Stripping very low frequency communication signals with minimum shift keying encoding from streamed time-domain electromagnetic data

摘要

The objective of this research is to eliminate very low frequency (VLF) communication signals with frequencies in the range of 15-25 kHz from streamed time-domain electromagnetic data. The frequency-domain approach of notch filtering or bandwidth limitation is unsatisfactory when early delay-time data are required in a time-domain system. United States military VLF transmitters presently communicate 200 bits/s using minimum shift keying modulation, and it is possible to derive the encrypted bitstream from the data sampled at typical geophysical streaming rates. The method involves convolving the data with waveforms of frequency different by one-quarter of the bit rate above and below the carrier frequency and using the difference between the convolutions to predict the transmitted bits. The transmitted signal is then exactly recreated from the decoded bits, and the predicted signal at the receiver is subtracted from the data stream. It is possible to predict bit rates and encoding methods from other military transmitters through data analysis and again subtract the predicted signals from the streamed data. This procedure reduces the variance of data, implying that unwanted VLF signals have been reduced by a factor of three to nine in stations thousands of kilometers from a VLF transmitter. Much larger signal/noise improvements are predicted for stations within a few hundred kilometers of the VLF source. Lower degrees of improvement are noted from Indian and Chinese transmitters, which appear to have different encoding and modulation methods.