Discussion by Walter Liggett
The authors summarize what has been inferred about the physics of binary stars through a variety of statistical signal processing techniques. They affirm certain aspects of the most widely accepted model for accreting binary stars, conclude that clumps of matter must accrete at more than 400 per second, and point to aspects of the observed x-ray intensities that seem unexplained. As I read the paper, I wished for a more integrated summary of the signal processing results that would be more likely to suggest modifications of the accepted physical model. If such a summary were possible, it would involve both physical insight into what to look for and statistical insight into the performance of signal processing techniques. The purpose of this discussion is to consider this challenge.
KeywordsBeat Frequency Binary Star Signal Processing Technique Gaussian Random Process Quasiperiodic Oscillation
Unable to display preview. Download preview PDF.
- Donoho, D. L. On minimum entropy deconvolution. In D. F. Findley, editor, Applied Time Series Analysis II, page 565, Academic Press. New York, 1981.Google Scholar
- Donoho, D. L. Surprising nonlinear phenomena in recovery of signals, images, fields. Abstract in The IMS Bulletin, 20:161, 1991.Google Scholar
- Lii, K. S. and Rosenblatt, M. Non-Gaussian linear processes, phase and deconvolution. In E. J. Wegman and J. G. Smith, editors, Statistical Signal Processing, p. 51, Marcel Dekker, New York, 1984.Google Scholar