Abstract
This paper investigates two different intelligent techniques—the neural network (NN) method and the simulated annealing (SA) algorithm for solving the inverse problem of Rutherford backscattering (RBS) with noisy data. The RBS inverse problem is to determine the sample structure information from measured spectra, which can be defined as either a function approximation or a non-linear optimization problem. Early studies emphasized on numerical methods and empirical fitting. In this work, we have applied intelligent techniques and compared their performance and effectiveness for spectral data analysis by solving the inverse problem. Since each RBS spectrum may contain up to 512 data points, principal component analysis is used to make the feature extraction so as to ease the complexity of constructing the network. The innovative aspects of our work include introducing dimensionality reduction and noise modeling. Experiments on RBS spectra from SiGe thin films on a silicon substrate show that the SA is more accurate but the NN is faster, though both methods produce satisfactory results. Both methods are resilient to 10% Poisson noise in the input. These new findings indicate that in RBS data analysis the NN approach should be preferred when fast processing is required; whereas the SA method becomes the first choice should the analysis accuracy be targeted.
Similar content being viewed by others
Abbreviations
- SA:
-
Simulated annealing
- NN:
-
Neural network
- RBS:
-
Rutherford backscattering
- PCA:
-
Principal component analysis
- WT:
-
Wavelet transform
References
Vizkelethy G (1994) Computer simulation of ion beam methods in analysis of thin films. Nucl Instr Methods B 89:122–130. doi:10.1016/0168-583X(94)95158-6
Kótai E (1994) Computer methods for analysis and simulation of RBS and ERDA spectra. Nucl Instr Methods B 85:588–596. doi:10.1016/0168-583X(94)95888-2
Toussaint UV, Fischer R, Krieger K, Dose V (1999) Depth profile determination with confidence intervals from Rutherford backscattering data. N J Phys 1:1–13. doi:10.1088/1367-2630/1/1/001
Barradas NP, Jeynes C, Webb R, Keissig U, Grotzschel R (1999) Unambiguous automatic evaluation of multiple ion beam analysis data with simulated annealing. Nucl Instr Methods B 149:233–238. doi:10.1016/S0168-583X(98)00731-9
Bohr HG, Frimand K, Jalkanen KJ, Nieminen RM, Suhai S (2001) Neural-network analysis of vibrational spectra of N-acetyl l-alanyl N-methyl amide conformational states. Phys Rev E Stat Nonlin Soft Matter Phys 64:21905–21918. doi:10.1103/PhysRevE.64.021905
Subasi A, Kiymik MK, Akin M, Erogul O (2005) Automatic recognition of vigilance state by using a wavelet-based artificial neural network. Neural Comput Appl 14:45–55. doi:10.1007/s00521-004-0441-0
Cannas B, Fanni A, Manetti S, Montisci A, Piccirilli MC (2004) Neural network-based analog fault diagnosis using testability analysis. Neural Comput Appl 13:288–298. doi:10.1007/s00521-004-0423-2
Ölmez T, Dokur Z (2003) Application of InP neural network to ECG beat classification. Neural Comput Appl 11:144–155. doi:10.1007/s00521-003-0351-6
Übeyli ED (2008) Wavelet/mixture of experts network structure for EEG signals classification. Expert Syst Appl 34:1954–1962. doi:10.1016/j.eswa.2007.02.006
Ceylan R, Özbay Y (2007) Comparison of FCM, PCA and WT technique for classification ECG arrhythmias using artificial neural network. Expert Syst Appl 33:286–295. doi:10.1016/j.eswa.2006.05.014
Hornik K, Stinchcomb M, White H (1989) Multilayer feedforward networks are universal Approximators. Neural Netw 2:359–366. doi:10.1016/0893-6080(89)90020-8
Haykin S (1999) Neural networks: a comprehensive foundation, 2nd edn. Prentice Hall, Englewood Cliffs
Riedmiller M, Braun H (1993) A direct adaptive method for fast backpropagation learning: the RPROP algorithm. Proc IEEE Int Conf Neural Netw 5:586–591
Jolliffe I (2002) Principal component analysis. Springer, New York
Aarts E, Korst J (1989) Simulated annealing and Boltzmann machines: a stochastic approach to combinatorial optimization and neural computing. Wiley, Chichester
Mayer M (2007) SIMNRA user’s guide. Max-Planck Institute of Plasma Physics
Jeynes C, Barradas NP, Webb R (2005) The WiNDF manual, University of Surrey
Barradas NP, Vieira A (2000) Artificial neural network algorithm for analysis of Rutherford backscattering data. Phys Rev E Stat Phys Plasmas Fluids Relat Interdiscip Topics 62:5818–5829. doi:10.1103/PhysRevE.62.5818
Acknowledgment
The author (Michael M. Li) would like to acknowledge Dr. Chris Jeynes (University of Surrey, UK) for sending a trial license to use the WiNDF software package.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Li, M.M., Guo, W., Verma, B. et al. Intelligent methods for solving inverse problems of backscattering spectra with noise: a comparison between neural networks and simulated annealing. Neural Comput & Applic 18, 423–430 (2009). https://doi.org/10.1007/s00521-008-0219-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-008-0219-x