The noise emissions of a full-scale nose landing gear, measured in a wind tunnel and obtained from computational simulations, are compared with those of three regional aircraft types recorded in flyover measurements. The results from these three approaches are also compared with the predictions of two airframe noise models (Fink and Guo). The geometries of the nose landing gears in all cases were similar. Microphone arrays and acoustic imaging algorithms were employed to estimate the sound emissions of the nose landing gears. A good agreement was found between the overall trends of the frequency spectra in all cases. Moreover, the expected 6 th power law with the flow velocity was confirmed. On the other hand, strong tonal peaks (at around 2200 Hz) were only found for the flyover tests and computational simulations and are not present in typical noise prediction models. As the frequencies of the tones did not depend on the flow velocity, they are likely to be caused by cavities found in structural components of the nose landing gear. Removing these tones would cause overall noise reductions up to 2 dB in the frequency range examined. The noise emissions in the side direction did not present tonal peaks. The acoustic source maps showed that the dominant noise sources were located in the middle of the wheel axle, followed by the main strut and the bay doors. It is, therefore, recommended to further investigate this phenomenon, to include cavity-noise estimations in the current noise prediction models, and to eliminate such cavities where possible with the use of cavity caps, for example.