Connecting things in the city

Simulation of a realistic large-scale IoT service scenario in a city showed that up to 99 percent of devices located deep indoors could be reached with new cellular IoT technologies.

Cellular networks are well-suited to providing connectivity for emerging IoT applications due to their ubiquitous deployments, as well as their inherent characteristics, which include security and reliability. Currently, cellular networks’ main role is to provide mobile broadband coverage. Connectivity for IoT devices poses new coverage challenges for various use cases. The 3GPP Low-Power Wide-Area (LPWA) cellular technologies, Cat-M1 and NB-IoT, supporting a wide range of low cost devices, can be deployed on existing LTE (4G) networks to overcome these challenges. Both technologies are currently being deployed worldwide, enabling a range of new IoT services.

Cat-M1 is designed to support a wide range of IoT applications, including connected waste bins, alarms incorporating emergency voice assistance and fleet management. Cat-M1 provides theoretical peak uplink data throughput of around 1Mbps. However, there is a compromise between data throughput and coverage: the lower the bitrate the application requires, the further the coverage is extended for the application. The minimum connectivity target has been set to a maximum coupling loss (MCL)¹ of 160dB where the achievable uplink data rate is around 1Kbps.² This can be compared to an MCL of 144dB for broadband LTE with up to 1Mbps in downlink and a few 10s of Kbps in uplink.

NB-IoT is a narrowband solution designed to provide even better coverage and enables deployment of devices with an even lower cost than Cat-M1. It targets ultra-low-throughput IoT applications, such as smoke detectors and utility meters. The minimum connectivity target has been set to an MCL of 164dB where the achievable uplink data rate is around 300–400bps.³ Both technologies support the IoT use cases exemplified in the figure below.

Network coverage for IoT applications with limited demands on throughput, such as metering and monitoring use cases in a metropolitan area, was analyzed. Measurements from a commercially deployed LTE network⁵ for broadband services were used to calibrate a model for simulating broadband LTE, Cat-M1 and NB-IoT coverage. A three-dimensional model of a city was used, with close to 1,000 buildings per square kilometer with an average of 5 floors per building. Both line-of-sight and non-line-of-sight characteristics, including outdoor-to-indoor and indoor radio propagation models, were considered. Typical radio base station site characteristics were assumed, with intersite distances of approximately 500 meters.

IoT devices, with a density of around 20,000 per square kilometer, were uniformly distributed across the city, both outdoors and indoors, and corresponding signal strength attributed to the different environments. For example, basements located partly underground were modelled with an additional path loss⁶ of 5dB in addition to the signal attenuation indoors (10–30dB) and those fully underground (deep indoors) with 20dB.

The coverage was simulated for an IoT application on broadband LTE, Cat-M1 and NB-IoT. The same cell layout was used to calculate coverage for each technology. Network coverage was analyzed in two frequency bands: one lower band (800MHz) that has the advantage of stronger signal propagation for further coverage, and one higher band (2.6GHz) offering greater capacity. The table above shows the percentage of devices reached for each technology.

The 800MHz band modelling showed that in challenging radio signal propagation environments, such as deep indoors, both Cat-M1 and NB-IoT can reach up to 99 percent of devices. This can be compared to broadband LTE, which would reach 77 percent of mobile broadband devices. In the 2.6GHz capacity band, the coverage of both Cat-M1 and NB-IoT is also substantially better than the broadband LTE coverage of only 32 percent.

Coverage is enhanced for low-data-rate IoT devices by reducing the data rate to provide additional coverage. With 3GPP targets already exceeded in evaluations, the enhancements will enable IoT city deployments with up to 99 percent coverage of devices using cellular networks for connectivity.

Note: Update of an article published in Ericsson Mobility Report, June 2017

¹ Maximum coupling loss (MCL): Coupling loss is a measure of the attenuation of the radio signal between the transmitter and receiver. MCL is the largest attenuation the system can support with a defined level of service. This can also be used to define the coverage of the service

² An MCL of 159.7dB is a 3GPP target that has been evaluated and exceeded by the industry. See also industry white paper “Coverage Analysis of LTE-M Category-M1, Version 1.0, January 2017”

³ An MCL of 164dB is a 3GPP target that has been evaluated and exceeded by the industry

⁴ In this city scenario, only IoT services with limited demands on throughput, such as metering and monitoring use cases, are included. IoT services with stringent requirements on availability, reliability, delay and higher demand on throughput, for example traffic safety, automated vehicles and industrial applications, are also expected to be deployed in city environments

⁵ Mobile network of a major European operator in a metropolitan area

⁶ Path loss is the signal decrease that occurs as the radio waves travel through the air or through obstacles