GSMA IoT WebTalk Q&A: The Power of Partnership – Clear Skies Ahead for Mobile-Enabled Aviation

Chris Kucera, OneSky: Different countries have different regulations, but work like the LTE Aerial Profile is intended to standardize on an approach to remove concerns of interference with ground based users. I think when we see this being implemented by MNOs, we’ll start to see policy adapt to using LTE for drones.

Barbara Pareglio, GSMA: In Europe there are efforts from EASA to enable BVLOS in a safe and secure manner. Telco can help supporting in this area by, first of all, having a conversation with the national authorities and demonstrate the performances of the network. Also it is highly recommended to participate to the ACJA work tasks to help define the right KPIs, the more mobile operators participate the better is the result.

Chris Kucera, OneSky: I think the words “autonomous” and also being “connected” are somewhat conflicting with each other. A fully autonomous system doesn’t need connectivity. The need for connectivity arises from being able to have secure and reliable command and control links with the drone. The frequency of the availability depends on how policy and the amount of automation in the system. For a fully autonomous system, we may still need to command the drone to land in event of an emergency. Policy may suggest we do this within 1 minute, 5 minutes or even longer. The use of the link for policing could involve various use cases, like command and control, but also video for surveillance. I don’t see any technical reasons why mobile networks couldn’t be relied upon for this use case in Smart Cities.

Boris Resnick, GLONASS Union: Here is must be noted that today only remotely piloted flight is considered. Autonomous flight will be possible only in a longer-term, when all required technologies reach certain maturity. However, provided that there is a remote pilot behind, these are valid UAV applications.

Thomas Neubauer, TEOCO:Cellular connected drones will enable a range of new business opportunities within the Mobile Communication Industry. This is like IoT of the skies, applicable to many industry verticals. The value of the contribution for a Mobile Network operator depends on what the MNO wants to do. Stay with connectivity only, or deliver a wide range of high value services to this new industry.

Peter Musgrove, AT&T: Drones can be a valuable tool for providing enhanced capabilities to support the design, build, operations, and maintenance of cell sites and other network assets. Drones are also enabling innovative solutions such as airborne cell sites like AT&T’s Flying COW, that can be used to enhance network coverage, help restore communications in areas affected by natural disasters, and public safety operations.

Thomas Neubauer, TEOCO:The US will enhance that decision – moving towards BVLOS – and follow the European model that mandates Network Remote ID – as Broadcast is not sufficient for that.

Peter Musgrove, AT&T: It is true that cellular networking is not included in the near-term remote ID (RID) solution in the US due to the FAA RID broadcast rule requirement using unlicensed spectrum. In the longer term, it is certainly likely that a network RID solution will be needed for BVLOS and other applications. It is very interesting that the FAA RID rule did not prohibit network RID in licensed spectrum, so innovations are likely to occur on cellular network RID in licensed spectrum working simultaneously alongside broadcast RID using Wi-Fi or Bluetooth in unlicensed spectrum. This use of network RID will promote the benefits of connections with UTM/USS systems and support other needed capabilities such as detect and avoid, C2, and BVLOS support. A cellular network RID solution may be pursued via 3GPP R17 specs being worked currently for RID and tracking. In the near-term, ACJA will study how device-to-device Proximity Services (ProSe) originally developed for public safety and PC5/ sidelink capability originally developed for C-V2X may be applied to support UAV RID via cellular broadcast when FAA regulations may eventually allow broadcast RID in licensed spectrum. As an example of future cooperative standards work, we see that the addition of both cellular broadcast technology and cellular network RID technology (perhaps via reference to 3GPP specs) can be considered for a future revision of the ASTM F38 RID standard.

Barbara Pareglio, GSMA:At the moment in ACJA we are not covering navigation services.

Chris Kucera, OneSky: ASTM is working on a Navigation Standard as a part of other UTM standards.

