Data acquisition, storage, filtering, and playback
Publication Date: 2014-May-30
The IP.com Prior Art Database
Page 01 of 4
Title: Data acquisition, storage, filtering, and playback
Larger and larger amounts of geophysical data are acquired, for example in seismic exploration for oil and gas. It is known that geophysical data can usually be compressed after acquisition, very often by a huge amount. If data can be compressed however, it indicates that too much data has been acquired in the first place. Compressive sampling techniques have started to be adopted in geophysical data acquisition, but the amount of acquired data, albeit much reduced to conventional data acquisition, is still very large, such that it would be economically attractive to store only the compressed data. Compressive sampling is based on the identification of an invertible linear transform of the acquired data, such that the representation of the data in the transform domain needs less storage capacity than required for the actual data. This invention is first about acquiring and storing the data in the transform domain representation, i.e. making the data representation in the transform domain an integral part of the acquisition system. Secondly, the data in the transform domain can be filtered, i.e. certain aspects of the compressed data can be enhanced over other aspects of the data by a manipulation of the transform domain data representation. Thirdly, a "data playback system" will be outlined, which can re-create the actual, eventually filtered data as outlined above, at the actual spatial sampling locations. This system actually applies the inverse transform mentioned above to the, eventually filtered, compressed representation of the data. The data replay system however will also have the ability to create interpolated data at spatial locations which had not been covered by sensors. This can be used to create for example data on a regular grid as required by many data processing algorithms. The interpolation is based on the inverse to the linear transform that would have been used if the data would have been acquired in the interpolated locations.
Conventionally, geophysical data, for example seismic data, is stored as acquired, i.e., each seismic trace is stored in digital format, usually a 32 bit floating point format, for each individual sensor, or analog group-formed set of sensors. The purpose of geophysical data acquisition traditionally is to directly acquire a "meaningful" signal, usually requiring a very fine sampling of the later. Modern data acquisition systems in this way create huge amounts of data, at considerable cost in storing the data. Recent developments in sampling theory, i.e., especially in the field of compressed sampling, generalize the sampling process to any linear measurement of the signal and allow under certain circumstances to reduce the amount of data to be acquired, as well as a further reduction in the amount of digital information about the data, from which the actual data can be reconstructed using computer...