathematics , wireless , combinatorics , physics , mathematical analysis , quantum mechanics , programming language , set (abstract data type)
This studies the capacity of a class of K ‐user Gaussian broadcast channels with additive i.i.d. Gaussian state sequences which are non‐causally known at the transmitter (GBC‐SKT). The transmitter wishes to reliably communicate both a common and a private message to each receiver, thus attempting to simultaneously pre‐code the common message against different states presented in channel outputs. The authors first derive the approximate capacity for the 2‐user GBC‐SKT to within 1.25 bits per channel use (bpcu) for all channel parameters and state variances. Then they extended the result to derive the approximate capacity for a class of K ‐user GBC‐SKT to within 1.5 bpcu. In the capacity‐approaching strategy, the channel input is obtained as the superposition of two codewords: a bottom codeword treating the channel state as noise and a top codeword pre‐coded against the channel state at each user for a portion of the transmission block length or power. For the outer bound, they use the fact that the capacity should be decreasing in state variance in order to get to the tighter outer bound for high values of state variances.