A Study on the Application of Rateless Coding in Non-Cellular MIMO Systems for Machine-Type Communication

Michael Karrenbauer, Sergiy Melnyk, Dennis Krummacker, Andreas Weinand, Hans Dieter Schotten

In: Rolf Findeisen , Sandra Hirche , Klaus Janschek , Martin Mönnigmann (Hrsg.). 21st IFAC World Congress. World Congress of the International Federation of Automatic Control (IFAC-2020) July 11-17 Berlin Germany Seiten 8243-8248 IFAC-PapersOnLine 53 2 Elsevier 2020.


For industrial applications, communication requirements are considerably different than in most other domains. Whereas in office environments and the consumer market networks are designed towards the maximum throughput, in Industry 4.0 scenarios, traffic inficts realtime requirements, demands particular high reliability and low latency. Industrial applications of the past as well as in most current production facilities are interconnected via cable. Wireless communication in this surrounding is still a big challenge, due to the mentioned communication demands. The high demands on reliability and latency require adaptations both to the MAC layer and to the baseband signal processing of the radio system being used. Of particular importance here is the error protection coding of the radio system, the investigation of which is the subject of the given work. This needs to achieve particularly high reliability but also must not consume too much computation time, while introducing only a reasonable amount of redundancy. In order to strive for this goal, this paper examines the utilization of Fountain codes (which belong to the family of rateless codes) with very short packets (in the order of less than 100 bytes). Investigated is the impact in terms of packet error rate. It is numerically shown that the joint erasure- and error-correction coding outperforms the sole usage of either one coding scheme. Interestingly, this is true in the presence of a Rayleigh channel but also in the presence of an AWGN channel. Finally, the performance of such a method is compared to that of space time block coding.

Weitere Links

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence