An Iterative Detection Aided Irregular Convolutional Coded Wavelet Videophone Scheme Using Reversible Variable-Length Codes and Map Equalization.

In this paper, we characterize the achievable performance of a wavelet encoding-based videophone system, which further compresses and protects the wavelet-based video codec´s bit-stream with the aid of reversible variable-length (RVLC) codes. More specifically, the video codec uses RVLC codes for compressing and protecting the video frame header. By contrast irregular convolutional codes (IRCCs) are used for unequal error protection of the video bitstream generated from the wavelet-subbands, when communicating over intersymbol interference (ISI) contaminated channels. At the receiver, the turbo-equalizer, the IRCC decoder and the RVLC decoder all employ the maximum a-posteriori (MAP) algorithm and their operations are performed in an iterative turbo-detection style. The proposed system was designed for maintaining a near-unimpaired video reconstruction quality for channel signal-to-noise ratios (SNR) in excess of about 5 dB, when communicating over a five-path dispersive AWGN channel.