Synchronization of Soqpsk-tg in Burst-mode Transmissions

In this paper, we present a maximum likelihood synchronization algorithm which jointly estimates frequency offset, symbol timing and carrier phase for shaped-offset quadrature phase-shift keying (SOQPSK) signals. We have considered a burst-mode transmission scenario in which a known training sequence is embedded in the beginning of each burst for the purpose of data-aided (DA) synchronization in a feedforward structure. The proposed algorithm first estimates the frequency offset independently from other parameters. The estimated frequency is then used to derive the symbol timing which is followed by the carrier phase estimation. The mean-squared error (MSE) of the proposed algorithm is computed via simulations. The results show that the proposed algorithm performs near the theoretical Cramér-Rao bound (CRB) at signal-to-noise ratios (SNRs) as low as 0 dB. INTRODUCTION Shaped-offset quadrature phase shift-keying (SOQPSK) is a physical-layer waveform that has seen extensive use in serial streaming telemetry (SST), and has been selected for future use in the integrated network enhanced telemetry (iNET) system. A key difference between SST and iNET is that iNET uses burst-mode transmission. The synchronization task becomes more challenging in burst-mode transmissions because there is little time for acquiring and locking onto the signal. In this paper, we introduce a maximum likelihood (ML) algorithm for feedforward DA synchronization of SOQPSK signals when transmitted over an additive-white Gaussian noise (AWGN) channel. Our algorithm jointly estimates the symbol timing, carrier phase and frequency offset based a known training sequence. The proposed approach first estimates the frequency offset from the collected samples corresponding to the training sequence. Once the frequency offset is known, the symbol timing is estimated, which is followed by the carrier phase estimation. The performance of the proposed method is investigated for two well-known SOQPSK schemes via simulations. The computed error variances reveal that our algorithm is capable of operating quite close to the theoretical Cramér-Rao bound (CRB) for all the synchronization parameters. The paper is organized as follows. The next section introduces the transmission model using SOQPSK signaling. Following that, we propose the ML estimation algorithm, which is derived based on the