5G positioning: Locating 5G devices anywhere

The positioning market is huge and continues to grow. While its value was USD 1.27 billion in 2024, it is projected to expand at a compound growth rate (CGR) of 28 percent from 2025 to 2033, according to Growth Market Reports [1]. 5G standalone networks provide consumers, enterprises, and the public sector with precise positioning services that can work seamlessly indoors and outdoors. 5G-based solutions that deliver both high-quality connectivity and positioning services through widely deployed cellular networks can significantly lower deployment, management, and maintenance costs compared to other positioning solutions that require specialized infrastructure.

Positioning is an increasingly important key enabler for logistics, industrial automation, advanced driver assistance systems, digital airspace services, digital representations of cities or factory floors, and many other applications, creating added value for various use cases. Figure 1 shows use cases divided into three categories:

• insights, provided by relatively coarse positioning accuracy for use cases like network optimization and anonymous traffic analysis for urban planning
• monitoring and tracking for use cases such as asset tracking and emergency services, which require medium positioning accuracy
• autonomy for use cases requiring high precision, such as operation of autonomous mobile robots, geofencing, collision avoidance, and navigation

All the three use case categories benefit from standardized interfaces between the application and the actual positioning mechanisms– network exposure application program interfaces (APIs). Different use case types benefit from different APIs or configurations of APIs and can be seen as an interface layer between applications and the 5G network.

Positioning based on global navigation satellite systems (GNSS) is nowadays taken for granted and provides sufficient accuracy in many outdoor areas. However, GNSS does not work well for indoor use cases, where a large fraction of 5G devices are located, or in other scenarios with limited line of sight (LOS) to satellites, like urban canyons. Indoor positioning solutions are more scattered and are based on technologies such as ultra-wideband (UWB), Bluetooth, Wi-Fi, vision, and light detection and ranging (lidar). A main drawback with several of these solutions is that they require additional infrastructure for positioning, which in turn drives increased costs for deployment, management, and maintenance.

5G positioning today offers accuracy and reliability that is already sufficient for many use cases, both indoors and outdoors, such as locating equipment in hospitals or factories, or locating an emergency caller. The positioning accuracy ultimately depends on the deployment density—how far apart the base stations are—as well as the amount of clutter in the environment, and measurement and computation capabilities of the base stations. The primary technology in 5G positioning is radio-based, relying on transmission and reception points (TRP) and their known locations in the environment. The device or user equipment (UE) position is unknown and can be positioned using radio transmissions or receptions at the TRP, which in cellular networks is also called a base station. Cellular base stations scale from large high-power radios with advanced antenna systems (AAS) to low-power radios for indoor deployments. Radio waves can be exploited to derive either their time of arrival (TOA) or angle of arrival (AOA) at the base station, see Figure 3 and Figure 4, respectively, and forward the origin of the radio wave.

The positioning support in 3GPP standards has continually evolved since Release 16:

• Release 16 supports a positioning accuracy of less than three meters for indoor scenarios
• Release 17 supports horizontal accuracy to within 20 centimeters
• Release 18 adds bandwidth aggregation, carrier phase, and GNSS augmentation to provide potential for centimetric-level precision[2]
• Release 19 introduces the artificial intelligence and machine learning (AI/ML) realm, which improves the non-line-of-sight (NLOS) performance of radio-based solutions

In practice, positioning errors below three meters with 90 percent reliability can be expected indoors in cluttered environments, while significantly better performance can be achieved in open spaces, even if several error sources come into play. Our current indoor tests in lightly cluttered environments show that sub-meter positioning accuracy is possible with 90 percent reliability.

In outdoor scenarios, positioning errors below 20 meters can be achieved when the device has LOS or near-line-of-sight to a base station equipped with an advanced antenna system, which can estimate the AOA of uplink radio signals. If the device to be positioned is not within LOS, errors are typically below 50 meters.

