OK, it is now time to consider a “real” wireless network!
Smartphones, complex IoTs or aggregates of IoTs (like vehicles), have been addressed to as “terminals”. In fact, it is now a few years that in several occasions this name no longer apples. Think at the tethering supported by your smartphone, thinks at the WiFi network in recent cars or at Alexa, AppleTV, … All of them are devices that doubles up as Network Node. You find the basic functions of routing, channel selection, data storage, processing that we are used to see in network nodes. Your smartphones act as a network node and access gateway for your computer, your vehicle WiFi is a (local) network supporting devices like your phone seen as a server, so that it can provide access to the data it stores as well as to some of the applications that were downloaded on it. Alexa connects and orchestrates a variety of IoT in your home (Alexa, switch on the light…).
This transition from being a “terminal” to becoming a network node is a crucial and disrupting evolution paving the way to 6G, and it is already here!
Actually, in the 5G architecture (not yet implemented, although we start to see the first signs in the “private 5G”) there is the possibility of handing off the session control to the device (that at this point is no longer a “terminal” but a network node). It means that the device -your smartphone- can request and negotiate network resources to set up a communication session and can change these resources and their aggregation through the session as new needs arise. This is not possible in the 4G architecture but it becomes possible in the 5G (the examples I gave before are possible using WiFi, using the Operator’s network it will not be possible, this is the reason why my car, equipped with AppleCar requires me to switch on the WiFi on the iPhone to manage the iPhone as a server).
The shift from “terminal” to “network node” kills the last hierarchy remaining in the telecommunications network. A smartphone that is a network node can use SDN (Software Defined Networking) to set up the chain/pool of resources required. Great from the application point of view, the final nightmare for the Telcos that have already seen their networks become dumb pipes and now might find out they will no longer provide connectivity services (the only ones left), rather becoming a commodity warehouse of bunches of network equipment on shelves that can be picked up on demand by (application) users and that of course will be in competition with other third parties network equipment (included those provided via NFV). Network functions like authentication, end to end security, store and forward,… can (and are starting to) drift away from the network (soft SIM, biometric identification,..) and be provided by third parties.
Of course, once we shift the focus from the network edge (base stations, poles, antennas…) to the “terminals” we are stepping into a real wireless network one that moves around on wheels (and legs)!
Moving around it requires a completely different approach to the management of communications resources. This is the area of the mesh networks, well explored in the military sector where the goal is to create a reliable, self-configurable network out of thousands of devices (that may be parachuted from above, as an example). This is an area where Small World theory, autonomous systems and the underlying Artificial Intelligence are leading the evolution (and there is plenty of research needed). These three areas also requires plenty of processing power and low power communications to sustain massively distributed processing.
Smartphones, vehicles, IoTs will become network nodes that can create a network bottom up (eventually, when needed connecting to the big pipes and reserving the resources required). Incidentally, this is the way Nature creates communications infrastructures (among a pride of lions, a flock of sheep and a swarm of bees!)
This will become a technical possibility (as mentioned, on a limited scale, it is already here) and, even more important, it will be a feasible approach to 6G from an economic standpoint, shifting the investment from Telcos to the edges (and to consumer electronics, i.e. to each of us).
An interesting twist brought by a network that stems from the edges and with terminals that double up into network node is the change of the traffic capacity paradigm. In today’s networks, the more terminals are connected the lower the capacity available to each of them (the capacity depends on the base station/frequency and spectrum available and this capacity has to be divided by the number of users at a given time). In a 6G network it is the other way round. Every additional device that can act as network node increases the overall capacity, so the more users the more capacity!