Writing on Dirty Paper with Resizing and its Application to Quasi-Static Fading Broadcast Channels

Abstract

This paper studies a variant of the classical problem of ``writing on dirty paper'' in which the sum of the input and the interference, or dirt, is multiplied by a random variable that models resizing, known to the decoder but not to the encoder. The achievable rate of Costa's dirty paper coding (DPC) scheme is calculated and compared to the case of the decoder's also knowing the dirt. In the ergodic case, the corresponding rate loss vanishes asymptotically in the limits of both high and low signal-to-noise ratio (SNR), and is small at all finite SNR for typical distributions like Rayleigh, Rician, and Nakagami. In the quasi-static case, the DPC scheme is lossless at all SNR in terms of outage probability. Quasi-static fading broadcast channels (BC) without transmit channel state information (CSI) are investigated as an application of the robustness properties. It is shown that the DPC scheme leads to an outage achievable rate region that strictly dominates that of time division.

To appear in IEEE International Symposium on Information Theory 2007

References in corpus (1)

• On the Capacity of Fading MIMO Broadcast Channels with Imperfect Transmitter Side-Information

Cited by in corpus (7)

• Ergodic Fading MIMO Dirty Paper and Broadcast Channels: Capacity Bounds and Lattice Strategies
• Relay Channel with Orthogonal Components and Structured Interference Known at the Source
• Filter and nested-lattice code design for fading MIMO channels with side-information
• An Upper Bound on the Capacity of Vector Dirty Paper with Unknown Spin and Stretch
• On the Capacity of the Dirty Paper Channel with Fast Fading and Discrete Channel States
• On the Dirty Paper Channel with Fast Fading Dirt
• Worst-Case Expected-Capacity Loss of Slow-Fading Channels