Age-Optimal Channel Coding Blocklength for an M/G/1 Queue with HARQ
Paper in proceeding, 2018

We consider a communication system in which a source transmits information updates to a destination node through a binary erasure channel (BEC). When a packet containing an information update, which consists of a fixed number of information bits, arrives at the transmitter, it gets queued in a buffer, to be encoded and sent over the channel. Before transmitting a packet, the transmitter selects a channel coding blocklength n and then uses an automatic repeat request (ARQ) protocol, whereby packets that are decoded incorrectly are repeated. The choice of the coding blocklength thus affects the end-to-end status age. However, this dependency is nontrivial since, on the one hand, the duration of a single transmission attempt is directly proportional to n, so the smaller n the better. On the other hand, a smaller value of the blocklength n yields a higher probability of decoding error, which increases the end-to-end status age. Employing recent finite-blocklength information-theoretic bounds and approximations on the rate achievable on a BEC for a given blocklength and a given error probability, we study the age-optimal design of this system. We find that for any nontrivial BEC, there exists an optimal blocklength that minimizes the average age and average peak age of information.

Author

Hakan Sac

Middle East Technical University (METU)

Tan Bacinoglu

Middle East Technical University (METU)

Elif Uysal-Biyikoglu

Middle East Technical University (METU)

Giuseppe Durisi

Chalmers, Electrical Engineering, Communication, Antennas and Optical Networks

IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC

Vol. 2018-June 8445909

19th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2018
Kalamata, Greece,

SWIFT : short-packet wireless information theory

Swedish Research Council (VR) (2016-03293), 2017-01-01 -- 2020-12-31.

Subject Categories

Telecommunications

Communication Systems

Signal Processing

DOI

10.1109/SPAWC.2018.8445909

More information

Latest update

1/9/2020 8