Commissariat a L'energie Atomique Et Aux Energies Alternatives

Centre National De La Recherche Scientifique

Université De Bretagne Sud

Mathieu Thevenin

Olivier Bichler

Karim Boudergui

Laurent Letellier

Gwenole Corre

Philippe Coussy

Olivier Thomine

Guillaume Latu

Vesna Simic

Thomas Peyret

Christophe Bobin

Robert French

Gwenolé Corre

Kevin Martin

A method for compressing digital data, including: extrapolating a value of each sample of data to be compressed as a function of a value of at least one preceding sample, to produce an extrapolated sample; differentiating between each extrapolated sample and the corresponding sample of data to be compressed, to produce a differentiated sample; and deleting redundancy between successive differentiated samples produced by the differentiating stage.

Digital data compression

A signal acquisition and processing method includes a step of acquiring signals in at least two acquisition channels (A, A, . . . , A); a step of detecting the acquired signals in at least two detection channels (V, V, . . . , V) associated respectively with the two acquisition channels; a step of distributing the detected signal or signals that result from the detection step in at least one computing tile available in a set of computing tiles; a step of storing the detected signal or signals in the computing tile or tiles; and a step of processing the signals stored in the computing tile or tiles in order to obtain data characteristic of the detected signals. The method can apply to the acquisition and processing of non-deterministic data (nuclear instrumentation, laser instrumentation, radar detection, etc.).

Method for acquiring and processing signals

A device for processing a data stream originating from a device generating matrices of Nl rows by Nc columns of data includes K computation tiles and interconnection means for transferring the data stream between the computation tiles. At least one computation tile includes: one or more control units to provide instructions, n processing units, each processing unit carrying out the instructions received from a control unit on a neighborhood of Vl rows by Vc columns of data, a storage unit to place the data of the stream in the form of neighborhoods of Vl rows by (n+Vc&#x2212;1) columns of data. The storage unit includes a block of shaping memories of dimension Vl&#xd7;Nc and a block of neighborhood registers of dimension Vl&#xd7;(n+Vc&#x2212;1), an input/output unit to convey the data stream between the interconnection means and the storage unit on the one hand, and between the processing units and the interconnection means on the other hand.

Device for the parallel processing of a data stream

Growing community of inventors

Saclay, France

Mathieu Thevenin