A close-to-capacity dirty paper coding scheme

Uri Erez*, Stephan ten Brink

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

224 Scopus citations

Abstract

The "writing on dirty paper" channel model offers an information-theoretic framework for precoding techniques for canceling arbitrary interference known at the transmitter. It indicates that lossless precoding is theoretically possible at any signal-to-noise ratio (SNR), and thus dirty-paper coding may serve as a basic building block in both single-user and multiuser communication systems. We design an end-to-end coding realization of a system materializing a significant portion of the promised gains. We employ multidimensional quantization based on trellis shaping at the transmitter. Coset decoding is implemented at the receiver using "virtual bits." Combined with iterative decoding of capacity-approaching codes we achieve an improvement of 2 dB over the best scalar quantization scheme. Code design is done using the EXIT chart technique.

Original languageEnglish
Pages (from-to)3417-3432
Number of pages16
JournalIEEE Transactions on Information Theory
Volume51
Issue number10
DOIs
StatePublished - Oct 2005

Funding

FundersFunder number
Deshpande Center for Technological Innovation, Massachusetts Institute of Technology

    Keywords

    • Dirty paper coding
    • EXIT chart
    • Interference cancellation
    • Iterative decoding
    • Minimum mean-squared error (MMSE) estimation
    • Mutual information
    • Repeat-accumulate (RA) codes
    • Vector quantization

    Fingerprint

    Dive into the research topics of 'A close-to-capacity dirty paper coding scheme'. Together they form a unique fingerprint.

    Cite this