The data rate is an important parameter in the tracking management of the electronic reconnaissance system. The size of the data rate determines the tracking accuracy, and also affects the system resource consumption. In order to balance the tracking accuracy and resource consumption, an adaptive data rate algorithm is needed. However, the current adaptive data rate algorithm of the active radar is not suitable for the electronic reconnaissance system because of the discontinuity of the radiation source in the time domain. In order to address these issues, the radiation source antenna scanning period as a constraint is introduced, and predictive covariance threshold algorithm is improved to be applied to electronic reconnaissance system. In the improved algorithm, the integer multiple of the scanning period of the radiation source antenna is used as the sampling interval, and the prediction covariance is obtained by the bearing-only tracking algorithm. When the prediction covariance is greater than the set threshold, the sampling is performed. The simulation results show that the method can adaptively get the sampling period according to the set tracking accuracy, save system resources.