© The Institution of Engineering and Technology
A tri-layer bus system-on-chip (SoC) and a butterfly-path accelerator are used to enhance system-level performance in a sequential minimal optimisation learning core. The tri-layer bus architecture is used to obtain an adequate transfer rate. The butterfly-path accelerator also uses symmetrical access to resolve bottlenecks during linear prediction cepstral coefficients extraction. This novel design increases speed and flexibility without substantially increasing area. For implementation in chip manufacturing, the SoC is synthesised, placed and routed using the TSMC 90 nm technology library. The die size is 2.09 mm × 2.09 mm, and the power consumption is 8.9 mW. Compared with the non-butterfly-path design, the simulation results show that the proposed architecture provides a 2.4-fold speed increase. In addition, clock down-sampling and voltage scaling reduce the power consumed by the proposed chip by a factor of 8.5. The experimental results confirm the improved speed and power that are provided by the proposed architecture and methods.
References
-
-
1)
-
19. Peng, S.-Y., Minch, B.A., Hasler, P.: ‘Analog VLSI implementation of support vector machine learning and classification’. Proc. IEEE Int. Symp. Circuits Systems, Seattle, WA, May 2008, pp. 860–863.
-
2)
-
6. Wang, N., Bayoumi, M.A.: ‘System-on-chip communication architecture: dynamic parallel fraction control bus design and test methodologies’, IET Comput. Digit. Tech., 2007, 1, (1), pp. 1–8 (doi: 10.1049/iet-cdt:20060080).
-
3)
-
21. Lin, P.-C., Wang, J.-C., Wang, J.F., et al: ‘Design and portable device implementation of feature-basedpartial matching algorithms forpersonal spoken sentence retrieval’, IET Signal Process., 2007, 1, (3), pp. 139–149 (doi: 10.1049/iet-spr:20060326).
-
4)
-
9. Pomante, L.: ‘HW/SW co-design of dedicated heterogeneous parallel systems: an extended design space exploration approach’, IET Comput. Digit. Tech., 2013, 7, (6), pp. 246–254 (doi: 10.1049/iet-cdt.2013.0026).
-
5)
-
22. Wang, J.-F., Wang, J.-C., Mo, M.-H., et al: ‘The design of a speech interactivity embedded module and its applications for mobile consumer devices’, IEEE Trans. Consum. Electron., 2008, 54, (2), pp. 870–876 (doi: 10.1109/TCE.2008.4560173).
-
6)
-
3. Kang, K., Shibata, T.: ‘An on-chip-trainable Gaussian-kernel analog support vector machine’, IEEE Trans. Circuits Syst. I, Regul. Paps., 2010, 57, (7), pp. 1513–1524 (doi: 10.1109/TCSI.2009.2034234).
-
7)
-
5. Wang, J.-F., Peng, J.-S., Wang, J.-C., et al: ‘Hardware/software co-design for fast-trainable speaker identification system based on SMO’. Proc. IEEE Int. Conf. Systems, Man, and Cybernetics, Anchorage, AK, October 2011, pp. 1621–1625.
-
8)
-
23. Lee, C.-H., Hsu, S.-B., Shih, J.-L., et al: ‘Continuous birdsong recognition using Gaussian mixture modeling of image shape features’, IEEE Trans. Multimed., 2013, 15, (2), pp. 454–464 (doi: 10.1109/TMM.2012.2229969).
-
9)
-
27. Vitkovski, A., Jantsch, A., Lauter, R., et al: ‘Low-power and error protection coding for network-on-chip traffic’, IET Comput. Digit. Tech., 2008, 2, (6), pp. 483–492 (doi: 10.1049/iet-cdt:20050060).
-
10)
-
7. Srinivasan, S., Li, L., Ruggiero, M., et al: ‘Exploring architectural solutions for energy optimisations in bus-based system-on-chip’, IET Comput. Digit. Tech., 2008, 2, (5), pp. 347–354 (doi: 10.1049/iet-cdt:20070063).
-
11)
-
1. Cortes, C., Vapnik, V.: ‘Support vector networks’, Mach. Learn., 1995, 20, (3), pp. 273–297.
-
12)
-
11. Cao, K.K., Shen, H.B., Chen, H.F.: ‘A parallel and scalable digital architecture for training support vector machines’, IEEE Trans. Very Large Scale Integr. Syst., 2010, 11, (8), pp. 620–628.
-
13)
-
15. Wu, G.-D., Kuo, K.-T.: ‘System-on-chip architecture for speech recognition’, J. Inf. Sci. Eng., 2010, 26, (3), pp. 1073–1089.
