1. Home
  2. Honda R&D Technical Review Vol.24 N...
  3. Construction of Knock Detection Log...

Technical Review e-Book: Summary

Construction of Knock Detection Logic by Pattern Recognition Using Short-time Fourier Transform

Article of Honda R&D Technical Review Vol.24 No.1


The project discussed in this paper focused on the frequency attenuation characteristic of knock and mechanical noise in order to develop a more accurate method of distinguishing between the two phenomena. Short-time Fourier transform (STFT), which makes it possible to identify changes in frequency over time, was studied as a method of analyzing this characteristic. The duration of knocking is 80-90 deg in terms of crank angle. The frequencies of the multiple vibration modes that knock generates decline as time elapses, with attenuation greater in the high-frequency vibration modes, until finally only the 1, 0th mode remains. Mechanical noise occurs over a wide frequency band, and its attenuation is rapid. Based on these results, the concept of time was introduced in order to help enable the attenuation of the frequencies to be identified, and a knock detection logic that judges knock on the basis of pattern recognition was developed. This logic is able to distinguish knock and mechanical noise with a high degree of accuracy even when the timing and frequency of the two phenomena match. As a result, it has been possible to increase robustness in relation to mechanical noise in comparison to conventional knock detection systems that judge knock based on the intensity of the frequency at which it occurs, helping to ensure stable functioning of the system even when mechanical noise occurs. This has made it possible to set the threshold value for judgment of knock lower than is the case in conventional systems, contributing to the reduction of knock noise.


(1) Zadnik, M., Vincent, F., Vingerhoeds, R., Galtier, F.: SI Engine Knock Detection Method Robust to Resonance Frequency Changes, SAE Paper, 2007-24-0054 (2007)
(2) Lonari, Y., Polonowski, C., Naber, J., Chen, B.: Stochastic Knock Detection Model for Spark Ignited Engines, SAE Paper, 2011-01-1421 (2011)
(3) Ham, Y. Y., Chun, K. M., Lee, J. H., Chang, K. S.: Spark-ignition engine knock control and threshold value determination, SAE Paper, 960496 (1996)
(4) Borg, J. M., Cheok, K. C., Oho, S. T., Saikalis, G.: Knock Signal Analysis Using the Discrete Wavelet Transform, SAE Paper, 2006-01-0226 (2006)
(5) Zhang, Z., Tomita, E.: Diagnostic of Knocking by Wavelet Transform Method Utilizing Real Signal as Mother Wavelet, SAE Paper, 2001-01-3546 (2001)
(6) Zhang, Z., Tomita, E.: A New Diagnostic Method of Knocking in a Spark-ignition Engine using the Wavelet Transform, SAE Paper, 2000-01-1801 (2000)
(7) Xiaofeng, G., Stone, R., Hudson, C., Bradbury, I.: The detection and quantification of knock in spark ignition engines, SAE Paper, 932759 (1993)
(8) Kearney, M. J.: Knock Signal Conditioning Using the Discrete Fourier Transform and Variable Detection Window Length, SAE Paper, 2007-01-1509 (2007)
(9) Wu, G.: A Real Time Statistical Method for Engine Knock Detection, SAE Paper, 2007-01-1507 (2007)
(10) Samimy, B., Rizzoni, G.: Engine knock analysis and detection using time-frequency analysis, SAE Paper, 960618 (1996)
(11) Draper, C. S.: The physical effects of detonation in a closed cylindrical chamber, N.A.C.A., p. 493 (1934)

Author (organization or company)

Kenji AKIMOTO(Automobile R&D Center)、Hirotaka KOMATSU(Automobile R&D Center)、Atsushi KURAUCHI(Automobile R&D Center)

We would like to get your opinion on this research paper. (This is only applicable to registered members.)

The readers of this research paper have also selected these research papers.

New In-Line 4-Cylinder Gasoline Direct Injection Turbocharged Downsizing Engine
Article of Honda R&D Technical Review Vol.28 No.1
Optimization of Combustion Chamber Shape in High-efficiency Stoichiometric Combustion Gasoline Engine
Article of Honda R&D Technical Review Vol.27 No.2
Technology for Realization of Gasoline Engine Brake Thermal Efficiency of 45%
Article of Honda R&D Technical Review Vol.27 No.2