Backlog-adaptive compression: Age of information

The end-to-end delay of streaming source coding is characterized by an age of information (AoI) metric that measures the number of symbol periods the decoder output lags behind the encoder input. The source encoder receives input source symbols one per unit time and sequentially outputs binary codewords to a constant rate channel that transmits bits to the decoder. We examine a system in which knowledge of the busy/idle state at the channel interface enables the encoder to switch among codebooks with different source blocklengths based on the backlog of symbols at the encoder. We start by introducing two source sequence parsing policies and show that in each of them the blocklength process can be modeled by a Markov chain. We show by experiments that blocklength adjustment based on the channel interface state provides lower average age than codes with fixed blocklength. Aiming to avoid unnecessary frequent blocklength changes by the encoder backlog, we propose maximum blocklength control scheme at the encoder to further reduce the average age.

[1]  Sergio Verdú,et al.  Optimal Lossless Data Compression: Non-Asymptotics and Asymptotics , 2014, IEEE Transactions on Information Theory.

[2]  Anant Sahai,et al.  The error exponent with delay for lossless source coding , 2006, 2006 IEEE Information Theory Workshop - ITW '06 Punta del Este.

[3]  Anthony Ephremides,et al.  Age of information under random updates , 2013, 2013 IEEE International Symposium on Information Theory.

[4]  Roy D. Yates,et al.  Update or wait: How to keep your data fresh , 2016, IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications.

[5]  Roy D. Yates,et al.  Timeliness in Lossless Block Coding , 2016, 2016 Data Compression Conference (DCC).

[6]  Eytan Modiano,et al.  Optimizing age-of-information in a multi-class queueing system , 2015, 2015 IEEE International Symposium on Information Theory (ISIT).

[7]  Michael B. Baer Source Coding for Quasiarithmetic Penalties , 2006, IEEE Transactions on Information Theory.

[8]  Roy D. Yates,et al.  Real-time status: How often should one update? , 2012, 2012 Proceedings IEEE INFOCOM.