-
14)
-
2. Campbell, W.M., Campbell, J.P., Gleason, T.P., et al: ‘Speaker verification using support vector machines and high-level features’, IEEE Trans. Speech, Audio Lang. Process., 2007, 15, (7), pp. 2085–2093 (doi: 10.1109/TASL.2007.902874).
-
15)
-
3. Takahashi, N., Guo, J., Nishi, T.: ‘Global convergence of SMO algorithm for support vector regression’, IEEE Trans. Neural Netw., 2008, 19, (6), pp. 971–982 (doi: 10.1109/TNN.2007.915116).
-
16)
-
4. Platt, J.C., Scholkopf, B., Burges, C., et al: ‘Fast training of support vector machines using sequential minimal optimization’, in (eds.): Schölkopf, B., Burges, C., Smola, A.: ‘Advances in Kernel methods—support vector learning’ (MIT Press, Cambridge, MA, USA, 1999).
-
17)
-
29. Alhussien, A., Wang, C., Bagherzadeh, N.: ‘Design and evaluation of a high throughput robust router for network-on-chip’, IET Comput. Digit. Tech., 2012, 6, (3), pp. 173–179 (doi: 10.1049/iet-cdt.2011.0082).
-
18)
-
16. Genov, R., Cauwenberghs, G.: ‘Kerneltron: Support vector machine in silicon’, IEEE Trans. Neural Netw., 2003, 14, (5), pp. 1426–1434 (doi: 10.1109/TNN.2003.816345).
-
19)
-
S. Chakrabartty ,
G. Cauwenberghs
.
A sub-microwatt analog VLSI trainable pattern classifier.
IEEE J. Solid-State Circuits
,
5
-
20)
-
10. Cao, L.J., Keerthi, S.S., Ong, C.-J., et al: ‘Parallel sequential minimal optimization for the training of support vector machines’, IEEE Trans. Neural Netw., 2006, 17, (4), pp. 1039–1049 (doi: 10.1109/TNN.2006.875989).
-
21)
-
26. Kondo, H., Nakajima, M., Masui, N., et al: ‘Design and implementation of a configurable heterogeneous multicore SoC with nine CPUs and two matrix processors’, IEEE J. Solid-State Circuits, 2008, 43, (4), pp. 892–901 (doi: 10.1109/JSSC.2008.917528).
-
22)
-
8. Hansson, A., Wiggers, M., Moonen, A., et al: ‘Enabling application-level performance guarantees in network-based systems on chip by applying dataflow analysis’, IET Comput. Digit. Tech., 2009, 3, (5), pp. 398–412 (doi: 10.1049/iet-cdt.2008.0093).
-
23)
-
25. Kuhn, H.W., Tucker, A.W.: ‘Nonlinear programming’. Proc. Int. Conf. Berkeley Symp. Mathematical Statistics and Probability, Berkeley, California, August 1951, pp. 481–492.
-
24)
-
17. Kucher, P., Chakrabartty, S.: ‘An energy-scalable margin propagation-based analog VLSI support vector machine’. Proc. IEEE Int. Symp. Circuits Systems, New Orleans, USA, May 2007, pp. 1289–1292.
-
25)
-
28. Zhang, L., Han, Y., Xu, Q., et al: ‘On topology reconfiguration for defect-tolerant NoC-based homogeneous manycore systems’, IEEE Trans. Very Large Scale Integr. Syst., 2009, 17, (9), pp. 1173–1186 (doi: 10.1109/TVLSI.2008.2002108).
-
26)
-
12. Kuan, T.-W., Wang, J.-F., Wang, J.-C., et al: ‘VLSI design of an SVM learning core on sequential minimal optimization algorithm’, IEEE Trans. Very Large Scale Integr. Syst., 2012, 20, (4), pp. 673–683 (doi: 10.1109/TVLSI.2011.2107533).
-
27)
-
24. Durbin, J.: ‘Efficient estimation of parameters in moving average Models’, Biometrika, 1959, 46, (3–4), pp. 306–316 (doi: 10.1093/biomet/46.3-4.306).
-
28)
-
13. Peng, C.-H., Chen, B.-W., Kuan, T.-W., et al: ‘REC-STA: Reconfigurable and efficient chip design with SMO-based training accelerator’, IEEE Trans. Very Large Scale Integr. Syst., 2013, 22, (8), pp. 1791–1802 (doi: 10.1109/TVLSI.2013.2278706).
-
29)
-
14. Kim, S.-N., Hwang, I.-C., Kim, Y.-W., et al: ‘A VLSI chip for isolated speech recognition system’, IEEE Trans. Consum. Electron., 1996, 42, (3), pp. 458–467 (doi: 10.1109/30.536144).
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-cdt.2013.0153
Related content
content/journals/10.1049/iet-cdt.2013.0153
pub_keyword,iet_inspecKeyword,pub_concept
6
6