This paper proposes a novel method for the simultaneous estimation of multiple interference phases from a single recording of the interference field in a multi-wave digital holographic interferometry set-up. The method involves the separation of signal components from the recorded interference field based on windowed Fourier transform based filtering and amplitude discrimination criteria. The proposed method possesses the advantages offered by windowed Fourier transform based filtering such as noise robustness and a high quality reconstruction of the signal components. The performance of the method is validated with numerical and experimental examples.