Space applications require the data logging system to be as efficient as possible as this kind of applications are characterised by extremely limited storage and communication budgets. Sensor data compression has a very important role and can be either lossless, or allow data degradation, in order to boost the compression ratio. Reducing data size becomes even more critical if it is required to monitor housekeeping data, because such monitoring requires to continuously gather data about several physical quantities, from several onboard sensors. Since an extremely high compression capability is required, losing some information becomes crucial in order to achieve the monitoring goal. On the other hand, information loss must not compromise data usability, as high compression implies deterioration of compressed data. This paper describes a high-performance, error controlled, lossy compression method based on Rice coding. The proposed solution provides high compression ratio, it allows to specify the maximum error on decompressed data, and its memory footprint and computational power requirements are extremely low. The proposed solution mainly targets slowly oscillating signals; however, the compression ratio achieved for quickly oscillating signals is still competitive, especially considering the extremely small memory footprint and low computational power requirements. The proposed compression method has been evaluated on real sensor data collected by the Ariane 5 launcher. The achieved reduction of data size, considering a maximum error to maximum signal value ratio of about 3%, is of 7000 times for temperature sensor – which is a slowly oscillating signal – and of 23 times for the vibration sensor – which is a quickly oscillating signal.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Rice Coding-Based Lite Compression Algorithm with Controlled Error for Sensor Data


    Contributors:


    Publication date :

    2018-06-01


    Size :

    4098554 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Data Compression and Error-Protection Coding

    British Library Conference Proceedings | 1995


    LITE Data Processing

    Powell, K. A. / Winker, D. M. | British Library Conference Proceedings | 1998