A Low-Power Hardware-Friendly Binary Decision Tree Classifier for Gas Identification
Abstract
:1. Introduction
2. Binary Decision Tree Classifiers
2.1. Overview
2.2. Hardware Considerations
3. Proposed DT Classifier Implementation
3.1. Threshold Logic Unit
3.2. Programmable Logic Unit
4. Results
4.1. Experiment Setup
4.2. Classification Performance
4.3. VLSI Implementation
5. Conclusions
PC No. | DT Type | Accuracy | Leaves | Node | Mul | Add/Sub | Memory (No. of Coefficients) |
---|---|---|---|---|---|---|---|
2 | axis-parallel | 88.33% | 13 | 12 | 0 | 12 | 12 |
2 | oblique | 89.24% | 9 | 8 | 16 | 16 | 24 |
3 | axis-parallel | 92.73% | 10 | 9 | 0 | 9 | 9 |
3 | oblique | 90.30% | 5 | 4 | 12 | 12 | 16 |
4 | axis-parallel | 99.55% | 7 | 6 | 0 | 6 | 6 |
4 | oblique | 94.55% | 3 | 2 | 8 | 8 | 10 |
no PCA | axis-parallel | 91.36% | 10 | 9 | 0 | 9 | 9 |
no PCA | oblique | 92.58% | 6 | 5 | 60 | 60 | 65 |
DT Type | Process | clk (ns) | Delay (ns) | Power (mW) | Area (103 μm2) | |
---|---|---|---|---|---|---|
Reference [6] | oblique | 0.18 μm | 5 | 1.61 | 25 | 100 |
This work | axis-parallel | 0.18 μm | 5 | 0.9 | 3.1 | 8.9 |
Classification Method | GMM [7] | Committee Machine [8] | Oblique DT [6] | This Work (Axis-Parallel DT) |
---|---|---|---|---|
Sensor type | 2×2 SnO2 array | 2×2 SnO2 array | 4×4 SnO2 array | 4 × 4 SnO2 array |
Target gas species | CH4,CO,H2, CO–CH4, CO–H2 | CH4,CO,H2, CO–CH4, CO–H2 | Ethanol, CO, H2 | Ethanol, CO, H2 |
Detection rate | 92% | 94% | 92.58% | 91.36% |
FPGA Implementation | ||||
No. of Slice FF | N/A | 12146 | 1176 | 233 |
No. of 4-LUT | N/A | 20115 | 1269 | 160 |
ASIC Implementation | ||||
process | 0.25 μm | N/A | 0.18 μm | 0.18 μm |
Area | 1.69 mm2 | N/A | 0.1 mm2 | 0.028 mm2 |
Power | N/A | 500 mW | 25 mW | 3.1 mW |
Acknowledgments
References
- Gutierrez-Osuna, R. Pattern analysis for machine olfaction: A review. IEEE Sens. J. 2002, 2, 189–202. [Google Scholar]
- Belhouari, S.B.; Bermak, A. Gas Identification using density models. Pattern Recognit. Lett. 2005, 26, 699–706. [Google Scholar]
- Beriman, L.; Freidman, J.H.; Olshen, R.A. Classification and Regression Trees.; Wadsworth International Group: Belmont, CA, USA, 1984; pp. 203–215. [Google Scholar]
- Struharik, R.J.R.; Novak, L.A. Evolving decision trees in hardware. J. Circuits Syst. Comput. 2009, 18, 1033–1060. [Google Scholar]
- Bermak, A.; Martinez, D. A reconfigurable hardware implementation of tree classifier based on a custom chip and CPLD for gas sensors applications. Proceedings of the IEEE TENCON, Chiang Mai, Thailand, October 2004; 1, pp. 32–35.
- Bermak, A.; Martinez, D. A Compact 3D VLSI Classifier using Threshold Network Ensembles. IEEE Trans. Neural Networks 2003, 14, 1097–1109. [Google Scholar]
- Shi, M.; Bermak, A. An Efficient Digital VLSI Implementation of Gaussian Mixture Models-based classifier. IEEE Trans. Very Large Scale Integr. Syst. 2006, 14, 962–974. [Google Scholar]
- Shi, M.; Bermak, A.; Belhouari, S.B.; Chan, P.C.H. Gas Identification based on a committee machine for Microelectronics gas sensors. IEEE Trans. Instrum. Meas. 2006, 55, 1786–1793. [Google Scholar]
- Quinlan, J.R. Induction of Decision Trees. Mach. Learn. 1986, 1, 81–106. [Google Scholar]
- Quinlan, J.R. C4.5: Programs for Machine Learning; Morgan Kaufmann Publishers: San Francisco, CA, USA, 1993. [Google Scholar]
- Hall, M.; Frank, E.; Holmes, G.; Pfahringer, B.; Reutemann, P.; Witten, I.H. The WEKA Data Mining Software: An Update. SIGKDD Explor. 2009, 11, 10–18. [Google Scholar]
- Murthy, S.K.; Kasif, S.; Azlzberg, S. Asystem for induction of oblique decision trees. J. Artif. Intell. Res. 1994, 2, 1–33. [Google Scholar]
- Lopez-estrada, S.; Cumplido, R. Decision tree based FPGA architecture for texture sea state classification. Proceedings of IEEE Conference on Reconfigurable Computing and FPGAs, San Luis Potosi, Mexico, September 2006; pp. 1–7.
- Struharik, R.J.R.; Novak, L.A. Intellectual property core implementation of decision trees. IET Comput. Digit. Tech. 2009, 3, 259–269. [Google Scholar]
- Sethi, I.K. Entropy net: from decision trees to neural nets. Proc. IEEE 1990, 78, 1605–1613. [Google Scholar]
- Alioto, M.; Palumbo, G. Model and Design of Bipolar and MOS Current-Mode Logic: CML, ECL and SCL Digital Circuits; Kluwer Academic Publishers: Norwell, MA, USA, 2005. [Google Scholar]
©2011 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).
Share and Cite
Li, Q.; Bermak, A. A Low-Power Hardware-Friendly Binary Decision Tree Classifier for Gas Identification. J. Low Power Electron. Appl. 2011, 1, 45-58. https://doi.org/10.3390/jlpea1010045
Li Q, Bermak A. A Low-Power Hardware-Friendly Binary Decision Tree Classifier for Gas Identification. Journal of Low Power Electronics and Applications. 2011; 1(1):45-58. https://doi.org/10.3390/jlpea1010045
Chicago/Turabian StyleLi, Qingzheng, and Amine Bermak. 2011. "A Low-Power Hardware-Friendly Binary Decision Tree Classifier for Gas Identification" Journal of Low Power Electronics and Applications 1, no. 1: 45-58. https://doi.org/10.3390/jlpea1010045