This thesis addresses the problem of cell loss when transmitting video data over an ATM network. Cell loss causes sections of an image to be lost or discarded in the interconnecting nodes between the transmitting and receiving locations. The method used to combat this problem is to use a technique called Error Concealment, where the lost sections of an image are replaced with approximations derived from the information in the surrounding areas to the error. This technique does not require any additional encoding, as used by Error Correction. Conventional techniques conceal from within the pixel domain, but require a large amount of processing (2N2 up to 20N2) where N is the dimension of an N×N square block. Also, previous work at Loughborough used Linear Interpolation in the transform domain, which required much less processing, to conceal the error.