Also found in: Dictionary, Wikipedia.
aperture synthesisA technique originally of radio astronomy devised by Sir Martin Ryle in the late 1950s. It is used to obtain high resolution images by combining signals from a number of relatively small antennas spaced over an area equivalent to a large aperture. Although originally used only for synthesized apertures that were ‘fully filled’ by smaller antennas, the term is also applied to instruments for which no clearly definable equivalent large aperture exists. The antennas are connected in interferometric pairs (see radio telescope) and the amplitudes and phases of the signals from a radio source are recorded on all antenna spacings necessary to cover the required area. The information is then processed in a computer where a Fourier transform is carried out to generate a radio map of the source.
In practice the rotation of the Earth is often used to move the antennas in space – a technique known as Earth-rotation synthesis. In such an arrangement, just 12 hours of observation generate all the baselines available from a particular configuration; the same baselines are repeated with the opposite sense during the next 12 hours. For a radio interferometer in which all the baselines are oriented east-west (e.g. the Ryle Telescope and the WSRT), each baseline produces an elliptical track of synthesized aperture. The major axis of the ellipse is determined by the baseline length, whereas the minor axis depends on the sine of the declination of the source. Low declination sources cannot therefore be mapped with a uniformly high resolution by an east-west interferometer. The VLA largely overcomes this difficulty by arranging its dishes in a Y-shaped configuration.
The computer processing involved in map-making can become very sophisticated, and special software (such as AIPS) is usually used. The Fourier transform of the synthesized aperture defines the synthesized beam (or dirty beam) of the telescope, with which the final image will appear to be convolved (see convolution). A synthesized aperture that is not fully filled will produce false features or artifacts in the map, and these must be suppressed using deconvolution algorithms such as Clean or MEM. An aperture consisting of only a few equally spaced strips will show bright, widely spaced concentric rings or grating responses (see array) around features in the map. Even a fully filled synthesized aperture may need to be graded, by gradually reducing the weight of contributions from the longer baselines. This reduces the side-lobe response of the telescope, which is the residual structure of the synthesized beam lying outside its central region. Other techniques such as self-calibration are used to remove artifacts generated by instrumental drifts and the effects of the atmosphere.