Implementation of video codec with programmable parallel DSP

The implementation of a video codec with a programmable parallel digital signal processor (DSP) is described. The advantages of DSP implementation over dedicated VLSI implementation are discussed. System design concepts and the architecture for a multicomputer-type DSP system called NOVI are reviewed. This system has been developed as a prototype for program development and performance evaluation of parallel processing systems. Intraframe DCT (discrete cosine transform) codec is implemented by NOVI as an example of a transform-type video codec. The processing speed of the implemented codec is examined. This codec can process 9 frames/s (one frame consists of a 256*240 pixel image) using 34 PEs (processing elements): 16 PEs for the coder, 16 PEs for the decoder, and 2 PEs for the video I/O. Packet assembly and multiplexing are also performed by the codec.<<ETX>>

[1]  Corporate Inmos Limited.  Transputer Reference Manual , 1988 .

[2]  William K. Pratt,et al.  Scene Adaptive Coder , 1984, IEEE Trans. Commun..

[3]  Kevin P. McAuliffe,et al.  The IBM Research Parallel Processor Prototype (RP3): Introduction and Architecture , 1985, ICPP.

[4]  K. Ramachandran,et al.  A 30 MHz programmable CMOS video FIR filter and correlator , 1988, 1988., IEEE International Symposium on Circuits and Systems.

[5]  H.J. Mattausch,et al.  Video chip set for data rate compression by filtering and DPCM coding , 1988, 1988., IEEE International Symposium on Circuits and Systems.

[6]  Sadayasu Ono,et al.  Architecture and performance of a multicomputer type digital signal processing system 'NOVI' , 1988, ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing.

[7]  Francis Jutand,et al.  A one chip VLSI for real time two-dimensional discrete cosine transform , 1988, 1988., IEEE International Symposium on Circuits and Systems.

[8]  B. Lee.  A new algorithm to compute the discrete cosine Transform , 1984 .

[9]  P. Wiley.  A parallel architecture comes of age at lastl hypercube machines reach the market on the promise of executing billions of operations a second to solve difficult simulation problems , 1987, IEEE Spectrum.

[10]  Inmos Limited,et al.  Occam Programming Manual , 1984 .

[11]  Hiroshi Deguchi,et al.  A parallel processing scheme for tree-structured image generation system , 1988, 1988., IEEE International Symposium on Circuits and Systems.

[12]  N. Ohta,et al.  A load balancing technique for video signal processing on a multicomputer type DSP , 1988, ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing.