A pyroshock signal characterization method based on shock-waveform dictionary

In space engineering, devices (including products) need to work in pyroshock environments during their complete service life. The verification of the full compatibility of vulnerable devices with the pyroshock environment is necessary. Shock response spectrum (SRS) as the current shock specification tool in space engineering has many limitations. However, an SRS curve may be associated with different pyroshock signals that can cause different effects. Therefore, there is a great need to develop characterization methods for pyroshock signals. In this paper, the shock-feature relevant waveform is used as the basis function to approximate pyroshock signals. A parametric dictionary design method is proposed using a random search strategy, in which the designed dictionaries are close to equiangular tight frames. The pyroshock approximation problem is turned into a multi-objective optimization problem and solved by a modified convolutional sparse coding algorithm. Reconstructed signals approach pyroshock not only in the time domain but also in the SRS domain. Parameters of shock-waveforms can be used for pyroshock characterization, simplification, comparison and tracking. The superiority of the proposed methods is validated by the analyses of a pyroshock signal measured in-flight. •A new method to characterize pyroshock signals is developed using physics-based shock waveform dictionary.•It can simultaneously reconstruct signals in both time domain and shock response spectrum (SRS) domain.•It is an efficient and reliable method to characterize pyroshock signals and can provide a new way for shock testing specification.