Some of the synchronization problems encountered when recording data with different systems are presented, with a focus on the simultaneous measurement of video and digital data, followed by off-line synchronization of the information streams. Research in the context of the Visual Lab project [1] has led to the development of the Pseudo Random Noise (PRN) code approach, which combines excellent synchronization accuracy with a relatively simple set-up and a low synchronization information data-rate.

Every post-processing synchronization method relies on the overlap of information between different simultaneous measurements. The main difference between the methods is the amount of data used to record the overlap. At the two extremes, one can record an (absolute) time source for each measurement, or simply rely on the overlap of information channels in different measurements. An example of the latter is the similarity of eye-blinks measured with a camera system (e.g. eye-tracker) and eye-blinks detected through the measurement of the vertical EOG. All possible approaches have been implemented in the NLR 'Human factors Evaluations, data Analysis and Reduction Techniques' (HEART [2]) data analysis program, to determine the best option.

This program concluded that the required overlap of information could best be provided by a special binary code, called PRN [3]. The PRN code possesses some interesting characteristics, which make it suitable for recording on diverse media (digitally, or within audio and video streams), without the requirement for specialised and/or complex converters. It also allows the retrieval of time-differences with a high resolution (in the order of 5 ms for a PRN chip-rate of 2 Hz and a measurement length exceeding 100 chips, i.e. 100/2 = 50 seconds). The PRN code tested is also used by the Global Positioning System (GPS, a satellite position determination system), which determines time-differences between a transmitter and a receiver with extreme accuracy (in the order of nanoseconds).

The development of the PRN synchronization technique is not finalised. Several enhancements and refinements to the technique are still possible. However, the current implementation already demonstrates extremely high precision without many system constraints, and therefore seems to be a practical way to allow multiple recordings with diverse media (audio, video and digital information).

References

1. 'Visual Lab project', Apr 1999 - Dec 2001. http://www.noldus.com/
2. Hoogeboom, P.J. (2002). Human factors evaluations, data analysis and reduction techniques. Proc. 2nd IEEE Benelux Signal Processing Symposium (SPS-2000), Hilvarenbeek, The Netherlands, March 23-24.
3. Gold, R. (1968). Maximal recursive sequences with 3-valued recursive cross-correlation functions. IEEE transactions on Information Theory, vol. IT-14, 154-156.

Paper presented at Measuring Behavior 2002 , 4^th International Conference on Methods and Techniques in Behavioral Research, 27-30 August 2002, Amsterdam, The Netherlands

© 2002 Noldus Information Technology bv