Reversible seismic data compression

This paper is concerned with reversible seismic data compression techniques for transmitting data from exploration field to a research center via satellite. The statistical character istics of seismic data, including Vibroseis data and impulsive data, are examined. Based on these characteristics, various compression techniques, including prediction, orthogonal transforms, and digital coding methods have been investigated. Explicit methods are proposed for compressing the Vibroseis and impulsive data. These methods have been simulated for extensive real data. The compression performance is measured by the compression ratio and the signal-to-noise ratio. The results indicate that a compression ratio ranging between five to one and six to one with at least a 30 dB SNR can be achieved. Facts of real-time design that is, execution time, core size requirement, depth of queue, and transmission time are examined. For transmitting the compressed seismic data using a 7-bit block counter with continuous ARQ, 5000 bits per frame are recommended for a 9600 bps channel bandwidth.