GNSS technique helps pedestrians navigate 'urban canyons'

A new smartphone-based global navigation satellite system (GNSS) positioning technique has been developed, with the aim of overcoming the challenges of so-called ‘urban canyons’. By enhancing measurement redundancy, the technique addresses the shortcomings of traditional GNSS systems in complex urban landscapes where tall buildings often block or reflect satellite signals, leading to inaccuracies in pedestrian navigation.

GNSSs are essential for many location-based services, but their effectiveness is severely compromised in urban canyons — ie, streets that are flanked on both sides by tall buildings such as skyscrapers. Dense clusters of tall buildings cause multipath and non-line-of-sight (NLOS) errors, leading to significant positioning inaccuracies, particularly in the cross-street direction. This makes it challenging for pedestrians to determine which side of the street they are on — a critical element for accurate navigation and safety. Furthermore, existing solutions, such as 3D building models or deep learning-based signal classification, either require impractical resources or extensive training data. There is thus a clear demand for a more accessible, reliable approach to enhance pedestrian navigation in such environments.

Researchers from Shenzhen University, The Hong Kong Polytechnic University (PolyU) and Wuhan University have now developed a new GNSS positioning technique, which they described in the journal Satellite Navigation. The new method, dubbed ‘sidewalk matching’, uses smartphone sensors combined with pedestrian maps to improve positioning accuracy in challenging urban canyons. By dynamically identifying the side of the street and filtering out NLOS signals, this technique significantly improves GNSS accuracy, making it ideal for real-time pedestrian applications.

The team’s algorithm works by identifying which half of the sky has more line-of-sight (LOS) signals, enabling it to determine the correct side of the street. By analysing the carrier-to-noise ratio (C/N0) and satellite azimuth angles, the system can filter out faulty measurements using a sliding window method, meaning that only reliable data is used. A key feature is the incorporation of pedestrian dead reckoning (PDR), which uses smartphone accelerometers and gyroscopes to validate GNSS positions, especially at intersections where traditional methods may struggle. Extensive tests conducted in Hong Kong demonstrated that the technique provides positioning accuracy of less than 5 m — far superior to traditional GNSS systems, which can have errors exceeding 18 m in urban canyons.

“This sidewalk matching technique represents a significant leap forward in urban pedestrian navigation,” said lead researcher Dr Duojie Weng, from Shenzhen University and PolyU. “By combining smartphone sensors with pedestrian network data, we’ve created a practical solution that doesn’t require costly 3D models or large training datasets. This makes it an accessible, cost-effective solution for a wide range of real-world applications, from pedestrian safety to ride-hailing services.”

The sidewalk matching technique has far-reaching implications for various pedestrian-centric applications. It promises to significantly improve pedestrian collision avoidance systems by providing real-time, precise positioning even in densely built urban areas. Ride-hailing services can leverage this technology to enhance pick-up and drop-off accuracy, while jaywalking monitoring systems can more reliably determine which side of the street a pedestrian is on. Additionally, the method could be a game changer for visually impaired individuals, providing greater independence and safety in urban environments. Thanks to its reliance on widely available smartphone technology, this innovation has the potential to become a standard feature in future navigation systems, revolutionising how pedestrians navigate cities.

Image credit: iStock.com/yongyuan