Thomas Neubauer, TEOCO: First of all, 5G will come with a range of new capabilities that have a positive impact for BVLOS drone operations. However, most applications in the upcoming years will be perfectly fine with 4G. Beamforming and Massive MIMO is one of these capabilities that will further enhance the capabilities, enhancing the performance and flexibility towards Drones. Network Slicing will have an even bigger positive impact.

Thomas Neubauer, TEOCO: Cellular networks are operating in licensed frequency bands that have to be and are being protected from external interference sources, irrespectively of the vehicles being on the ground or up in the air.

Chris Kucera, OneSky: Vehicle testing should identify any interference from propulsion systems and other integrated equipment before flying operationally.

Thomas Neubauer, TEOCO: Within ACJA we make sure today that the future use of complementary communication technologies – once available and meeting the requirements in terms of bandwidth, size, weight, energy consumption and cost – can build on the same interface definitions developed today in ACJA.

Chris Kucera, OneSky: When used in conjunction with UTM, there are benefits from strategic planning. If the vehicle can fly autonomously without communications, the UTM can help ensure deconfliction from other aircraft. We should know when and where communications will be available. The SDSP concept from WT2 defines interfaces that can support thic concept of sharing coverage information.

Boris Resnick, GLONASS Union: 5GHz satellite C2 data links may complement cellular networks outside their coverage.

Chris Kucera, OneSky: As mentioned in the session, we expect cellular to become necessary as BVLOS regulations start to open up. A majority of these missions will need certified drones and we are still awaiting these. Once certified drones are available with adequate detect and avoid systems, we will start to see a sharp increase in demand for aerial cellular connectivity services. In the next 1-2 years, I would think the numbers would still be low (<10% of existing COTS drones) but increasing rapidly towards a majority being connected within the next 5-10 years.

Barbara Pareglio, GSMA: ACJA at this stage is not looking at the service such as swarming. ACJA is trying to tackle the different service in order of importance and as much as possible in accordance with the regulations. However, mobile ecosystem player can leverage and share their experience on platooning on vehicles on the ground.

Thomas Neubauer, TEOCO: We already see today that mobile network operators are working with complementary terrestrial and satcom providers to provide seamless connectivity. There are several solutions being currently investigated, tested and implemented.

Chris Kucera, OneSky: The use of UTM for strategic deconfliction can mitigate some issues with detect and avoid. The use of SDSP technology defined by WT2 in ACJA will help share network coverage so they can plan for outages, but still fly autonomously through areas where there isn’t any service available terrestrially. SatCom can also provide service in areas where terrestrial isn’t available.

Thomas Neubauer, TEOCO: The business cases are out there today. Commercial deployments are actually reality today, such as industrial inspections on critical infrastructure, power and energy distribution networks, public safety and first responders saving lives, parcel delivery (such as medical delivery, blood samples and also other), and so on. In ACJA we have more than 25 mobile network operators actively participating. They are all preparing today for the time when regulatory clearance comes – which will be like an On/Off switch. The EU member states have just agreed on a new drone regulation in Europe that will come effective with January 2023. It makes network remote ID a mandatory service. This makes it a reality in 20 months!

Peter Musgrove, AT&T: Given recent BVLOS waivers, we can likely anticipate infrastructure, agriculture, and security will move first.

Stefano Faccin, Qualcomm: At present we have not seen wider adoption of eUICC in UAVs, however eUICC are definitely a valid solution to simplify the deployment of UAVs with cellular connectivity.

Barbara Pareglio, GSMA: 3GPP has already performed some analysis on the impact and they also provided some mitigations that are possible.

Boris Resnick, GLONASS Union: There is a standard way to address mission criticality for aviation data links, first of all Operational Safety and Performance Assessment, and that will be done in cooperation with aviation SDOs, with corresponding requirements as a result. Generally, aviation comms are disjoint with any other comms, including those related to PPDR. MNOs will be responsible to handle UAS and PPDR comms simultaneously and maintain proper requirements for both.

