IoT positioning in LTE standardization

Positioning had its day in the sun during LTE standardization in 2016 – not only regarding emergency calls but also supporting IoT devices. Here we look at the key terms and technologies of IoT positioning when it comes to LTE.

cityscape night view

Positioning has been a topic in LTE standardization since 3GPP Release 9 – mainly to fulfill regulatory requirements for emergency call positioning. What really brought it into the limelight in Release 14, however, are the benefits provided for IoT-use cases, as illustrated below: Smart ‘things’ (such as wearables), Transport ‘things’ (asset tracking) and Sensing “things” (such as environmental monitoring):

Graphic 1


The devices involved require low-power consumption and the possibility to communicate in the most challenging locations in terms of coverage. They may not support GPS or other Global Navigation Satellite Systems (GNSS), and they might be located indoors or even deep indoors – the “thing” might be in a basement or even in a mine shaft. In general, the low power and complexity of these devices, their limited bandwidth, and the high-positioning accuracy demand expected in the use cases, evoke many challenging discussions for the research community.

What we will look at with this post is support for further-enhanced machine-type communication (feMTC) and narrowband IoT (NB-IoT) - both representing candidate solutions of Low Power Wide Area (LPWA) technology with low-power consumption and aimed for extended and enhanced coverage for low-rate MTC applications.
For an overview of IoT terms related to 3GPP releases, you can check the blog posts “Celluar IoT alphabet soup”, and “Narrowband IoT standardization soon finalized” published last year.

Basic positioning support for enhanced MTC devices was available in 3GPP Release 13:

  • Enhanced Cell Identity (ECID) is a Release 9 positioning method in which the UE can report reference signal received power and quality (RSRP/RSRQ) and Rx-Tx time difference along with the serving cell ID for positioning purpose.
  • Observed Time Difference Of Arrival (OTDOA) is also a Release 9 positioning method, in which the UE measures the downlink Positioning Reference Signal (PRS) time difference of arrivals of several neighbor cells in comparison to a reference cell. The positioning estimation is based on multilateration of these measurements, also known as Reference Signal Time Difference (RSTD).

Moving to Release 14, several positioning enhancements have been considered for but not limited to Cat-M1 and Cat-M2 devices, which are UE categories for feMTC. Release 14 enhancements include finalization of the measurement requirements and the OTDOA enhancement with more densified PRS configurations and more frequent PRS transmissions.

Denser PRS occasions in time compensates the narrower bandwidth and by allowing accumulation over time an acceptable positioning accuracy is expected. An optional frequency hopping is also standardized to add frequency diversity.

Below you will see an example of different PRS configurations including a 20MHz, 160ms periodicity LTE PRS where each positioning occasion is one sub-frame, a 5MHz PRS with occasions of two sub-frames and 80ms periodicity, and a 1.4MHz PRS with occasions of four sub-frames and 40ms periodicity.

Graphic 2


Another new OTDOA support is the possibility of multiple PRS transmission configurations aligned with legacy LTE. It can be expected that there may be UEs supporting multiple PRS configurations and hence exploiting the multiple PRS transmission configurations in favor of a higher positioning accuracy.

Note that the Release 13 NB-IoT device did not support any positioning feature aside from simple Cell Identity (CID). Therefore, both ECID and OTDOA positioning support are introduced in Release 14 NB-IoT, after discussions on the suitable positioning method candidates. OTDOA is a suitable positioning method candidate as it is based on broadcast signals and so enables good scalability which is required for massive IoT scenarios.

Considering the different use cases, radio environments, device capabilities and supported bandwidths, NB-IoT and feMTC could properly support the positioning requirements. NB-IoT is considered to be more suitable for low-complexity device use cases, while feMTC can be used for use cases that have more demanding positioning requirements and deployments in challenging radio environments.

Considering OTDOA support for even narrower band requires extensive accumulations. However, for a 200 kHz NB-IoT carrier, limited downlink radio resources are available for positioning purposes. In Release 14, a new reference signal was designed for positioning of NB-IoT, referred to as Narrowband Positioning Reference Signal (NPRS), which has synergies with PRS of LTE and feMTC.

There are three parts (types) of NPRS configuration methods defined. Part A is a bitmap of the same length as valid subframe configuration (10 or 40 bits) in which NPRS will be assigned to invalid subframes and indicated as “1” in the bitmap. See below for an illustration of an example NPRS bitmap. Part B is like the configuration of LTE PRS, and the one shown below, while the number of sub-frames containing NPRS in one NPRS occasion can be 10, 20, 40, 80, 160, 320, 640 and 1280, and the periodicity of NPRS occasion can be 160ms, 320ms, 640ms and 1280ms. The third type is called Part A + Part B, in which a sub-frame contains NPRS if indicated by both configuration types.

Graphic 3

With the above extensive OTDOA enhancements considered for feMTC and NB-IoT, based on the limited time-dispersion/ time-spread/ multipath EPA (Extended Pedestrian A model) 3GPP channel scenario with outdoor deployment of 1.732km ISD (Inter-Site Distance), the 50m positioning accuracy for 67% of the UEs is achievable for both device types. However, for the same scenario setup, feMTC is expected to reach better positioning results compared to NB-IoT due to wider bandwidth.

One of the positioning advantages of the introduced PRS configurations compared to is their coverage enhancement capabilities, which enables detecting and measuring cells in deep indoor channel conditions. The Signal to Interference and Noise Ratio (SINR) threshold requirement for both reference and neighbor cells in RSTD measurements are set to -15dB for feMTC and NB-IoT devices, while the same parameter target is -6dB for reference cell and -13dB for neighbor cells in LTE. The performance of multilateration technique used in OTDOA depends on the number of available cells for time-of-arrival estimation. The conclusion is that these devices can achieve positioning support in scenarios in which the legacy LTE had trouble finding the minimum three cells for trilateration.

Extensive standardization efforts in two different Release 14 Work Items – feMTC and NB-IoT – have enabled positioning support for these devices whilst taking their low-complexity and low-power characteristics into account. Meanwhile, the positioning solution would be futureproof and scalable in terms of massive connectivity of IoT use cases. There are already many IoT use cases that require positioning estimation as a core aspect of their performance, and among the available and future IoT solutions, positioning capability can be a feature that can differentiate the favorability of one certain IoT solution.


ABOUT THE CONTRIBUTOR
The Ericsson blog

In a world that is increasingly complex, we are on a quest for easy. At the Ericsson blog, we provide insight, news and opinion to help make complex ideas on technology, business and innovation simple. If you want to hear from us directly, please head over to our contact page.

Contact us