In this paper, a microscale high-frequency ultrasonic transducer was prepared by combining traditional planar ultrasonic phased-array technology and micro processing technology. The piezoelectric ceramic material PZT was used as the functional material of the transducer. The number of the arrays was 72, the width of each array was 50 μm, the pitch of each array was 70 μm, and the length of each array was 3 mm. The PZT chip was finely ground to a thickness of 130 μm and could reach a frequency of 10 MHz. The experimental platform of micron-scale precision was set up for a beam-forming lateral sound field test and imaging experiment to validate the theoretical analysis. The echo imaging test showed that a mold with a feature size of about 400 μm could be imaged well.

Time-of-flight (ToF)-based 3-D target localization is a very challenging topic because of the pseudo-targets introduced by ToF measurement errors in traditional ToF-based methods. Although the influence of errors in ToF measurement can be reduced by the probability-based ToF method, the accuracy of localization is not very high. This paper proposes a new 3-D target localization method, Iterative Maximum Weighted Likelihood Estimation (IMWLE), that takes into account the spatial distribution of pseudo-targets. In our method, each pseudo-target is initially assigned an equal weight. At each iteration, Maximum Weighted Likelihood Estimation (MWLE) is adopted to fit a Gaussian distribution to all pseudo-target positions and assign new weight factors to them. The weight factors of the pseudo-targets, which are far from the target, are reduced to minimize their influence on localization. Therefore, IMWLE can reduce the influence of pseudo-targets that are far from the target and improve the accuracy of localization. The experiments were carried out in a water tank to test the performance of the IMWLE method. Results revealed that the estimated target area can be narrowed down to the target using IMWLE and a point estimate of target location can also be obtained, which shows that IMWLE has a higher degree of accuracy than the probability-based ToF method.