Stefano Faccin, Qualcomm: The model leveraging multiple MNOs subscriptions in a UAV to support enhanced reliability of C2 and remote identification is being studied in WT4. Answering from a technical point of view, and not speculating on the business model, this would consider a UAV registered simultaneously to two MNO networks, but using just one network at any given time. 3GPP has developed some features for this, and is further developing others (e.g. MUSIM) for 5G in particular, which can be leveraged to deliver higher communication reliability as needed. However, one aspect to notice is that there is no definitve conclusion that having at any time an active and a stand-by MNO connections would lead to increased reliability, unless there is considerably different coverage being provided in different areas by the two MNOs; it can however help in case of different load conditions on the different operator radio networks.

Barbara Pareglio, GSMA: Location services are well standardised by 3GPP and there are several options. However, which of the location service is provided it is a choice of each individual Mobile Operator. The basic service is the Cell-ID which is the one with the lowest accuracy and it is depending on the frequency and other factors. Then there are other services that provide a better accuracy up to about 50 metres in LTE for ATDOA. For 5G the accuracy increases also due to the nature of higher frequency and smaller segments. There are also other techniques to help to increase the accuracy, but there are not standardised by 3GPP at this stage and they are proprietary. ATDOA systems are not widely deployed in Europe while they are for example in the US.

Chris Kucera, OneSky: I think the question is about “certified” drones being built with a 5G modem. My guess is that we see 4G LTE devices in the near term and redundant satcom systems before we see 5G adoption. My guess is that 5G would be 2-3 years away at the very least.

Stefano Faccin, Qualcomm: Avoiding such situations require two main solutions: ensuring a strong and enforceable solution for UAV remote identification, and coniter-UAS measures. C-UAS is outside the scope of ACJA, thus this answer will focus only on the first aspect. Ensuring Remote Identification is implemented and enforced, associated with strong security solutions that do not allow malicious UAV operators to spoof the information broadcast by a UAV, will enable simpl, fast, and reliable identification of the pilot/entities associated to a UAV, thus deterring malicious pilot/operators. However, a strong security solution must be in place and, sadly, regulators at present have not focused on the security aspects of remote identification.

Gerry Libuano, Verizon: National regulations on Broadcast Remote ID will address such problems of rogue drones by requiring drones to transmit location and identification information which will allow law enforcement to address potential threats to airports and critical infrastructures.

Gerry Libuano, Verizon: Majority of the drones will operate below 400 ft AGL but there are some drones that may operate at 800 – 2000 ft altitudes.

Gerry Libuano, Verizon: Power control of aerial UE is the most simple way to control interference to the neighboring sites. Other methods such as UE directional antenna can also be considered but has dependency on the antenna size vs frequency of operation, antenna switching to find best serving cell.

Stefano Faccin, Qualcomm: Using SIM/IMSI is actually an incorrect approach altogether. Yes, a UAV is a UE equipped with a modem and SIM card, and as such has a subscription belonging to an MNO. However, at the application level the UAV uses the services of an USS/USSP, and is assigned an aviation-level identifier that is used for a variety of purposes.

The UAV is known to the USS/USSP not based on its IMSI (which need to be maintained secret and not exposed outside a 3GPP system, exactly as in the case of IoT devices), but based on an “higher layer” identifier. Such ID is also used for Networked Remote ID, when the UAV reports information to the USS/USSP. It is further used for Broadcast Remote ID, and sent in broadcast for other devices to receive and correlate the broadcasted UAV ID to the USS/USSP serving the UAV and that can provide aviation level information (e.g. actual UAV ID, pilot ID, license ID, etc.) to the entity that received the broadcast ID. Especially for Broadcast ID, using IMSI is absolutely not feasible. Therefore, we must leverage solutions where different identifiers are used (e.g. from a basic UAV serial number, to a more complex Session ID), which are separate completely from the SIM/IMSI.

