Timing Recovery for DOCSIS 3.1 Upstream OFDMA Signals
Xiao, Yayi 1990-
Data-Over-Cable Service Interface Specification (DOCSIS) is a global standard for cable communication systems. Before version 3.1, the standard has always specified single-carrier (SC) quadrature-amplitude modulation (QAM) as the modulation scheme. Given that the multi-carrier orthogonal frequency-division multiplexing (OFDM) technique has been increasingly popular and adopted in many wired/wireless communications systems, the newest cable communication standard, DOCSIS 3.1, also introduces OFDM as a major upgrade to improve transmission efficiency. In any digital communication systems, timing synchronization is required to determine and compensate for the timing offset from the transmitter to the receiver. This task is especially crucial and challenging in an OFDM system due to its very high sensitivity to synchronization errors. Although there have been many studies on the topic of OFDM timing synchronization, none of the existing methods are not directly applicable to DOCSIS 3.1 systems. Therefore, the main objective of this research is to develop effective and affordable timing synchronization algorithms for the DOCSIS 3.1 upstream signal. Specifically, three timing synchronization algorithms are proposed to comply and take advantage of the structure of the ranging signal (i.e., the signal used for synchronization purpose) specified in DOCSIS 3.1 standard. The proposed methods are evaluated under a realistic multipath uplink cable channel using computer simulation. The first algorithm makes use of the repetitive pattern of the symbol pairs in the ranging signal. The locations of the symbol pairs are determined by calculating a correlation metric and identifying its maximum value. The second and third algorithms are developed so that they exploit the mirrored symmetry of the binary phase-shift keying (BPSK)-modulated time-domain samples, corresponding to the first non-zero symbol in the ranging signal, and look for the exact location of the symmetry point. The first algorithm, with very low hardware complexity, provides reasonable performance under normal traffic and channel conditions. However its performance under a severe channel condition and heavy traffic is not satisfactory. The second and third algorithms provide much more accurate timing estimation results, even under the severe channel condition and heavy traffic flow. Since the second algorithm requires an enormous increase in hardware complexity, a few options are proposed to reduce the hardware complexity but it is still much higher than the complexity of the first algorithm. Applying the same complexity reduction techniques it is demonstrated that the third algorithm has similar hardware complexity to the first algorithm, while its timing estimation performance remains excellent.
DegreeMaster of Science (M.Sc.)
DepartmentElectrical and Computer Engineering
CommitteeKarki, Rajesh; Dinh, Anh; Chen, Daniel
Copyright DateOctober 2016
OFDMA timing recovery