Efficient multiplierless architecture for frame synchronization in DVB-S2 standard

Abstract

The first challenging step of the demodulation of the DVB-S2 signal with function of VCM (Variable Coding and Modulation)/ACM (Adaptive Coding and Modulation) is the detection of the Physical Layer (PL) header. PL header is transmitted using π/2-BPSK modulation and is composed of a fixed part (26 bits of Start Of Frame (SOF)) and a variable part (64 bits codeword of PL Signaling (PLS) code that defines the structure of the PL frame). Since the 90 bits corresponding to the PL header are affected by noise, the carrier frequency offset and the phase noise, the synchronization task in a DVB-S2 receiver is thus a critical task. In this paper, we present a properties of the Hadamard code used to encode the information of the PLS code to reinforce frame detection before knowing the actual value of the PL code. Moreover, we propose to perform the computation in the polar domain in order to avoid the need of multiplier and thus, to obtain a very low cost implementation. The associated decoder architecture is presented together with the measured performance at several SNRs.

A polar-based demapper of 8PSK demodulation for DVB-S2 systems

“…In terms of performance, both maxlog approximation and polar approximatoin lead to similar results. The main application domain of the polar LLR computation is for "polar based" demodulator [2], since it avoid the use of a CORDIC to transform back polar coordinates to cartesian coordinnates.…”

“…In [2], the authors presented a synchronization method using polar coordinates. In this paper, we extend the use of polar coordinates to also compute the LLR.…”

Improved Multiplierless Architecture for Header Detection in DVB-S2 Standard

“…In [2], the Hadamard structure of the PLS field is fully exploited and combined with the SOF field to achieve better detection performance. In [3] a hardware implementation of a method called JOINT that combines SOF and PLS detectors is proposed: a detection is declared when both SOF and PLS detectors agree. The extra cost of hardware induced by the PLS detector allows reducing the P M D for more than 45% compared to SOF detectors for a P F A = 10 −6 and E s /N o = -3 dB.…”

“…In this section variations of original SOF and PLS detectors presented in [3] are proposed for improving performance and reduce implementation complexity.…”

“…The remaining parts of this paper is organized as follows: Section II gives a brief state of the art header detection algorithms. Section III presents the proposed variations to SOF and PLS algorithms presented in [3]. Section IV introduces Global Summation method that combines SOF and PLS detections, and compares its performance to already published solution.…”