Multipath Time of Arrival Estimation Algorithm Based on Successive Interference Cancellation in NB-IoT Systems

Abstract

Narrow-band Internet of Things (NB-IoT) is a radio access technology of 5G with super coverage, low power, low cost, massive connections and a reduced set of functionalities. The NB-IoT systems are widely applied in industry such as smart transportation and smart community. Many NB-IoT industrial applications need a positioning feature for tracking and equipment monitoring. The Observed Time Difference of Arrival (OTDOA) is a downlink positioning method in NB-IoT systems, in which a User Equipment (UE) measures the Time of Arrival (TOA) of narrowband positioning reference signals (NPRS) from multiple base stations (eNodeBs). These TOAs from several neighbor eNodeBs are subtracted from a TOA of a reference eNodeB to form Reference Signal Time Difference (RSTD) measurements, which are reported to the networks for the UE positioning. A UE can obtain TOA by measuring the time delay of the first tap in the estimated channel impulse response. The performance of TOA/OTDOA estimation suffers from 180 kHz bandwidth limitation, low sampling-rate and low-cost in NB-IoT systems, especially under multipath channel and inter-cell interference environment. This paper proposes an enhanced multipath time of arrival estimation algorithm to improve the accuracy of TOA/OTDOA estimation by detecting the first channel tap in an estimated channel impulse response, estimating the frequency offset and channel parameters of each received signal, cancelling interfering signals from inter-cells iteratively and interpolating the correlations of time-domain NPRS signal evaluated at low sampling rate to a high sampling rate of 61.44MHz. The performance of the proposed method is evaluated via simulation. The performance of TOA/OTDOA estimation is shown to outperform the performance of conventional TOA estimation techniques that uses signal-to-noise fixed threshold or non-iterative method.