Moreover, in the lifetime of an MNO subscription for a UAV, the UAS operator may change the USS/USSP, and this a close tie between the SIM/IMSI and the identities used to identify the UAV is not an acceptable solution. Finally, strong solutions for security related to the UAV identity are needed, to ensure that the information provided by a UAV (especially in the case of broadcast remote ID) are not spoofed and can be easily validated.

Stefano Faccin, Qualcomm: Based on the investigations so far, supporting UAVs does not require a dedicated spectrum for UAV to ground communications. However, there are advantages in the use of dedicated spectrum, as several regulators are considering, to separate the UAV traffic at altitude from normal mobile devices on the ground. It is very plausible that UAVs will be using a combination of MNO spectrum (on the ground) and dedicated spectrum (e.g. in aviation band) when in flight.

Stefano Faccin, Qualcomm: The use of “dedicated” networks is surely one of the plausible solutions that could be adopted. Whether however we are talking actual private networks or a dedicated set of resources of an MNO, various options are possible. With 5G, NPNs (Non-Public Networks) and SNPNs (Standalone NPNs) are a definite possibility for certain deployment scenarios, typically more dedicated to one specific UAV mission type and role (e.g. private enterprise inspection of facilities and infrastructure). For more generic scenarios, 5G network slicing may be a tool for dedicating radio and core network resources for UAV and UAM traffic, isolating such traffic from other types of devices.

Stefano Faccin, Qualcomm: Spectrum use is the main challenge.

Chris Kucera, OneSky: I think adoption of drones and UTM is as great, if not greater. Without BVLOS drones and mandates to use UTMs/USS/USP, we won’t have a need for cellular.

Stefano Faccin, Qualcomm: I would say that the main challenges are guaranteeing the quality of C2 links, which we believe can be supported by current deployments, ensuring UAV devices do not interfere with other devices in the network (for this, appropriate radio features for aerial vehicles need to be in place; they have been standardized for LTE/4G and need to be deployed – see also the LTE aerial profile work of WT3 – and need to be standardized for NR/5G), and solid security for remote identification, in particular Broadcast Remote Identification, to ensure that UAVs can be securely identified and tracked. At present, we believe that there isn’t sufficient focus on developing a scalable, robust, and internationally acceptable solution for security, the way e.g. was done for C-V2X.

Stefano Faccin, Qualcomm: That would rely on the UTM infrastructure (North America), or the equivalent U-space infrastructure (Europe), and other similar regional infrastructures, for authorizing and monitoring UAVs.

Gerry Libuano, Verizon: ACJA Work Task 1 provided its inputs via liaisons to SA2 in support of UAS connectivity, Identifcation and Tracking in both the 5G System and the 4G Evolved Packet System. ACJA will advocate for the 5G NR enhancements in support of Aerial UEs (UAVs) as a priority feature in 3GPP for Release 18; ACJA Work Task 3 will then develop a 5G NR Aerial Profile based on Release 18 specifications, building on the LTE Aerial Profile.

Chris Kucera, OneSky: I don’t understand the issue with using cellular for maritime, but the application of SDSP could be good to help maritime plan for coverage/availability. This may already be covered with existing MNO service coverage maps as they are built for users at sea level.

Boris Resnick, GLONASS Union: Maritime comm requirements are generally a subset of aviation requirements. Inland navigation may benefit from cellular connectivity, but that is certainly out of current scope of ACJA.

Chris Kucera, OneSky: The cellular network is robust and actually, the number of additional drone customers will likely be much lower than the current number of ground mobile customers on a network. That plus the increase in capacity from 5G will probably take care of any capacity issues. What we may see is the need for additional localized services, like obstruction monitoring, interference monitoring, weather broadcast, etc. to support more complete set of aviation services that could help aviators as they fly from one airspace to another. The infrastructure provided by existing ATM equipment doesn’t readily support the drone use case at lower altitude.