Being an engineer has probably never been more interesting than it is today, with such rapid advances in technology and how this impacts both our daily life and our work at Ericsson. New devices, inventions, solutions, and tools appear every day alongside the ongoing digitization of all kinds of industries.
It’s hard for anyone to be updated, but to help, our CTO, Erik Ekudden, has just published a new article examining what he sees as the five most significant trends in technology evolution.
Inspired by his list, I have some personal reflections on technology evolution.
The 5 tech trends identified by Ericsson’s CTO:
- An adaptable technology base – Blending technologies in new ways to unleash next generation computational networks
- The dawn of true machine intelligence – Moving from cognitive machine intelligence toward augmented human intelligence
- End-to-end security and identity for IoT – A holistic approach to trust in all dimensions
- An extended-distributed IoT platform – Acceleration towards a distributed and connected IoT platform
- Overlaying reality with knowledge – Immersive communication that ties user experience to the physical world
Blending technology areas
Compared to previous communications systems, 5G will be able to exploit frequencies also in the mm-wave range, due to the huge amount of spectrum available at those frequency ranges. To exploit these frequencies, beamforming and massive number of antennas are needed to handle the increased propagation loss. At these high frequencies, very tight integration is key, and tens or hundreds of tiny antennas and RF chains need to be integrated on a small chip. This is challenging but also opens up new areas of very tight interaction between hardware and algorithm development. I expect to see a lot of activity on integration aspects and radio-resource management algorithms when the cellular industry is expanding into the mm-wave domain.
Machine intelligence is becoming a reality
Let’s focus on machine intelligence. This is a topic that has been discussed for many years, but it is only recently, with the advent of powerful computers and, maybe even more importantly, the communication possibilities enabling collection of large amounts of data from different places, that it has started to become a reality.
We have so far only seen the beginning of an evolution allowing machines to help and assist us in solving problems; problems that in many cases are not easy to define and solve with traditional algorithms. These systems can tirelessly analyze large amounts of data and present conclusions in a wide range of scenarios, spanning from assessing diseases to optimizing communication networks.
Given the huge amounts of data, the problem of moving this data between different functional units is challenging and various “memory-centric” rather than “processor-centric” architectures are being discussed. Specialized hardware for frequent but complex task are also becoming increasingly common, for example graphics co-processors utilized for image recognition in self-driving cars.
Security – a cornerstone of the IoT
Machine-type communication, in which different machines, from simple door locks to complex industrial production machines communicate, can greatly simplify our daily life as well as improve the production flexibility in factories. This is basically our vision of the connected future – anything that benefits from a connection will have one.
However, this also calls for a high degree of security. If I can remotely control my door lock at home, for example to allow a craftsman to enter my house, I definitely want a secure enough solution such that I know the right person and not a thief opens the door. Other, probably even more important examples, are remote-controlled medical devices (who wants their pacemaker to be hacked?) and traffic safety. Security for IoT devices will be a very important area and needs to be addressed from the start, not as a last-minute add-on.
Why we need standardized platforms
Machine-type communication has so far to a large extent been based on proprietary solutions for a specific use case. However, for the mass-market to truly take off, standardized platforms are required. Compare, for example, with the smartphone explosion, which were enabled by two main factors: the availability of ubiquitous communications in the form of 3G and 4G networks, and the availability of a standardized software platform in the form of Android and iOS. For machine-type communication, the connectivity aspect can be addressed by the recent eMTC and NB-IoT technologies together with the network slicing concepts starting to be deployed. Simplicity is important; devices must be truly “plug and play”, for example automatically configured through the network at power-on and associated with the appropriate network slice. If this vision can become true, all kind of industries can benefit from connectivity and be assured that the wireless connection will “just work”.
How 5G enables virtual and augmented reality
The emerging 5G networks with their high capacity and low latency will also enable virtual and augmented reality. For example, technicians working on a difficult task in the field can receive guidance by more experienced colleagues. On the more playful side, the gaming and entertainment industry is likely to pick up these trends as well, coupling the virtual world in the game to the real world outside the computers. Also in this case some form of standardized end-to-end platform is required. The communication technology is available, but the fragmentation when it comes to cameras and displays needs to be worked upon.
The trends and areas enumerated by Erik are exactly what drive the agenda for us in Research. So please explore these five trends in more depth in Erik’s new article in Ericsson Technology Review. And then we would love if you would join the conversation.