Degree Name

Doctor of Philosophy


School of Electrical, Computer and Telecommunications Engineering


Self-mixing interferometry (SMI) is a promising non-contact sensing technology, which has attracted much research attention in the last few decades. The SMI effect takes place when a small portion of laser is back-scattered from an external target and re-enters the laser diode (LD) internal cavity. A typical sensing system using an SMI configuration consists of an LD as the laser source, a target to reflect the laser, a photo diode (PD) to capture the optical signal. This configuration indicates the SMI-based sensing technology’s merits of minimum part-count scheme, low cost in implementation and ease in optical alignment. Various SMI-based sensing applications have been reported, including the measurement of displacement, velocity, vibration, laser related parameters, thickness, mechanical resonance, imaging, material parameter measurement, near-field microscopy, chaotic radar, acoustic detection, biomedical applications, etc.

Displacement measurement is one of the important applications of SMI-based sensing, which is widely used in defence, security and many industrial application scenes. In this thesis, by reviewing the related literature, the existing problems and measurement restrictions of the conventional SMI-based displacement measurement are recognized. To lift the restrictions, a new algorithm for micro-displacement measurement using modulated injection current is proposed and presented in Chapter 2. In this algorithm, the phase variation of an SMI signal caused by displacement within each modulation period is considered to be time varying. A set of measurement formula is derived for calculating displacement basing on Fast Fourier Transform (FFT) and its inverse operation on the observed SMI signals. Based on the Nyquist sampling theorem, the highest frequency of the reference and SMI signals to be digitized should not exceed half of the sampling frequency. Hence the proposed technique is able to measure the displacement of target vibrating at the frequency up to half of the sampling frequency. The proposed method is able to yield N points of displacement results within each modulation period of injection current, where N is the number of samples of the SMI signal available. In contrast, the typical existing methods can only produce a single measurement point on each period of modulated injection current. Therefore, the proposed method is able to measure the displacement of a target vibrating at a frequency N times higher than the existing methods, which is a significant performance improvement. The effect of windowing on SMI signal and its influence on retrieving displacement information is also investigated in this thesis. Optimal selection for a window function is analyzed and presented for reducing measurement error induced by windowing.



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.