Wyner-Ziv coding based on TCQ and LDPC codes

This paper considers TCQ and LDPC codes for the quadratic Gaussian Wyner-Ziv problem. After TCQ of the source input X, LDPC codes are used to implement Slepian-Wolf coding of the quantized source input Q(X) given the side information Y at the decoder. Assuming ideal Slepian-Wolf coding in the sense of achieving the theoretical limit H (Q(X)|Y), it is shown that Slepian-Wolf coded TCQ (SWC-TCQ) performs 0.2 dB away from the Wyner-Ziv distortion-rate function D_WZ*(R) at high rate. This result mirrors that of entropy-coded TCQ in classic source coding and establishes the connection between performances of high-rate Wyner-Ziv coding and classic source coding. Practical designs with TCQ, irregular LDPC codes (for Slepian-Wolf coding) and optimal estimation at the decoder perform 0.83 dB away from D_WZ*(R) at medium bit rates (e.g., ≥ 2.3 b/s). With 2-D trellis-coded vector quantization, the performance gap to D_WZ*(R) is only 0.66 dB at 1.0 b/s and 0.47 dB at 3.3 b/s.