Lempel-Ziv-Markov Chain Algorithm Modeling using Models of Computation and ForSyDe

Abstract

The data link is considered a critical function of modern aircraft, responsible for exchanging information to the ground and communicating to other aircraft. Nowadays, the increasing amount of exchanged data and information brings the need for network usage optimization. In this sense, data compression is considered a key approach to make data packages size smaller. Regarding the fact that avionics systems are safety-critical, it is fundamental not losing data nor performance during the compression procedures. In this context, manufacturers and regulatory agencies usually follow DO-178C guidance. Targeting model-based embedded design guidelines, DO-178C includes a supplement document, named DO-331. In this paper, we describe a widely used data compression algorithm, the Lempel-Ziv-Markov Chain algorithm (LZMA). Regarding formal model-based design, we argue that the synchronous dataflow model of computation captures the algorithm behavior more directly. The Formal System Design (ForSyDe) methodology is used to model the LZMA.

Analysis and Identification of Possible Automation Approaches for Embedded Systems Design Flows

“…A comprehensive case study was based on Reference [42] published by the authors of this work. That paper modeled the LZMA compression using the ForSyDe SDF MoC library, but not representing part of the system dynamic behavior.…”

“…A first simplified LZMA model was introduced by the author of the present research in Reference [42] using the SDF MoC. That simplified LZMA modeling considered a set of assumptions to be valid.…”

A Novel Approach to T-Cell Receptor Beta Chain (TCRB) Repertoire Encoding Using Lossless String Compression

“…A compression algorithm is defined as either lossless (i.e, when a compressed file is decompressed, the output matches the original file) or lossy (i.e., when a compressed file is decompressed, the output is epsilonclose to the original data, but not identical). Standard lossless compression algorithms that are being used in many domains [5,4,1,37,15] are based on deriving a token-based mapping to reduce the compressed file size [16,18,7,35,10,3,13]. Yet, there are no published studies attempting to combine these concepts from information theory, explicitly leveraging the various lossless data compression algorithms as feature extractors.…”