Learn More
—We present and analyze a joint source-channel coding strategy for the transmission of a Gaussian source across a Gaussian channel in n channel uses per source symbol. Among all such strategies, the scheme presented here has the following properties: i) the resulting mean-squared error scales optimally with the signal-to-noise ratio, and ii) the scheme is(More)
We study the fusion of data collected by multiple heterogeneous sensors that work cooperatively to achieve a common goal. This paper presents fast algorithms to fuse the sensor data. We map the problem into a graphical model and then develop a fast message-passing scheme to fuse the data. We simulate scenarios with 150 sensors and 200 targets that are(More)
An analog source is to be transmitted across a Gaussian channel in more than one channel use per source symbol. This paper derives a lower bound on the asymptotic mean squared error for a strategy that consists of repeatedly quantizing the source, transmitting the quantizer outputs in the first channel uses, and sending the remaining quantization error(More)
We consider source coding with a fidelity criterion, channel coding with a channel input constraint, and the combined problem of reproducing a source across a noisy channel. All three cases face a similar tradeoff between resource and performance, and the operating point with the highest performance per resource is of particular interest. In the case of(More)
Although more and more data is collected automatically, many interfaces still require manual input. When we, for example, enter our daily calorie intake or calculate our ecological footprint, we often have to guess the weight of the food or what distance we have covered with our car. In this paper, we propose a solution to overcome the problem of forcing(More)
  • 1