Abstract: Ultra-short baseline is the necessary device to achieve the accurate positioning of towing equipment, and it is also the important equipment to do the deep-sea resource research in the international seabed area at the present. Because of the influences of acoustic signal propagation loss,multipath effect and other equipment noise, underwater positioning data has many outliers, which has seriously affected the credibility and the use of positioning data.To solve this problem, we propose and implement a correction algorithm based on correlation analysis to reject outliers and random regression which fill and correct positioning data. We use positioning data measured from the international seabed area to validate the reliability of the algorithm.At the same time, we use simulation comparison method for small scale and short time positioning data forecast, and therefore a good result has been achieved. The experimental results show that the algorithm is fast and effective. It is expected to be applied to the international seabed area survey, and can provides direct service to the international seabed resources investigation for accurate positioning equipment.

Key words: ultra short baseline, underwater positioning, abnormal points, correlation analysis, fill correct