A Comparison of "Direct RF Sampling" and "Downconvert & Sampling" GNSS Receiver Architectures

The new direct RF sampling GNSS receiver front concept has been analyzed, experimentally tested, simulated, and compared to a traditional superheterodyne front end that uses mixing to down-convert the signal before sampling. The goal of this study has been to demonstrate that signal power and carrier tracking are not adversely influenced by the use of direct RF sampling. The direct RF sampling strategy that has been examined uses sampling rates that are much lower than the carrier frequency but that are more than twice the information bandwidth. Such a system employs a bandpass filter in front of the analog-todigital converter to avoid sensitivity loss via aliasing of out-of-band noise and interfering signals. This design approach has been evaluated analytically to determine the effects of sample clock jitter, and experimental and simulation results have been used to confirm the analytical results. The experiments have used computer data acquisition systems to record the outputs of direct RF sampling front ends that were connected to roof-top patch antennas. The results indicate that direct RF sampling yields equivalent performance to superheterodyne mixing if sample timing jitter is kept below well defined limits.