Symbol Rate Estimation for DVB-S2 Broadcasting

Abstract

In this paper we present a Maximum Likelihood (ML) approach to the symbol rate estimation of a linearly modulated signal, as those found in satellite broadcasting with DVB-S [1] or DVB-S2 [2]. We focus in the general case for which all the sync parameters, including the mentioned baud rate, are unknown at the receive processing unit. This blind approach finds its main application on the passive analysis of signals, including automatic classification of modulations and quality monitoring. As a motivating starting point, note that the symbol rate can differ among different broadcasters for the same type of service. Let us consider, for example, the signals used in digital TV satellite broadcasting employing DVB-S or the new upcoming standard DVB-S2. It is the purpose of this paper to show how to exploit the structure of the signal to anticipate the underlying symbol rate without resorting to the trial and error decoding for all possible symbol rates.