26 January 2009 Influence of noise to PCS particle sizing with Tikhonov regularization inverse algorithm
Author Affiliations +
In photon correlation spectroscopy (PCS) particle sizing techniques, the choice of regularization parameter in the inverse algorithm of Tikhonov Regularization can be classified into prior and posterior strategies. For prior strategies, the choice of regularization parameter is made before calculating the regularized solution, but the posterior strategies, based on some principles, make the choice of regularization parameter matching the error level of original data during the calculation for the regularized solution. The critical issue of regularization method lies in the proper balance and tradeoff between the accuracy and stability of the solutions, in other words, the regularization parameter chosen should match the error lever of autocorrelation function. Using prior and posterior strategies, scattered light signal of 50nm, 100nm and 300 nm particles were simulated and their autocorrelation functions inversed respectively. If noise existed, the stability was weakened while the deviation of peak value increased. For prior strategy, the noise influence to the inverse is obvious when noise factor is 0, 0.01, 0.02, 0.04, 0.05, 0.06, 0.08, 0.1 and 1, but inverse results can be obtained on large noise level even the noise factor is at 1. However, when using posterior strategy, the effect on inverse result was less when error was small, but convergent inverse results cannot be obtained when the noise level is high.
© (2009) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jin Shen, Jin Shen, Yanting Cheng, Yanting Cheng, Qiuyan Han, Qiuyan Han, Wei Liu, Wei Liu, Jingling Song, Jingling Song, } "Influence of noise to PCS particle sizing with Tikhonov regularization inverse algorithm", Proc. SPIE 7156, 2008 International Conference on Optical Instruments and Technology: Optical Systems and Optoelectronic Instruments, 71560V (26 January 2009); doi: 10.1117/12.810810; https://doi.org/10.1117/12.810810

Back to Top