The use of simulations to examine various trade-offs and estimating the scientific return from the space VLBI missions RADIOASTRON and VSOP is discussed. The impact of the on-board spacecraft constraints is determined.

A space project for studying ecological Earth problems is being carried out by means of radio techniques. A 30 m prototype antenna has already been deployed and tested. The radio telescope will be launched in 1994 into a circular orbit haying an altitude 600 km and an inclination of 65°. The planned mission time is ≥ 1.5 year; 25% of this will be available for VLBI observations at wavelengths 6 and 18 cm.

It is shown that the influence of a baseline knowing error on the fringe rate change during coherent averaging time interval is extremely increasing for space VLBI due to more angle velocity of a space baseline rotation comparatively with ground VLBI. It is shown on the example of Radioastron orbit that the reguirement to its orbit for such a short wavelength as 1.35 cm is determined by unknowing of the fringe rate change during coherent averaging time interval rather than by delay ambiguity and this reguirement is equal several meters.

We have measured the opacity at the Owens Valley Millimeter Array with tip curves at frequencies from 95 to 248GHz. These measurements have been interpreted using a model and then scaled to a standard frequency of 100GHz. The unique aspect of this data is the wide range of frequencies covered and the use of receivers with different sideband gain ratios.

The measurements being made to pick a site for the NRAO Millimeter Array are described. The instruments, techniques and preliminary results are discussed, along with the strategy for picking the final site for the array.

An overview of the software systems developed for the broad-banded Ryle Telescope.

Two important episodes in the early development of interferometry in radio astronomy are traced in detail. The first is the use of the sea-cliff interferometer at the Radiophysics Laboratory in Sydney, first by Pawsey for solar observations and later by Bolton for radio star surveys. The second is the development of the Michelson interferometer and the phase switch by Ryle in Cambridge. This also was employed for important observations of the sun and radio stars.

In the spring of 1964, having qualified for a sabbattical leave from the University of Illinois and having recently completed two years as Chairman of the Visiting Committee of the National Radio Astronomy Observatory, I inquired of the NRAO staff as to whether a desk and possibly some augmentation of my University half-salary might be available at Green Bank for the following academic year. Instead I was invited to join the staff as a fulltime employee and to take a coordinating role in the development of the “very large array” as Chairman of the Design Committee. The University granted me a year’s leave of absence which eventually stretched to four years.

A method of improving image fidelity by using observations at multiple frequencies is described. We discuss the power and possible limitations of the technique. Results of narrow frequency range observations with MERLIN and global VLBI are presented. We conclude with a consideration of the possible future impact of this technique.

The Cambridge Low Frequency Synthesis Telescope (CLFST) has baselines which are offset from the E-W direction by about 3°. The wide-field mapping problem has been overcome by the use of beam-sets, or artificial maps consisting of grids of synthesized beams, which incorporate the effects of both the geometric and chromatic aberration. By appropriate interpolation we can find a good approximation to the beam-shape at any point on a map.

The general relation between interferometric data and sky brightness requires the use of three-dimensional Fourier transforms. Only in special cases does the standard two-dimensional Fourier relations result in images free of distortions. For the VLA, use of two-dimensional transforms introduces errors which are important at low frequencies and high resolutions. We review options and progress in solving the computational problems associated with the full solution.

We discuss critically the methods used to complement synthesis maps with short-spacing information derived from single-dish observations. The conclusion is that from a single synthesis field the correct result cannot be derived, a large area has to be synthesized.

Many synthesis arrays currently in use have at least a few redundant baselines which can be used to derive model-independent telescope corrections. We discuss the relative merits of redundancy-based versus standard selfcalibration methods with respect to sensitivity to noise and interference. We then derive an approach to optimally combine the two methods for partially redundant arrays

We propose an application of the bispectrum analysis technique in millimeter interferometry, which is used in the speckle masking technique. For the millimeter interferometry, in which atmospheric phase fluctuations are very serious, not only bispectrum phase but also bispectrum amplitude is good observable. We have developed an algorithm to reconstruct images directly from bispectrum. We have made good maps with this algorithm in the case of both simple and compact sources.

The Molonglo Observatory Synthesis Telescope (MOST), located near Canberra, Australia, is an east-west earth-rotation synthesis array which, unlike conventional synthesis arrays, generates multiple real-time fan-beams and forms images by the back-projection of the recorded intensities rather than by Fourier inversion of visibilities. The images produced in this fashion are often marred by the presence of radial artefacts emanating from strong sources due to residual calibration errors. An algorithm devised to self-calibrate these errors is discussed.

The procedure for polarisation measurements with radio synthesis telescopes is fraught with uncertaincies and shaky assumptions. Therefore, their accuracy will be limited to a few percent at best. It is investigated here whether it will be possible to improve this accuracy by a factor 10-100 with the WSRT, which has excellent polarisation characteristics, very small closure errors, and redundant spacings. The latter can be used to calibrate the instrument in a way which is less dependent upon assumptions about the polarisation of the calibrator source. They also cause more reliable convergence of the SELFCAL model, by reducing the number of free parameters. Although the investigation is not yet finished, the tentative conclusion is that it may be possible, with crossed dipoles, to derive completely model-independent values for Q/I, U/I and V/I.

The Giant Metrewave Radio Telescope (GMRT) that is being built to operate at metre wavelengths, will encounter a number of problems that are specific to synthesis imaging at low frequencies. Some of these problems are highlighted and possible solutions that are being considered, are discussed.

At low frequencies, the typical fields of view over which imaging is required is several degrees across, for which (a) the visibility and brightness distribution cannot be assumed to be related by a two-dimensional Fourier Transform, and (b) the propagation delays introduced by the atmosphere cannot be assumed constant within the field of view. These are adequately represented in a method discussed in this paper. Although compute-intensive, the method is suitable for implementation on parallel processors.