The One Centimetre Receiver Array (OCRA) will be a approximately 100-element close-packed horn array mounted at the secondary focus of a large paraboloidal radio telescope. With its large number of simultaneous beams OCRA will be able to carry out unbiased surveys of the radio sky at sensitivity levels which are currently impractical; it therefore offers the potential for making new astronomical discoveries. OCRA will use conventional waveguide horns and the radio frequency technology will be based on that being developed for the Low Frequency Instrument (LFI) on the Planck Surveyor satellite. The principal design problems to overcome are fluctuations in the atmospheric transmission during the observations and the intrinsic 1/f noise in the wide-band (approximately 10 GHz) receivers. The baseline design concept involves approximately 50 pairs of horns with each pair connected to an independent correlation receiver. An alternative concept involves approximately 100 total power receivers with a moving tertiary mirror to modulate their beam patterns on the sky at a rate faster than either the fluctuations in the atmosphere or the 1/f noise. The relative advantages and disadvantages of these different approaches are under investigation.