Research on robust real-time detection fusion method of multi-source measurement data

Aiming at the interference of complex electromagnetic environment and the unrobustness of traditional method in the real-time processing of multi-source electronic measurement data, this paper proposed a detection fusion method based on median estimate, considering system error registration, cross check and filtering together. The paper put forward a real-time median estimating algorithm of slowly varying sequence based on the sorting of doubly linked list, which was used for estimating the median of error between sensors to get the robust method of system error registration, used together with summation model and error calibration estimation for estimating the median of measurement data difference to get the robust method of filtering. The methods were tested with measurement data to be effective in engineering application.