Incoherent laser off-target quantity detection relies on quadrature phase demodulation, and the application of a Kalman low-pass filter can significantly improve the phase discrimination performance of digital quadrature demodulation phasemeters. This paper proposes a solution based on Sage-Husa adaptive filtering, which uses adaptive factors to adjust the state prediction covariance array to effectively reduce the model errors and improve the filtering accuracy, to address the problem that the accuracy of the Kalman low-pass filter decreases when the noise statistics information is unknown. The adaptive Kalman filtering method may substantially improve the phase identification performance of a digital phase-locked demodulator and reduce the decoding error of off-target amount under low signal-to-noise ratio, according to Matlab simulation studies.