Research On The Algorithm Of Inteferogram Spectrum Reconstruction

Spectrum imaging technology is a powerful tool in exploring both the outer space and ourown planet earth, it can acquire two dimensional space information and one dimensionalspectrum information of its target simultaneously, producing massive amount of spectrum datawhich requires huge quantities of computation. In this paper we start from two aspects, one ishow to reconstruct the spectrum of the target with high quality from the hard-woninterferogram, another is how to effectively transport the related data in the process, and wewill study the two problems in both theory and practice.In this paper, to improve the theory of the conventional inteferogram spectrumreconstruction algorithm, we first discussed basic spectral estimation concepts and algorithms,then pointed out that the conventional spectrum reconstruction algorithm is virtually the samewith Blackman-Tukey (BT) algorithm in spectral estimation. Then we use two differentmethod (Auto-Regressive parametric model and eigenvector spectral estimation method) toreconstruct the power spectrum of light signal from interferogram, and compared the resultswith conventional method for spectrum reconstruction. From the comparison we concludedthat the modern spectral estimation methods can well be adopted in Fourier Transform (FT)spectrum reconstruction and would perform excellently, especially when to get higher spectrumresolution from a short series of recorded data. Later we discussed the disadvantages andrestrictions of the two methods used in the simulation. From these results and discussions, wesuggested to implement various new spectral estimation method in FT spectrum reconstructionto get better performance.To address the massive amount of data transported in the reconstruction and the complexityof the computation, distributed system is adopted in order to improve the processing speedwhen dealing with large amount of reconstruction. After a brief introduction of distributedsystem’s concepts and features, we build a simple distributed inteferogram reconstructionsystem, later experiment validated the reconstruction results and showed that the system iseffective; on this system we then discussed several problems about the performance andsecurity of the distributed system.