Network exposure and the case for connected drones
Until recently, hefty price tags, flight distance constraints and questionable video quality have limited the use of inspection drones in industrial use cases. However, with the new network exposure capabilities unlocked by 5G Core, service providers can overcome these limitations and expand their cellular IoT services offering for enterprises.
Just like global positioning systems (GPS), drones are no longer an exclusive military technology – they are now powerful tools for many businesses. The consumer market has also witnessed the evolution of the photography and film industry, and now drones are increasingly being used for civil applications, particularly within the agricultural, construction, insurance, oil and gas and law enforcement sectors.
This presents a new market opportunity for communication service providers – and it is too big to be ignored.
The challenge of using drones in industrial contexts
Historically the industrial use of drones has been challenging for several reasons:
- Limited range communications over unlicensed spectrums with restrictions to visual line of sight (VLOS)
- The quality of service (QoS) of real-time data transfers (e.g. video) decreases with distance
- Costly and uncontrolled security and quality for remote video transfers
- The need to support AR/VR real-time services to provide a more immersive experience
The network exposure capabilities that come with 5G Core (5GC) can address these challenges – as demonstrated in this But before we look at how it’s done, we must first understand what network exposure is.
What is network exposure?
2G, 3G and 4G signaling exchanges information in one of two categories: information exchange between two network functions or between any network function within the user equipment (UE). The challenge with IoT is that it involves multiple types of UE with different communication models – as opposed to one type of device.
This resulted in an effort to define machine type communication for radio and packet core. The visible standardization for 4G and 5G is covered by 3GPP, and defining the set of APIs for 5GC deployments – which is not like the one-size-fits-all transport network that we saw with earlier 3GPP technologies – was a major improvement. The need to optimize 4G networks for IoT traffic and protect them from the IoT-generated challenges has resulted in a hurried effort to define the enhancements for machine type communication (MTC).
Most of these enhancements provide tools for radio network optimization, but there is a visible standardization effort to define features for the Evolved Packet Core (EPC) and 5GC. These standardizations are defined as service capability exposure functions (SCEF) for EPC and network exposure functions (NEF) for 5GC. These functions enable data to be transported within signaling messages via control plane, eliminating the need for a costly user plane connection. The 4G SCEF enables non-IP data delivery (NIDD) with some functions for border security, while the 5G NEF enables small data delivery on top of its security border functionalities. The conclusion? Even though 3GPP SCEF and NEF do not seem very alike, the NEF is – in effect – a newer and better SCEF.
In a 5G connected world, the customer’s application needs to communicate with the service provider’s network functions. NEF allows service providers to handle valuable data coming from application functions (AF) in a secure way with the right allocation and utilization of resources. The network exposure function provides signaling scenarios to enable exchange of information from and to an external application function in a controlled, secure way.
The value chain for connected drones and inspection use cases
The connected drone is powered by an ecosystem of industry players, and it’s important for them to work together closely to enable new industrial drone use cases. For example, in an asset inspection use case the value chain will look like the below: where service providers offer connectivity and network exposure solutions in partnership with the drone application provider. The enterprise will use that application along with the service provider’s connectivity to offer inspection services. The enterprise can own the drones and the operation itself, or use external parties to manage the services.
In the current regulatory situation, a backup pilot that’s ready to take control must be within the line of sight of each drone. This limits the scalability of drone operations for European countries as per the latest published rules for drone operations in European cities by EASA (European Union Aviation Safety Agency). EASA has used the term “U-space” to describe the management of unmanned aircraft traffic – in this case the drone - and ensure safe interaction with other entities using the same airspace, like airplanes or helicopters.
The features already present in the mobile network would enable a single remote pilot to operate several autonomous drones at different sites, which is a much more scalable solution.
The way service providers and application providers tackle these challenges will be different depending on their role in the value chain.
The challenges for service providers
A big challenge for service providers will be exposing a suite of easy-to-use APIs that are able to address specific needs and facilitate the improvement of remote management activities for connected drones. Other challenges include:
- Secure communication between users and drones and access management
- Dynamic change of quality of service (QoS) for HQ real-time data transfer (e.g. video) when necessary
- Dynamic switch of data traffic from/to the drone (e.g. remote control) into a low latency slice when necessary
- Analytics showing the best pattern with maximum network coverage
The challenges for application providers
The main challenges for application providers involve improving the drone management system applications by using simple APIs (hiding the network complexity), and controlling the level of security and performance of the connectivity service between applications and drones. Other challenges include:
- Granting data integrity, avoiding intrusions and controlling the rights in the communications between drones and apps
- Improving the quality of the video from the drone when necessary
- Enabling optional remote control of the drones when over the Visual Line of Sight (VLOS) with AR/VR experiences
- Optimizing the flight pattern based on network coverage
How network exposure can address these challenges
Exposure service-based architecture (SBA) enables the creation of specific communication flows between NF and NEF (based on standard i/f), and focuses on exposing network capabilities though standard Network APIs.
