Various data compression methods are widely used in the processing of information in radio communication systems. The most commonly used methods are quasi-reversible data compression, used, for example, in the processing of radio telemetry information. Such methods are most effective in processing slowly changing signals. For processing broadband signals, such methods are of little use due to the fact that they do not allow to obtain an acceptable compression ratio. Therefore, for processing broadband (fast-changing) signals, it is necessary to use other processing methods that can significantly reduce the redundancy of data transmitted over communication channels. The article is devoted to data processing methods based on the inversion information space. Such methods make it possible to process broadband signals and at the same time achieve sufficient data compression coefficients when transmitting information over communication channels. This is especially important when processing non-stationary broadband signals in the onboard system of spacecraft. A feature of such systems is the need to process information in real time in conditions of a priori uncertainty about the type of statistical properties of the measured processes. The article gives the concept of elementary inversions, inversion keys of relations. The article presents the results of research on data compression methods based on the inversion information space.