In the context of the internet of things, dynamic loading of device data information has gained greater importance over time. The proposed algorithm is a Geometric Constraint-based Least Squares Estimation (GC-LSE) positioning algorithm. Accuracy of positioning data is improved by enforcing distance boundary constraints and equality constraints between device reference points and unknown nodes. This strategy makes it possible to monitor operation data of surge arresters and transformers online. Variable fractional delay (VFD) filter is developed on the basis of Farrow structure. Data are processed in two stages, i.e., integer point delay then fraction multiple delay to allow dynamic information loading of IoT devices. Once the positioning errors have been corrected and filtered delays have been processed, the mean information loading time to power IoT devices is only 2.30 seconds with a success rate of 99.19-100.00 percentage, which proves that the optimized loading strategy is effective.
