2020 via time machine: networks and systems

Edit Note : This is the second of a two-part series on this event. The first post can be read here.

Last week’s IEEE Technology Time Machine Symposium brought together leading academics, engineers, executives, and government officials from around the world, to engage in presentations and dialogue regarding the evolution of technology over the next decade. In yesterday’s post, I reviewed some of the insights regarding devices and technologies. Today, we’ll address networks and larger-scale systems such as smart grids.

Networks

Wireless continues to get better, with more bandwidth in more places. LTE offers a several-fold improvement over HSPA or HSPA+ in data rates as well as a ten-fold reduction in latency. However, operator executives such as Telstra CTO Dr. Hugh Bradlow and NTT Docomo VP Dr. Minoru Etoh pointed out that peak bandwidth is what customers tend to focus on, but total network capacity is the main challenge in providing an excellent customer experience: it’s nice to own a Lamborghini, but won’t get your there any faster at rush hour.

A single user may easily be able to get an HD video stream wirelessly over LTE, but Telstra studies have shown that only a few dozen end-users of a base station can do that simultaneously: due to LTE sector throughput limitations under good conditions. Moreover, this is not some temporary technology glitch or fault of underinvestment, but a challenging limit due to usable frequencies based on radio signal propagation characteristics and information theory.

This insight calls into question the approach being proposed where millions or billions of very “dumb” thin-client devices are wirelessly linked to entertainment and intelligence in the cloud. For devices, networks, and the cloud to function effectively at scale, smart trade-offs will need to be made in real time between when to render computationally challenging scenes in the cloud and send the scene over the network, and when to send raw information over the network for local rendering. Operators will also have to figure out how to manage priorities across users. Consequently, Bradlow sees intelligent traffic management as the key challenge of emerging wireless networks.

On the other hand, putting more cells with a smaller coverage radius could work, such as femtocells or even using Wi-Fi, but this means more wired networks to backhaul the traffic, reducing some of the ease-of-deployment benefits of wireless networks. Prof. Hequan Wu, former vice-chairman of the Chinese Academy of Engineering, pointed out that China Unicom’s mobile data traffic had recently grown 62 percent. In a single quarter. China Mobile’s grew 10 times its previous rate — in a single year. Mobile video is the challenge to solve, everything else is just rounding error.

For the Shanghai World Expo 2010, 10,000 mobile video cameras were installed on trucks and buses for security purposes. Consider the future hurdle of every passenger car sending or receiving several mobile video streams, as each passenger streams an on-demand movie or participates in a video conference. Moreover, Wu said that in some large Chinese cities, densities were up to 140,000 users per square kilometer.

HP’s Dr. Peter Hartwell dove into sensor networks, pointing out that the Internet of Things becomes really useful when dynamic data is used for real-time decision-making. A broad variety of sensors (temperature, humidity, power use) will all be integrated with processing and wireless connectivity into a single chip, enabling new applications, ranging from eHealth to smart grids. Hartwell pointed out that, as with cell phones, size will shrink and features will multiply. A variety of technologies are being incorporated here, such as “energy harvesting” to scavenge power from the environment, e.g., from vibration.

Putting it all together.

While the status and trends of many point technologies was addressed in depth, the consistent theme running through the conference might be said to be a more intelligent world, based on more information from more devices being used in real time to optimize the human experience while enhancing efficiency and environmental sustainability. Nokia Services EVP Dr. Tero Ojanpera stated that anonymized, real-time data from mobile devices is the “ultimate collective intelligence,” enabling everything from optical vehicular traffic routing to locating a popular restaurant to creation of accurate maps. George Arnold, national coordinator for smart grid interoperability at NIST, observed that smart grid efforts will turn the traditional approach of building capacity to meet demand on its head, focusing instead on shaping demand to fit within capacity. Light bulbs with embedded Wi-Fi chips will be able to not only report that they are on, but reduce their output when the power infrastructure is stressed. And, the smart grid approach — which Arnold observed is the use of IT and communications technologies for utilities — isn’t limited to just electricity, but is also being applied to other utilities such as water and natural gas.

Telecom Italia’s Roberto Saracco described the potential of a “mirror world,” a virtual world that is not just a play world in another galaxy or for social networking, but an exact virtual duplicate of the real-world populated with data from a variety of sensors. Such an environment might mean that by the time this conference is held in 2020, everyone will have the opportunity for face-to-face interaction—without having to physically travel to Hong Kong, or wherever, and the ability to shake hands—remotely, via haptic interfaces.

A hopeful, positive attitude was pervasive at the event, with NEC’s President Dr. Nobuhiro Endo outlining a strategy and future which is “friendlier to people, and friendlier to the planet.” Pervasive information and insight, emerging technologies, crowd-sourced intelligence, cloud-based global optimization, greener approaches, reduced power, and ease of use through natural interfaces may yet help solve some of the problems facing the world today.

Joe Weinman leads Communications, Media, and Entertainment Industry Solutions for Hewlett-Packard. The views expressed herein are his own.

Related content from GigaOM Pro (subscription req’d):