The main technical components of 5G positioning are shown in Figure 2. The passive positioning service is inherently available in cellular networks as the position of a device can be determined based on regular signal strength measurements. Also, knowing in which network cell the UE is present gives an approximate position, which can be good enough for certain use cases. Since cell sizes depend on the distances between base stations, the achievable positioning accuracy is different in dense urban, suburban, and rural areas.

Outdoor use cases can be supported by measuring the AOA and estimating the distance to the device. If increased accuracy is required, the positioning accuracy can be improved by cell site densification.

Indoor use cases require less coverage area and are supported by techniques such as time difference of arrival, which use low-power and low-complexity base stations.

GNSS accuracy outdoors can be further bolstered by adding assistance for real-time kinematic (RTK) positioning via the cellular network, which greatly improves accuracy.

The main advantage of 5G positioning is that the same infrastructure and devices are used for both connectivity and positioning. This results in simpler deployment and reduced maintenance compared to solutions such as Bluetooth and UWB, which require additional infrastructure for positioning.

Another advantage is that 5G positioning provides seamless positioning service with only a negligible service interruption when moving between indoor and outdoor scenarios, as shown in Figure 5. This is important in use cases such as logistics, asset tracking, national security, and public safety scenarios.

Many sensor-based positioning solutions, such as inertial measurement units (IMU), cameras and lidars on an autonomous guided vehicle, provide positioning in native local and/or relative coordinates, thereby requiring translation into the global coordinate system. Whereas 5G base stations are already located in the global coordinate system thus positioning estimates are also naturally provided in global coordinates.

While GNSS-based positioning is taken for granted, GNSS jamming and spoofing threats have increased recently and revealed weaknesses in this solution. Further, ionospheric activity causes disturbances and coverage in deep urban canyon use cases is also an issue. 5G can also warn devices about GNSS issues via unicast or broadcast information. 5G positioning is less susceptible to external interference as it uses a terrestrial infrastructure with significantly higher received signal strength than GNSS, making it more secure against jamming and spoofing.

In radio-based positioning, the LOS path or first direct path from the transmitter to the receiver is the desired signal, while any other reflections and refractions from the environment may make the position estimate erroneous if they are falsely identified to be the LOS path. Detecting the LOS radio path in multipath environments is one of the main challenges with radio-based 5G positioning. Note that the LOS path could be through a physical object, such as a window, if the power is large enough for the receiver to distinguish it from noise. In a practical sense, the first path above the noise floor has a good probability of being the LOS path, and if so, further processing determines the TOA or AOA of this path.

Figure 6 depicts the NLOS phenomenon in the spatial and time domains. A good LOS detection algorithm can reliably find the first LOS path in the channel impulse response (CIR). Note that in multi-antenna and multi-beam systems, there will be multiple CIRs associated with each antenna and/or beam for the positioning algorithm to process.

Another challenge is the potential lack of LOS links, both in outdoor and indoor environments. In areas where the deployment is designed for data traffic capacity, there may be LOS to sufficiently many base stations to perform TDOA-based positioning. On the other hand, in areas with less dense deployments, positioning solutions that combine angular and range measurements are typically more robust, as LOS is required only between a single base station and the device. 

Another potential challenge for 5G positioning is the synchronization between base stations used for TDOA measurements. The TDOA method relies on precise time synchronization between base station antennas to cancel out the device clock jitter in the uplink transmission time. While 5G networks can easily cope with this jitter (in the order of hundreds of nanoseconds) for connectivity, positioning requires synchronization in the order of 1 nanosecond. This method is well-suited for simple non-AAS radios.

The following table provides a high-level overview of the advantages and disadvantages of different positioning technologies.

a. Does not work indoors, underground, tunnels, urban canyons, and so on.

b. Challenging to support over large area, as the environment must be pre-mapped and connected to reference data with precise location information.

c. All positioning technologies that require a separate infrastructure lead to increased capex and opex, due to the need to deploy separate infrastructure, as well as its life cycle management.

d. The positioning service only has local coverage. There is no general outdoor solution for wide-area coverage.

