Study on real-time sensor signal transmission based on technology of polynomial predictive filtering

Transmission uncertainty is an unavoidable drawback of network communication. Former study results of real-time sensor signal transmission could not be applied generally without considering the problem from the perspective of signal features. Effects resulting from transmission uncertainty are alleviated by a method which is presented in the paper. Polynomial predictive filtering (PPF) and channel coding technology are used to solve problems of real-time sensor signal transmission. In this paper a strategy is introduced, followed by some samples discriminated and transmitted selectively, which are generated from the remote sensor. These samples transmitted through the network can be restored by prediction in receiver, with the signal reconstructed. From the perspective of users, it seems that signal sampled by a local sensor. By constructing a virtual local sensor, transmission uncertainty can be transformed into output errors of the virtual local sensor. While the network bandwidth can be utilized effectively by transmitting samples selectively.