Mobile networks are uniquely positioned to improve the safety of drone operations and also to enable beyond visual line-of-sight applications. Our particular interest is in low altitude drones, which fly from around 200-400 feet up to 1,000 feet, because this is where existing terrestrial mobile networks are able to provide coverage. We are focusing our work on two aspects:
- Radio connectivity for drones in order to understand what the issues are, and how we can solve them.
- How we can support Unmanned Aircraft System (UAS), Traffic Management (UTM) or simply drone traffic management with mobile network services.
Watch the video for an overview of our work:
In this post, we will explore radio connectivity.
Mobile connectivity aids drone operations
Mobile connectivity can be added to drone operations in two primary ways. The baseline is to use a remote controller over a direct radio link to control the drone.
The first way is to add mobile connectivity to the remote controller, or simply attach a mobile phone to the controller, but still use a direct link to the drone to control its flight. This way the airspace awareness of the pilot can be improved as real-time airspace information, airspace restrictions, and alarms and warnings can be shown to the pilot during the operation.
The second method is to connect the drone to the mobile network and use mobile connectivity for command and control. This improves safety because all the real-time information from the drone can be sent over the network to the drone traffic management system. This also enables beyond visual line-of-sight operation because the whole network can be utilized to control the drone.
Next, we look into the details of the second scenario, i.e., when the drone is connected and controlled over the mobile network.
Three challenges in radio connectivity for drones
There are several aspects that make working on radio connectivity for drones both interesting and challenging. The first aspect is coverage. Existing mobile networks are optimized for terrestrial broadband communications. This means that base-station antennas are tilted down to reduce inter-cell interference while optimizing terrestrial coverage. Hence drones in the sky are served by the side-lobes of the base-station antennas. So, the first natural question is, whether the coverage is then sufficient in the sky?
The second aspect is related to interference. In the sky, the radio propagation is closer to free-space propagation. Therefore drones generate more interference to the networks in the uplink, and they also experience more interference in the downlink. This calls for interference mitigation techniques.
The third aspect is related to mobility support. The cell association pattern in the sky is quite different from the cell association pattern on the ground. This naturally raises the question of how we optimize mobility support for drones in the sky.
In the figure below, we share some of our simulation results. We consider a rural macro scenario, in which we have 10 MHz LTE bandwidth at 700 MHz carrier frequency. Each base station has two cross-polarized antennas at a height of 35m with 6 degrees of down-tilt. The base-station antenna pattern used in the simulation is illustrated in the upper-right figure here.
The bottom left figure shows the distributions of the downlink coupling gain at three different heights. One is at 1.5m, which is ground level. The second is at 40m, which is 5m higher than the base-station antenna height. And the last is at 120m, which is the current FAA limit for drone flying. So we can see at both 40m and 120m that the five percentile downlink coupling gain is larger than its counterpart on the ground level. This implies that the more favorable propagation in the sky can make up for the antenna gain reduction.
The figure on the bottom right shows the downlink SINR distribution at the three different heights. From this figure we can see the downlink SINR distributions at both 40m and 120m are statistically worse than their counterpart at ground level. This implies that the signal quality is worse in the sky, and we may need some interference mitigation techniques in order to fly drones.
We recently had the opportunity to demonstrate and provide an overview of our work at the Ericsson Startup Day in Silicon Valley, and to discuss our findings with industry experts on a panel at the IEEE Vehicular Technology Conference in Toronto. You can access the slides of our positioning talk at IEEE VTC 2017.
To better understand the potential of LTE for drone communication, we also submitted a 3GPP study item proposal together with DoCoMo in March. The study item is supported by 35 companies including major network vendors, major UE vendors, and operators worldwide. You can read more about the 3GPP technical study report here: 3GPP TR 36.777, Enhanced LTE support for aerial vehicles.
You can also read more about:
- The radio network aspects in our recent paper: “The Sky Is Not the Limit: LTE for Unmanned Aerial Vehicles”.
- An invited talk on LTE connected drones at Texas Wireless Summit 2017.
- How Ericsson explains performance of drones on LTE networks.