The Pulse-Pre-Pump Brillouin Optical Time Domain Analysis (PPP-BOTDA) based distributed fiber optic sensing (DFOS) technique has a unique advantage of achieving full-distributed structural strain with cm-order spatial resolution, which is ideal for fine stress monitoring of the pile foundation. The high spatial resolution helped to detect local changes and defects while inevitably led to noisy data meanwhile, especially a certain part of anomalous sensing data. Therefore, a robust algorithm of random sampling consistency (RANSAC) was employed to extract and process the DFOS data. In a field static load test, the variation of strain along a precast pile was measured with PPP-BOTDA interrogator. The axial force and frictional resistance were calculated based on the RANSAC processed strain data. The pile-soil interface properties were estimated as well. It shows the feasibility and good performance of PPP-BOTDA in the measurement of precast piles stress test. The RANSAC proves to be an effective method for processing DFOS data with a certain number of outliers. This work presents a valuable reference for DFOS based pile stress test and data analysis which will promote its application in other related structural health monitoring purpose.

No abstract