and consumer sector networks
Wireless sensor networks (WSNs) are an emerging technology that will have a major societal, environmental and financial impact. By 2020, it is predicted that there will be more than 50 million interconnected nodes worldwide. WSNs seek to capture compressible analog signals and aim to transmit them reliably to one or more nodes, while maintaining low power consumption. Technical challenges include achieving communication reliability under delay constrains in harsh environments and limiting power consumption to ensure extended lifetimes. Current methods are often highly application-specific. In addition, they do not work well in environments with unknown signal-to-noise (SNR) ratio, have limited performance in multiuser scenarios and lack feedback functionality.
The invention is a low complexity architecture for communicating captured sparse signals in WSNs. The application-independent physical layer design leverages sparsity existing in many physical signals to parsimoniously represent them. By preserving the relative bit importance during transmission, it achieves graceful tradeoffs between distortion and channel SNR, resulting in increased robustness against channel errors across a wide range of SNR values in a rateless fashion. The invention was proved to be optimal in terms of distortion in the high SNR regime in point-to-point links and can efficiently serve multicast scenarios. The device is applicable to a wide range of applications and performs close to an idealized layered transmission scheme in terms of reliability and end-to-end distortion.
and signal model independent communication architecture
service of multiple receivers at their highest possible information rate
receiver feedback needed for optimal rate selection
computationally intense algorithms in transmitting sensor nodes