Some use cases require highly accurate and reliable positioning, such as geofencing for improved safety when humans are moving close to autonomous machines, and autonomous guided vehicles moving on a factory floor or in a warehouse. For such use cases, cellular positioning alone is often not the most accurate and cost-efficient solution. However, as 5G positioning can easily be integrated with other technologies, the accuracy of such a solution can be improved by augmenting it with additional sensors, such as motion sensors, lidars, or cameras on the device, floor maps of the factory, hospital, or mine, or external cameras overlooking a parking garage.

It's worth adding that these sensors have certain limitations, such as motion sensor drift, bad lighting, and indistinguishable features in the visual environment. Cellular positioning can help mitigate these limitations, although this kind of positioning has its own challenges that significantly differ from the ones listed above. As such, the combination of cellular positioning and additional sensors provides an accurate and robust solution.

The real-time data from these sensors can be transported with a cellular connection using a global time reference, which can be combined with the cellular positioning estimates in a network node. Integrated solutions, combining cellular positioning with additional sensors, called sensor fusion, also benefit from different technologies that complement each other. For example, visual simultaneous localization and mapping (SLAM) does not work well in an environment with few visual features, while cellular positioning does, but may instead have difficulties in cluttered environments.

When augmenting cellular positioning with additional sensors, understanding how reliable the different information sources are to be able to combine them in a good way is paramount. An error ellipsoid added to each positioning measurement can be used to indicate the integrity of the measurement to the application, considering factors such as the approximate distance to surrounding base station(s), the bandwidth used, and whether the device is within LOS of the base station(s).

The 5G system can also improve GNSS performance by providing the correct information to the device. This solution is called GNSS-RTK [10].

Ericsson recommends a 5G-based positioning solution for the following reasons:

• It is an integrated positioning and connectivity solution that also utilizes the 5G connectivity infrastructure for positioning and removes the need for separate infrastructures.
• It delivers similar positioning performance for different types of 5G devices.
• It provides a secure and reliable positioning service leveraging 5G security features for identity protection, authentication, device security, and so on.
• It enables seamless indoor and outdoor positioning.
• It can easily be integrated with and augmented by sensors such as inertial measurement units and lidars.
• It gives positioning information in global coordinates.

The recommended outdoor solution is based on AAS radios and utilizes angle and range information through beam selection and distance estimates from a single base station. However, this solution may also be augmented by multiple measurements from surrounding base stations. As these settings allow locating devices using only one base station, the solution works well in deployments dimensioned for connectivity services, where LOS to multiple base stations, as required for TDOA-based positioning solutions, rarely occurs.

The recommended indoor solution is based on timing measurements, specifically uplink TDOA, from multiple base stations used for triangulation. This solution works well for base stations without AAS capabilities, which are typically used indoors, where angle information is not available. For local indoor deployments, the density is much higher than in outdoor macro networks. In addition, the number of LOS links between the device and surrounding base stations is typically sufficient for TDOA-based positioning.

While the positioning of a specific device on demand often provides the best accuracy available, it limits use cases that require the positioning of all devices in the network. Such use cases may relate to counting or listing all devices in an area or providing population density. For these instances, a better solution relies on passive or event-based positioning data available in the mobile network.

For asset tracking, autonomous robots and vehicles, and so on, where positioning is required both indoors and outdoors, the recommended positioning solution depends on the device capabilities as well as the required positioning performance. Where moderate accuracy is enough, such as within logistics, and where the device cost is crucial, 5G positioning is the best choice.

Positioning with an indoor 5G infrastructure can easily be combined with outdoor positioning using the already existing outdoor macro coverage, thereby providing a seamless indoor and outdoor positioning service. With the location management function as a single point of control in the network and key algorithms in the radio access network (RAN), software upgrades to enhance positioning performance can be done with less downtime and have an immediate effect on all devices. This is crucial as enterprise and industry scenarios may have hundreds of devices to support.

