Fast Decoding of ARINC 717 Flight Data Recordings
Florian Schwaiger, Florian Holzapfel · AIAA Scitech 2021 Forum · 2021
View Video Presentation: https://doi.org/10.2514/6.2021-1982.vid To support big data applications for flight data analysis, we present an improved method for the decoding of binary ARINC 717 recorder files. Our tool transcribes flight parameters from any given dataframe layout of the raw file into one of several tabular open data standards (e.g. .parquet, .csv). Our improved method cuts execution time by a factor of up to 61 and memory utilization by a factor of up to 15 – compared to the previous implementation presented in 2016. It basically removes the performance bottleneck from the data intake stage for big data cloud platforms, as demonstrated in the SafeClouds.eu project. It can further decode much larger files than before, e.g. 2 GB on a desktop PC system, thanks to several performance optimization techniques. Thus, we can more easily parallelize and distribute the tool to remote compute clusters. Further, the tool provides new features, such as the estimation of the most important flight parameters from an undefined dataframe layout. Beyond simple detection of corrupt subframes, the tool can repair some parameters on bit level, e.g. fixing the alignment of multipart parameters (e.g. GPS position) or unrolling wrapped values (e.g. radio altitude).