6G Vision for Public Safety
15.09.2020We recently joined the 6G Flagship Ecosystem and you might be wondering why. 6G is far away. Most markets are only testing the preceding generation 5G solutions. Thus, there are very few concrete projects to run. Why did we want to be a part of the ecosystem?
Working with networks, all kinds of connections and collaboration is like a second nature to us. We want to be a part of envisioning the future of wireless connectivity. Further, this ecosystem and the research is championed by the university of Oulu also based here in Finland making it very easy for us to collaborate.
Lately we have run into publications presenting a vision for 6G and we think this will be a good way to kick off our participation. Since we have the most experience from public safety, we decided to focus on thinking what 6G could bring about in that arena. In a sense, we are trying to see where the current trends in our home market can lead to, when technology is radically improved. In the end we identified three core areas: Virtual Presence, Autonomous Systems and Computational Fog.
Virtual Presence
Medical and other expert knowledge has only increased in specialization over the decades. We see that this trend will continue. What it has already meant for, especially the emergency services, is the need to reach back to the expertise of specialists that cannot arrive at the scene of the emergency. In the first generation this meant calling for advice over voice radio and lately it has meant streaming video and vital signs to ER doctors waiting for the ambulance to arrive. Some of our customers are even running 360-degrees cameras inside ambulances to provide a full view of the patient to the hospital and enable the doctors to remotely direct the procedures performed by the paramedics.
As we extrapolate this trend and remove the technical barriers especially in data transmission rates, we believe that two things will happen. First, full digital twins will be built of patients even from an ambulance. An enormous amount of data will be measured and transmitted to enable a medical simulation inside a hospital that allows for testing treatments virtually before they are applied in real life.
Second, teams of experts will be able to utilize virtual reality to enter the scene of the emergency. In some cases, for example, in dangerous situations, there will not even be a single human onsite. This may seem like not too hard a challenge even for current technology, but the way we see this happening with 6G is that there is a team of experts from around the world collaborating in the same environment. Not only will they need to see the scene in real time, but they need to be able to also manipulate each other and common artifacts as part of their work. Similarly, the persons onsite will utilize Augmented Reality to see and interact with the virtual participants. Keeping a full team of persons all in sync with the reality will be a tremendous challenge for future mobile networks.
Autonomous and Remote Systems
A common misconception is that autonomous cars and other machines would not require connectivity. Even though many of the activities are done on the device, there is a considerable need for highly reliable redundant connectivity. In the public safety domain, autonomous systems are most commonly deployed in search and rescue type of missions with flying drones leading the way.
We believe that 6G latency and throughput will make it possible to build many of such systems in away that all of the processing is done on the backend. This will make running some machine learning applications a lot easier and most importantly limit the cost of the autonomous unit even to a point where the unit may be disposable. Thus, large swarms of autonomous units will be able to enter, for example, forest fires without the need to be recovered. However, this will not only put a strain on the data rate, but on the network architecture. In such emergencies that cover larger geographical areas, the range from a base station will require the network to be extendable by emergency vehicles and the drones themselves in a mesh networking fashion. Increasing the range from the base station directly will hit the limits of physics unbreakable by technical innovation.
Far ahead of this solution, we will however see cases where even the most precise systems are remotely operated over wireless connectivity. This is possible already for most systems with the upcoming 5G connections as long as the network reliability risk has been taken care of. The use cases range from flying drones indoors at high speeds to remotely operated medical procedures.
Computational Fog
Perhaps the biggest change anticipated with 6G will however be a horizontal enabler what we are here calling computational fog. This network will have many elements, including vehicle routers, that are able to offer generic computing capacity for calculations that are not related to the functioning of the network itself. This alone is not new, but commonly referred to as edge computing. However, as currently the discussions pit distributed edge devices and their computing power against the centralized cloud and single calculations are allocated to the desired locations during application architecture design, this will change completely.
With 6G it is possible that the compute capacity will be the limiting factor instead of the network, which is almost unheard of historically in mobile use cases. Load balancing will need to be done on compute tasks, so that individual computation will search the network fog to look for a suitable platform that is able to offer the computational muscle in time for the need. An application of a user in a car at a parking lot calls out “I need approximately 100MFLOPS to be calculated in the next 100ms, who’ll pick it up?” and a near-by IoT-device picks up the tasks without the devices ever before knowing of each other. In a public network this might require, for example, a blockchain based ledger and market to allocate the costs of the calculations fairly, but network operators and public safety officials carry an advantage in being able to optimize computation within their own fog.
The optimization is going to be, however, a lot more demanding than just the deadline and the size of the task to be completed in terms of processing operations and the differences in latency. As the tasks get larger, the network will also need to consider how these parameters may change during the execution as the participants may be moving at high speeds. Likewise, the optimization will need to consider the availability of data. For many artificial intelligence tasks, the classic case of having the most detailed case specific data available locally needs to be contrasted with having the broadest view in the backend. While public safety users will benefit from this “AI capacity provided by the network”, it is amazing to thing how much it will also require “AI capacity to direct the network”.
To get to a computational fog, 6G will need to offer a practically fully connected mesh network between all users, which in itself will open up a lot of possibilities. For example, giving the public safety use a priority in the network, will mean that there are always countless of routes to get through, as long as you can physically connect to all of the available networks. This will likely also create physical spaces that are fully wireless connected. For example, buildings that are not just intelligent with installed automation, but where every user can communicate with everything. A police officer or a fire fighter will have an amazing understanding of the building or port area they enter as soon as they connect with the network.
We eagerly look forward in participating the creation of these and other unforeseen innovations with out partners and customers!
Our vision relies on the work of others:
https://arxiv.org/ftp/arxiv/papers/2004/2004.14850.pdf