标题： 关于衰落信道中缓冲传输的端到端失真 显示英文标题

标题： On the End-to-End Distortion for a Buffered Transmission over Fading Channel

摘要： 本文研究了通过衰落信道传输的模拟源端到端失真/延迟权衡问题。模拟源被量化并存储在缓冲区中，直到被发送。就缓冲区延迟而言，有两种极端情况：无延迟和无限延迟。我们观察到，引入缓冲区延迟可以获得显著的功率增益。我们的目标是研究这两种极端情况之间的状况。利用最近提出的 \emph{有效容量} 概念，我们推导出了该权衡关系的闭式公式。对于单输入单输出（SISO）的情况，得到了失真-延迟曲线的渐近紧致上界，当 $\mathcal{D}_\infty \exp(\frac{C}{\tau_n})$ 时，该上界趋于无限延迟下的下界，其中 $\tau_n$ 是归一化延迟，$C$ 是常数。对于更一般的多输入多输出（MIMO）信道，我们计算了失真与信噪比（SNR）的指数——高信噪比（SNR）情况下期望失真的指数衰减速率。数值结果表明，引入少量延迟可以节省大量传输功率。 显示英文摘要

摘要： In this paper, we study the end-to-end distortion/delay tradeoff for a analogue source transmitted over a fading channel. The analogue source is quantized and stored in a buffer until it is transmitted. There are two extreme cases as far as buffer delay is concerned: no delay and infinite delay. We observe that there is a significant power gain by introducing a buffer delay. Our goal is to investigate the situation between these two extremes. Using recently proposed \emph{effective capacity} concept, we derive a closed-form formula for this tradeoff. For SISO case, an asymptotically tight upper bound for our distortion-delay curve is derived, which approaches to the infinite delay lower bound as $\mathcal{D}_\infty \exp(\frac{C}{\tau_n})$, with $\tau_n$ is the normalized delay, $C$ is a constant. For more general MIMO channel, we computed the distortion SNR exponent -- the exponential decay rate of the expected distortion in the high SNR regime. Numerical results demonstrate that introduction of a small amount delay can save significant transmission power.