This does not happen every day. We are living the design and implementation of a new technology that is called to revolutionize mobile communications as we know them until now.
The principle of industrial design that dictates that "design follows function" could be applied to the design of this new technology and the subsequent analysis of requirements, although contrary to what usually happens in industrial design, the use cases for the 5G are not yet fully defined and that could be an issue.
The success or not of 5G technology may depend on the clear definition and compliance with the objectives defined by the use cases. In case the use objectives for which the technology is designed are not met, we would live an evolution more than a technological revolution, a prevalence of design over functionality.
Until the mid 20s it is unlikely that extensive 5G deployments will be seen in production, so until then, what we will be living will be a process of technological evolution similar to the technological leaps experienced previously.
Next we will make a brief summary of use cases, design objectives, and 5G requirements, focusing only on a small number of aspects related to functional characteristics and technology design. As we will see below, these design objectives impose a set of technological requirements to be implemented.
Finally, we will see a small practical case of application of this new technology.
Potential use cases of 5G technology can be classified into three large groups according to the requirements established by the new generation applications:
- More bandwidth in mobile access.
- Lower latency and high reliability in communications.
- Mass machine to machine communications.
New applications may require one or multiple of the above characteristics, so the classification is flexible. The ITU-T has established an orientation classification of the types of applications depending on the functionality demanded.
The 5G New Radio (NR) specification in Standalone operation mode (5G end-to-end) includes a new set of specifications for the 5G Network Core that goes beyond what is defined in the Non-Standalone specification (4G integration). The Non-Standalone specification allows 5G NR deployments using existing 4G systems, and therefore, has its limitations and restrictions.
5G NR in Standalone mode includes the following service specifications:
- enhanced Mobile Broadband Service (eMBB): applications that require large amounts of mobile bandwidth, such as broadcasting UHD video, for example.
- Ultra-Reliable Low-Latency Communications (uRLLC): applications in real time, such as autonomous driving or telemedicine.
- massive Machine Type Communications (mMTC): applications requiring massive connectivity between machines, being able to suppose a real alternative to WiFi connectivity. It would apply, for example, to the entire IoT world (agriculture, vehicles, smart city).
Applications that are part of critical functions will require, in addition to one or more of the above functionalities, a wide 5G deployment. That is why some applications can not be implemented in an initial 5G phase and could be delayed until the 5G coverage and implementation is broad enough. This may apply to existing applications as well as to others that we don’t know yet.
The 5G design objectives can be divided into functional and structural objectives. Both are related, although it is possible to establish clear differences between them.
The functional objectives correspond to those functions that the technology will perform, analyzing the fulfillment of the design objectives initially marked. The functional objectives would be latency, throughput, the number of simultaneous connections to the network or aspects related to mobility.
The structural objectives go far beyond the functional ones, since it corresponds to the analysis of each of the parts that make up the system and how they relate to each other to enable compliance with the functional objectives. Architecture aspects, softwarization, and network slicing fall into this category.
The functional design objectives are sometimes conflicting, which represents an additional challenge in the design of the system as a whole. For example, throughput and number of connections per unit area or latency and mobility at high speed.
Based on the use cases described above, we will briefly analyze the requirements of 5G from a service point of view.
Enhanced mobile broadband services (eMBB)
- Flexibility and resilience to support ultra high bandwidth (UHB) services
- Simplification in the management mechanism of sessions and carriers.
- Most efficient multicast traffic transport methods.
- Distribution of network functions to the end of the network (network edge).
- Network slicing.
Massive machine type communication-based services (mMTC)
- Support a massive number of IoT devices in an efficient way.
- Minimize congestion caused by a massive number of simultaneously connected devices.
- Maintenance of end-to-end QoS even when the number of devices connected to the network simultaneously is very high.
- Management of various traffic patterns efficiently
- Unicast, multicast, anycast.
- Traffic to low volume bursts generated by a massive number of devices.
Ultra-reliable and low latency communication-based services (uRLLC)
- Very low latency and high reliability in data transport
- Improved service levels with very low latency.
- Improvement of mobile signaling with the aim of improving said latency (for the drastic reduction of end-to-end latency the complete deployment of the core 5G is required).
The Autonomous Vehicle
The potential of 5G technology as an enabler in the development of the autonomous vehicle (AV) is still unknown. However, what can be deduced are the effects that the support of communications between the vehicle and the network (V2N) could have on current communications networks, something that would mean multiplying up to x100 the current levels of mobile traffic. Nowadays, the current infrastructures are not designed for handling such amount of traffic.
Depending on the degree of autonomy of the vehicle, the needs for data communications vary, both in volume and nature.
Experts have defined up to 5 levels of autonomous driving. Each of these levels describes the extent to which the vehicle assumes the driver's tasks and responsibilities, as well as the interaction between the vehicle and the driver.
- Level 0: there is no autonomous driving. The driver controls all aspects of driving, with total lack of help from the system. Nowadays this could be considered “legacy” driving.
- Level 1: assistance to the driver. The systems provide information to the driver, but never takes control of the vehicle (notice involuntary lane change).
- Level 2: partially automated driving. The vehicle can take control in certain situations, but the driver is responsible for driving the vehicle (autonomous parking, fatigue detectors). Driver needs to be aware of what the vehicle is doing in every moment.
- Level 3: highly automated driving. In certain situations, the driver may neglect driving for some periods of time and focus in some other activity like reading or office work.
- Level 4: complete automated driving. The vehicle takes control and drives autonomously most of the time, but the driver must be able to take control if they decide to do so.
- Level 5: autonomous driving. The vehicle assumes all driving functions of the vehicle. No driver is necessary, all occupants of the vehicle are considered passengers.
The levels of autonomous driving (3 to 5) are still in the testing phase today, and although there are already commercial models that incorporate some of the features described above, their operating model is based primarily on on-board processing of the vehicle itself.
Market Share Evolution
It is expected that the market penetration rate of autonomous vehicles (levels 3 to 5) will exceed 50% from the year 2035. Leaving aside the cultural and/or ethical reasons implied by the concept of autonomous driving, the causes of such a slow incorporation in the market of this new technology could be due, among other causes, to the following:
- Lack of bandwidth required for V2N communications.
- Support for the connectivity of a huge number of vehicles.
- Ability to react in real time to events that occur while driving.
The communication needs of the autonomous vehicle can be classified into:
- Communications between the vehicle and the infrastructure (V2I).
- Communications between vehicles (V2V).
- Communications between the vehicle and the network (V2N).
Of the three, the one that concerns us here is the communication between the vehicle and the network: V2N. The purpose of this type of communication is to connect the vehicle to the network to provide, for example, data to the navigation system with real-time traffic status information, vehicle software updates, access to multimedia content, or access to processing systems of data in the cloud for those tasks related to autonomous driving that can not be solved by means of V2I or V2V communications.
V2N communications need, to a greater or lesser extent, each of the requirements described above: large amount of bandwidth per vehicle, the possibility of connecting a large number of vehicles to the network (especially in areas with a high population density), and a minimum latency to be able to perform autonomous driving tasks in those cases in which communications between vehicles (V2V) or between vehicle and infrastructure (V2I) are not applicable.
We are in an initial phase of 5G, and the future is still too uncertain. The current applications can benefit without a doubt from the technological evolution to 5G, but the most interesting is still to come, because historically technology has always acted as a facilitator for new applications to emerge.
The success of the 5G deployment will depend on the applications that will be developed in the future, but the future development of applications will also depend on the 5G design and implementation being a success and the requirements being fulfilled.
Until the fully automated car takes over, drive carefully.