Characterisation and Calibration of Smartphone GNSS Antenna for Enhanced Positioning

Global Navigation Satellite Systems (GNSS) are integral to modern positioning technologies with applications spanning multiple industries that require precise positioning. While geodetic-grade systems achieve high precision through advanced hardware, smartphones with their compact and cost effective GNSS antennas face significant limitations. These limitations arise primarily from the embedded nature of their components, the absence of calibration standards, and susceptibility to environmental and electronic interference. This research investigates the characterisation and calibration of embedded smartphone GNSS antennas with a specific focus on the Xiaomi Mi8, to enhance positioning accuracy through systematic calibration techniques. The study is structured into three progressive stages. Stage 1 establishes experimental controls and investigates smartphone GNSS parameters. Key considerations include hardware and software configurations, signal isolation, reproducibility of test conditions, and the influence of the parameters such as tilt, heading, Wi-Fi, and Bluetooth were systematically tested. A Xiaomi Mi8 smartphone equipped with the Broadcom BCM47755 dual-frequency GNSS chipset was selected as the test device due to its advanced capabilities. Raw data collection and subsequent parameter tests were conducted in controlled environments, including anechoic chambers and open sky field setups. To ensure the reliability and repeatability of results, a Record and Playback System (RPS) was utilised for signal replication. Stage 2 focuses on developing calibration models tailored to the Xiaomi Mi8 GNSS antenna. The Phase Centre Offset (PCO) and Phase Centre Variation (PCV) were quantified using absolute calibration techniques in collaboration with the Robotic Calibration Facility at Geoscience Australia. This involved extensive testing to isolate the antennas characteristics from other hardware components and environmental factors. Challenges addressed included smartphone battery longevity, session logging time limitations and smartphone to calibration software communication. The calibration outcomes provide precise metrics for the Xiaomi Mi8 GNSS antenna, setting a foundation for the subsequent evaluation phase. Stage 3 evaluates the impact of the calibration model on positioning performance across various scenarios including application usage and device orientation. The calibration models were integrated into a bespoke MATLAB data positioning program, and their effectiveness were analysed for positioning accuracy with the application of the Phase Centre Correction. Comparative tests demonstrated that the calibrated Xiaomi Mi8 GNSS antenna achieved measurable improvements in accuracy, particularly for scenarios involving complex device orientations. The research revealed several critical findings. First, it confirmed that uncalibrated smartphone GNSS antennas contributed significantly to positioning errors, with deviations largely attributed to unknown PCO and PCV values. Second, the study highlighted the impact of smartphone specific factors such as internal electronic interference and device orientation on GNSS positioning performance. Third, the calibration model demonstrated its ability to enhance accuracy by accounting for these factors, particularly with the application of a dual file antenna calibration technique for the tilted smartphone. This study makes a substantial contribution to the field by presenting alternative ways to calibrate low-cost embedded GNSS antennas resulting in increased positioning accuracy. This methodology has the potential to make it feasible for use in consumer-grade devices. Moreover, the findings provide insight into optimising GNSS performance in smartphones, with potential applications extent extending to other compact devices such as wearable technology and IoT systems. Future research could explore the integration of multi-frequency and multi-constellation capabilities to further enhance the accuracy a smartphone GNSS antennas. The study also underscores the need for industry standards in smartphone GNSS calibration to ensure consistency and reliability across devices and manufacturers. In conclusion, this research provides a robust framework for improving the positioning accuracy of embedded GNSS antennas in smartphones by addressing key challenges through systematic testing and calibration modelling. By advancing the understanding and quantification of antenna specific parameters, it paves the way for significant improvements in positioning technologies benefiting a wide array of industries reliant on GNSS applications.<p></p>