Low-Complexity Sub-Optimal Cell ID Estimation in NB-IoT System

This paper deals with cell ID estimation in narrowband-internet of things (NB-IoT) system. The cell ID value is carried by
the narrowband secondary synchronization signal (NSSS). We suggest a low-complexity sub-optimal estimator, based on the auto-
correlation of the received observations. It is up to thirty times less complex than the optimal maximum likelihood (ML) estimator
based on cross-correlation. In addition, we present three methods allowing the receiver to take advantage of the different repetitions
of the NSSS. They are based on a hard decision after every estimation, a soft combination of the different observations of the NSSS,
and an hybrid mix between the two firsts, respectively. The advantages and drawbacks of the presented techniques are stated, and a
performance analysis is proposed, which is further discussed through simulations results. It is shown the that different methods reach
the performance of ML after several repetitions for a lower overall complexity.