LAUSR

In the era of Internet of Things, so many wireless networked sensors and smart devices are embedded in infrastructures. For so many different wireless sensors, accurate times tamps are important to interpret data. In general, time stamps can be obtained by atomic clocks and GPS. The paper used the natural timestamping which is based on the grid frequency estimation. For grid frequency estimation, the paper proposed maximum merging model in fractional cepstrum domain for dealing with complicated power grid channel interference in China grids with data measured by the wireless sensor networks. The proposed method outperforms existing power grid frequency measurement algorithms, because the existing do not consider the facts that harmonic is time-varying and the pass band of some recording devices do not cover grid frequency. Through analyzing the results of analog frequency measurement, actual voice frequency measurement and recording audio tampering detection, this method is better than the one Bykhovsky proposed. It can be preferably applied to the grid frequency measurement, recording location recognition and audio tamper detection.