DoA and ToA Estimation, Device Positioning and Network Synchronization in 5G New Radio: Algorithms and Performance Analysis

Location information plays a significant role not only in our everyday life through various location-based services, but also in emerging technologies such as virtual reality, robotics, and autonomous driving. In contrast to the existing and earlier cellular generations, positioning has been considered as a key element in future cellular networks from the very beginning of the fifth generation (5G) standardization process. Even though the earlier generations are capably of providing coarse location estimates, the achieved accuracy is far from the expected even sub-meter positioning accuracy envisioned in the context of 5G networks. In general, 5G new radio (NR) networks provide a convenient infrastructure for positioning by means of wider bandwidths, larger antenna arrays, and even more densely deployed networks especially at high millimeter wave (mmWave) frequencies. Building on dense 5G NR networks, this thesis focuses on the development of novel network-centric positioning frameworks by exploiting the existing NR reference signals. The contributions in this thesis can be grouped into topics based on the considered frequency ranges and the employed beamforming (BF) schemes therein.

First, novel cascaded algorithms for sequential device positioning are proposed assuming 5G NR networks operating at the lower sub-6 GHz frequency range and equipped with digital BF capabilities. In the first stage of the cascaded solution, two sequential estimators are proposed for joint direction of arrival (DoA) and time of arrival (ToA) estimation facilitating the received reference signals. Thereafter, the second-stage sequential estimators employing the obtained DoA and ToA estimates are proposed for joint positioning and network synchronization resulting in not only device location estimates, but also clock parameter estimates that are obtained as a valuable by-product. Such a choice stems from the fact that the ToA estimates are not feasible for positioning as such due to the clock instabilities in low-cost devices and the insufficient level of synchronization in the cellular networks. Second, a similar cascaded algorithm for joint positioning and network synchronization is proposed in the context of dense mmWave 5G networks and fundamentally different analog BFs. In particular, a novel joint DoA and ToA estimator is proposed by fusing information from multiple received beams based on a novel beam-selection method. In addition, the theoretical performance limits are derived and compared to those obtained using the digital BFs. The cascaded framework is completed with the second-stage positioning solution in a similar manner as in the case of digital BFs.

The performance of both frameworks is evaluated and analyzed in various scenarios using extensive computer simulations relying on the latest 5G NR numerology and a ray-tracing tool. Overall, this thesis provides valuable insights into practical positioning algorithms and their performance when relying solely on the 5G NR networks and available signalling therein. The obtained results in this thesis indicate that the envisioned sub-meter positioning accuracy is technically feasible using NR-based solutions.