A Concept for Data-Aided Carrier Frequency Estimation at Low Signal-To-Noise Ratios

In this paper, we consider data-aided carrier frequency estimation for burst mode transmission in low signal-to-noise ratio (SNR) environments. It is well known from literature that it is advantageous for the accuracy of data-aided carrier frequency estimation to choose separate packets of pilot symbols as pilot constellation. We here introduce a new concept to treat such pilot symbol constellations. This concept lowers the SNR threshold for considered frequency estimator and, therefore, allows for accurate frequency recovery at low SNR. Furthermore, it is demonstrated that the approach improves the accuracy of carrier frequency estimation also in the presence of fading.