Scalable Real-Time Video Compression for Data Link Applications

This project will develop a preliminary design for an operationally effective, scalable, real-time wavelet-based compression system for data link applications that can process still frames, continuous line scanned imagery and video sequences. The system will feature a single decoder design capable of decompressiong any of the image formats; a common compressed data representation (bitstream); dynamically adaptable compression rates; scalability in both spatial resolution and frame rate; capability to code a subset of the source image at higher quality than surrounding areas; and integrated channel coding techniques for resistance to noise, fading and interference. The design of data encoders and decoders will empasize construction with a small number of common application specific integrated circuits (ASICs) to be developed in a later phase. System performance will be demonstrated using non-real-time software simulations. The relationships between image quality, compression level, latency, channel corruption and processor resource requirements will be derived.