Several use cases—including autonomous mobile robots, asset tracking, three-dimensional (3D) positioning of uncrewed aerial vehicles, and smart buildings that activate air-conditioning based on the location of people in the facilities—involve devices with markedly different capabilities. This may result in significantly different and unknown positioning accuracy. Therefore, the best option is to utilize localization solutions not requiring radio-level cooperation from the devices, as well as device-agnostic positioning methods. These methods rely on uplink communication signals and measurement reports, originally designed for communication purposes, as part of 3GPP conformance testing.

Making the positioning performance agnostic to the device type also has its benefits, since computing the device position in the network ensures a uniform localization performance that is practically independent of the device’s capabilities and use cases. In addition, device-agnostic network-based positioning is future-proof, fully supporting the upcoming 3GPP 6G specifications, which will improve cellular positioning with enhanced accuracy, reliability, and lower latency.

While the underlying positioning technologies are key to realizing the use cases, exposure application program interfaces (APIs) in 3GPP and CAMARA is key. The Open Mobile Alliance Mobile Location Protocol API, which has served the location industry for more than 20 years enabling on demand & real-time location, is now being complemented by more industry 4.0 friendly APIs, such as Ngmlc. The introduction of CAMARA APIs will enable the application development community to excel in the use of safe and trusted network location data based on end-user consent. Rigid and scalable procedures for user consent management associated to exposure of location data to network applications is also an important component for assured privacy.

6G networks are expected to provide significantly better positioning capabilities and performance for connected objects through AI solutions, such as improved LOS detection and AI-based fingerprinting solutions that compare measured signals with a map of pre-recorded signal fingerprints. Such solutions will be especially important in areas without LOS to surrounding base stations.

To achieve precise positioning, 6G networks will offer advanced support for combining positioning information from various sources, including cellular position, images or video from camera sources, lidar, inertial measurement units, and GNSS assisted by the cellular network. This improved support encompasses better accessibility to positioning information and enhanced capabilities for integrating it seamlessly within the network. Additionally, 6G networks can provide up-to-date maps of the local environment, providing information about how a user or device can move, which in turn results in more accurate positioning.

One of the major changes in 6G is the introduction of integrated sensing and communication (ISAC), which not only allows for positioning non-connected objects [11] [12], but also creates new use cases, for example in the safety domain.

Digital representations of the local environment can be created by combining the positioning capabilities provided by 6G networks. With precise positioning, which can locate connected objects with high accuracy, and sensing to pinpoint non-connected objects, digital representations of, for example, a city or a shopping mall, can be generated and kept up to date by tracking buses, cars, pedestrians, as well as other assets.

Ericsson recommends 5G-based positioning solutions because 5G devices anywhere can be located by utilizing the 5G infrastructure built for mobile connectivity services. By deploying indoor 5G networks in large buildings with limited interference from outdoor macro networks, a seamless positioning service with the ability to position 5G devices both indoors and outdoors can be achieved.

5G positioning alone, based on the infrastructure built for mobile connectivity services, provides sufficient performance for most use cases and can already achieve a positioning accuracy of 20 to 50 meters outdoors and one to three meters indoors. For better positioning accuracy, there is the option to densify the 5G network. Another solution is to fuse the positioning information of 5G and sensors such as cameras, lidars, and inertial measurement units. Furthermore, AI/ML-based enhancements can also be considered to improve positioning accuracy and reliability thanks to network-centric solutions readily supporting sensor fusion and AI/ML technologies.

Positioning is a key feature of 5G standalone networks and, through the use of AI, is expected to evolve in 6G. By introducing sensing capabilities, 6G will also be able to position non-connected objects, such as drones flying too close to an airport. In conclusion, the 3GPP standard will continue to provide a framework that allows for further enhancement and innovation.