Improved adaptive forward-backward matching pursuit algorithm to compressed sensing signal recovery
- 83 Downloads
As a novel two-stage greedy approximation algorithm, Forward-Backward Pursuit (FBP) algorithm attracts wide attention because of its high reconstruction accuracy and no need for sparsity as a priori information. However, the FBP algorithm has to spend much more time to get a higher accuracy. In view of this, an Improved Adaptive Forward-Backward Matching Pursuit (IAFBP) algorithm is proposed in this paper. In the forward stage, the IAFBP algorithm uses an adaptive threshold to select the appropriate number of atoms into support set, so that the number of selected atoms is more random. In the backward stage, the projection coefficient of the atoms is taken as the basis of rejection, and the deletion threshold is introduced to reject the atoms adaptively, so that more right atoms are retained in each iteration and the reconstruction speed can be accelerated. At the same time, it overcomes the excessive backtracking phenomenon existing in the adaptive process and improves the accuracy of the algorithm. The simulation results of one-dimensional sparse signals and two-dimensional images show that the IAFBP algorithm has more advantages than the FBP algorithm in reconstruction performance and computational time.
KeywordsCompressed sensing Matching pursuit Forward-backward search Adaptive threshold Signal reconstruction
The work was supported by the National Natural Science Foundation of China (No. 51575472), by the Natural Science Foundation of Hebei Province of China (No.E2019203448), the scientific research program of Hebei Education Department (No. ZD2015049) and the scientific research program for Talents Returning from Overseas of Hebei Province (No. C2015005020).
Compliance with ethical standards
Conflict of interest
The authors declare no conflict of interest.
- 5.Guo H, Han S, Hao F et al (2017) SOSP: a stepwise optimal sparsity pursuit algorithm for practical compressed sensing. Multimed Tools Appl 77(7):1–24Google Scholar
- 8.Karahanoğlu NB, Erdoğan H (2013) Optimal forward-backward pursuit for the sparse signal recovery problem. Proceeding of 21th signal processing and communications applications conference (SIU), 21–24Google Scholar
- 11.Meng X, Zhao R, Cen Y (2016) A modified regularized adaptive matching pursuit algorithm for compressed sampling signal reconstruction. Signal Process 32(2):186–192Google Scholar
- 12.Meng Z, Pan Z, Li J et al (2019) Improved backtracking regularized adaptive matching pursuit algorithm and its application. Chinese High Technol Lett 29(02):110–118Google Scholar
- 14.Peng Y, He Y, Lin B (2012) Noise signal recovery algorithm based on singular value decomposition incompressed sensing. Chin J Sci Instrum 33(12):2655–2660Google Scholar
- 18.Wang L, Zhou L, Ji H et al (2014) A new matching pursuit algorithm for signal classification. J Electron Inform Technol 36(6):1299–1306Google Scholar
- 19.Wang F, Sun G, Zhang J (2016) Acceleration forward-backward pursuit algorithm based on compressed sensing. J Electron Inform Technol 38(10):2538–2545Google Scholar
- 20.Yan C, Li L, Zhang C, et al (2019) Cross-modality bridging and knowledge transferring for image understanding. IEEE Trans Multimedia, to be publishedGoogle Scholar
- 21.Yao S, Wang T, Shen W (2015) Research of incoherence rotated chaotic measurement matrix in compressed sensing. Multimed Tools Appl 76(17):1–19Google Scholar