| Research Projects Supported by HKU's High Performance Computing Facilities |
| Researcher: |
| Mr King-fung Ho, Department of Physics |
| Project Title: |
| Deconvolution of Positron Annihilation Coincidence Doppler Broadening Spectra Using an Iterative Projected Newton Method with Non-negativity Constraints |
| Project Description: |
| A generalized least-square method with Tikonov-Miller regularization and non-negativity constraints has been developed for deconvoluting two-dimensional coincidence Doppler broadening spectroscopy (CDBS) spectra. A projected Newton algorithm is employed to solve the generalized least-square problem. The use of suitable deconvolution algorithms is an important issue in improving the quality of CDBS spectra in the low momentum range. A major factor in the success of any deconvolution venture is the quality of the input spectrum itself. Owing to the noise arising from the stochastic nature of the counting process, the more counts in the spectrum, the more true to the convoluted functional shape. A good modern nuclear analog-to-digital converter can digitize into number of channels to 16000, giving CDBS image data of 16000x16000 pixels. |
| Project Duration: |
| 3 years |
| Project Significance: |
| The promising results improved the quality of the experimental data. Not only could this deconvolution method be implemented in mathematics or physics world, it could also be used to enhance the resolution of the electronic equipments with noise arising from the stochastic nature of counting. |
| Results Achieved: |
| Good retrieval of the underlying positron-electron momentum distributions in the low momentum region is demonstrated. The algorithm has been successfully used to deconvolute experimental coincidence Doppler broadening data from various metals. The resolution of the experimental data has been improved by a factor of 3. |
|
|
|
|
| Remarks on the Use of High Performance Computing Cluster: |
| Since a good CDBS image data comprises more than 25 million pixels, HPC Cluster could run the program that ordinary PC could not handle with. In addition to its superb large size of hard disk, parallel computing methods could also substantially speed up the running of the programs and thus save my research a lot of computational time. |
| Email Address: |
| h9614823@hkusua.hku.hk |