In an effort to accomplish the real-time acquisition of the laser ranging results of space debris during the daylight and enhance the observation success rate, this paper establishes a joint distribution model of noise and echo signals grounded on the intensity distribution law of the daylight background noise. Through an in-depth analysis of the measurement characteristics of single-photon detectors, a real-time recognition algorithm based on the disparity in statistical distribution is put forward. This algorithm partitions the observation data into intervals of equal length. It then employs the goodness-of-fit test of the geometric distribution to ascertain the data distribution law. Subsequently, it locates the interval in which the echo signal resides by analyzing the contribution degree of the chi-square statistic. The experimental outcomes indicate that under the circumstances of a laser frequency of 400 Hz and a background noise photon rate of 2 × 10⁷ photons per second, the algorithm is capable of achieving real-time recognition of the echo interval for an intensity of 0.09 echo photons per single pulse within 1 s. This breakthrough resolves the critical challenge of daylight echo discrimination in high-noise environments. This method overcomes the constraints of the traditional signal intensity threshold and offers a novel technical approach for the tracking and precise orbit determination of space debris in a low signal-to-noise ratio environment.

The high-precision motion state analysis of space targets has important scientific value and application potential in the fields of geodynamics, geodesy, space collision warning and avoidance, and the capture, recovery, and removal of space debris. With the increasing repetition rate of satellite laser ranging systems, the inversion and analysis of space target motion state based on high-precision and high-repetition-rate satellite laser ranging data has become a hot spot in current research. How to filter out noise and retain valid information from the high-repetition-rate, high-precision laser ranging data has become a challenge. The traditional polynomial fitting method has problems with low data accuracy and erroneous deletion of valid data when processing laser ranging data with complex patterns. To address this problem, this paper innovatively designs a new type of spline function and accordingly proposes a laser ranging data preprocessing method that can automatically adapt to the trend of data variation. The method is validated using various characteristic observation data provided by Changchun station (7237), and the results show that the novel spline method is superior to the traditional method in maintaining the integrity of data patterns and significantly improves the data accuracy. After two iterations of denoising, the RMS of the novel spline method is reduced to one-fourth of that before denoising and as low as one-eighteenth of that of the traditional method, and the accuracy is further improved with the increase of the number of iterations. This study provides a practical and reliable novel solution for the preprocessing of laser ranging data with complex patterns.