The composition of Network APIs enables the creation of service APIs designed for specific use cases. A service API is typically a combination of the following API categories: monitoring (detect something), insight (analyze and deduct) and enforcement (take action in the network).
Service APIs can be sold to application developers offering an API management system that’s able to secure registration, payment, life-cycle management and support, etc. If the service provider doesn’t have the API management system, then it has to be added as part of this solution.
Let’s see how network exposure along with service-based architecture can overcome the challenges we outlined one by one.
The drone application needs to authorize a drone that is using 3GPP-based mobile connectivity to get access to and be controlled by the drone application. Based on that, the drone is registered for an applicable service in the drone application and uses a mobile network subscription that is enabled for drone usage by the service provider.
To enable this, both the requesting drone (including hardware and software) and the mobile subscription used needs to be authenticated. This way an attack can be prevented, and it can be ensured that the drone is using the correct network subscription.
This can be achieved with SBA via the API management and gateway capabilities on top of NEF that offers different security mechanisms. This addressed to the security challenge stated above by authenticating and authorizing the drone application to communicate with the drone – and securing data integrity during the data transport across all exposure layers.
One thing to note is that all of these security requirements are applied beyond the definition of 3GPP 33.501 to handle connected drone cases – such as authorization and authentication with API invoker based OAuth, role based access control (RBAC), service level agreements (SLA) and throttling etc.
Quality of service
Network exposure for 4G and 5G using application-driven quality of service APIs available in both SCEF and NEF solve the challenge of the quality of real-time data transfer decreasing over distances. It also requires the involvement of policy and charging rules function (PCRF) and policy and charging enforcement function (PCEF) in 4G and 5G respectively. This API handles the quality control index mechanism that is used in 3GPP LTE networks to ensure bearer traffic is allocated to an appropriate QoS.
Different bearer traffic requires different QoS and different quality control index values. This enables partners/customers to secure low network latency and suitable bandwidth for specific UEs to ensure a high-quality user experience. It provides the guaranteed network bearer within each data communication session with a better user experience on demand with higher bandwidth and low latency. The QoS API applies lower latency, jitter and maximum data burst volume, and prioritizes data sessions for a drone video streams.
Remote control of drones
The next challenge for connected drones is limited range communications over unlicensed spectrums with restrictions to Visual Line of Sight (VLOS). This results in loss of control over the drone due to low network quality and minimizes the AR/VR real-time service for immersive experiences.
This can be solved via APIs that allow switching to a low latency slice in case requests for remote control (driving) and/or for AR and VR services are made. Network slicing separates the control plane from the user plane to move user plane functionality towards the network edge.
Functions such as speed, capacity, connectivity and coverage are allocated to meet the requirements of its primary objective.
When the drone application requests to take control of the drone remotely via application-driven dynamic slice selection APIs, it switches traffic into dedicated a network slice with ultra-low latency performance for uninterrupted low-latency communication between the application and the connected drone. The local break-out is dynamically triggered from drone application towards NEF in 5G Core and activates the key network functions to enforce the policy.
Autonomous drones and network coverage
There are further challenges in connected drone use cases for which the solution is still being explored – such as controlling the drone remotely via a drone application. This requires pre-defined flight patterns based on the best available network coverage to avoid any loss of connectivity. This requires mobile connectivity related information (e.g. radio signal strength, bandwidth etc.) from mobile networks that can create a network coverage map highlighting the no-fly zones for drones where the network coverage is not good enough to avoid loss of connectivity. The key to success may be network programmability along with 3GPP 5G SBA mentioned earlier. Here is some recommended reading for network programmability.
One more interesting fact is that in addition to civil drone industry use cases, extensive research is being conducted into enhanced network services and the role of unmanned aerial vehicles (UAVs) as a core network equipment with radio and backhaul capabilities is being emphasized. What 5G is currently enabling for consumer drones is just the tip of the iceberg. The future holds enormous opportunities and what we’re seeing today indicates that soon connected drones will directly impact human lives on daily basis. The journey from luxury to necessity will a short one in this case.
Want to understand more about Ericsson cloud core exposure?