The study of noise sources generated by aircraft flyovers in the vicinity of airports requires the deployment on the ground of large arrays of microphones, both in terms of the number of channels (typically several hundreds of microphones) and in terms of their span (several tens of meters). This type of test is in many ways similar to the pass-by noise antenna tests in the automotive, railway or maritime domains: the main difficulties consist in taking into account the trajectory of the source in a beamforming algorithm, in taking into account the frequency dispersions induced by the Doppler effect, and in evaluating the directivity of the source which is seen by the array under several angles of incidence. Deconvolution or identification techniques partially overcome beamforming drawbacks(poor resolution at low frequencies and side lobes), but have been developed for static sources, generally based on the estimation of averaged interspectra of stationary signals. The application to pass-by noise, which is transient by nature, is necessarily delicate and requires many compromises; the most advanced techniques to date are nevertheless based on this principle. An alternative has emerged in recent years, based on an iterative deconvolution processing in the time domain called CleanT. This method, inspired by the Clean method in the frequency domain, has the advantage of operating directly on the resampled signals in the time base of the source, correcting by construction the Doppler effect and allowing access to the information of directivity-at the expense of a significant computational cost. This method, recently proven on automotive and railway applications, is applied in this paper to the aeronautical case of flyover tests, first with simulated data and finally with in situ measurements.