Differential-Clustering Compression Algorithm for Real-Time Aerospace Telemetry Data

• Main Authors: Xuesen Shi, Yuyao Shen, Yongqing Wang, Li Bai
• Format: Article
• Language: English
• Published: IEEE 2018-01-01
• Series: IEEE Access
• Subjects: Real-time aerospace telemetry data; lossless compression; similarity metric; clustering
• Online Access: https://ieeexplore.ieee.org/document/8478298/

The volume of telemetry data is gradually increasing, both because of the increasingly larger number of parameters involved and the use of higher sampling frequencies. Efficient data compression schemes are therefore needed in space telemetry systems to improve transmission efficiency and reduce the burden of required spacecraft resources, in particular of their transmitter power. In this paper, a differentialclustering (D-CLU) compression algorithm for lossless compression of real-time aerospace telemetry data is proposed. Because of the temporal-spatial correlation characteristics of telemetry data, the use of a differential compression strategy can efficiently improve compression performance. However, differential compression faces two non-negligible problems, reliability and compression ratio, both of which may be solved by clustering. This is the approach pursued in the proposed D-CLU compression algorithm. The algorithm involves both clustering and coding. In the clustering stage, a one-pass clustering method based on a similarity metric is used to group the original data into clusters. In the coding stage, two traditional encoding algorithms, Lempel-Ziv-Welch and run-length encoding, are used to encode the data, based on the clustering results. Compared with the direct use of differential compression, the clustering-based differential compression algorithm can reduce the error propagation range, thus increasing reliability. The experimental results demonstrate that the proposed D-CLU algorithm can also achieve better compression performance than the other existing algorithms.