A mobile videophone image codec using wavelets and classified vector quantisation for mobile radio speech channels

A bandwidth efficient image codec is offered as a direct replacement for mobile radio speech codecs in second generation wireless systems, such as the Pan European GSM system. The image codec is contrived for Quarter Common Intermediate Format (QCIF) videophone sequences, and uses the 2D orthogonal wavelet transform to decompose the Displaced Frame Difference (DFD), into a four band structure using Quadrature Mirror Filters (QMF). The transformed coefficients are then compressed using classified vector quantisation (CVQ) and then multiplexed to a time division multiple access (TDMA) slot and modulated using Pilot Symbol Assisted Quadrature Amplitude Modulation (PSAQAM). The proposed image codec does not suffer from blocking effects that are visually disjointed and has therefore a major advantage over current discrete cosine transform (DCT) methods. A further advantage of using wavelet based transforms is that they require less hardware and are simpler to implement on a DSP chip than Fourier based methods. The method proposed is suitable for implementation in VLSI technology.