By: Brian Daniels, Senior Vice President, Z5G
Driving today is chaos.
Imagine New York City at rush hour. Today, drivers and cars make decisions independent of each other. Autonomous vehicles are evolving to solve for this chaos. Ultimately, automated traffic grids where vehicles, infrastructure, and IOT devices communicate will pave the way to safe and efficient mobility solutions. Significant network demands are imminent.
The autonomy of a vehicle is rated on a scale of one to five—one relies solely on a human driver and five is fully automated. Levels two to four vary in automation capabilities from partial, conditional and high automation. Fully autonomous vehicles will be driverless cars operating on any road in any condition. The ‘driver’ of a level five autonomous vehicle will actually just be a passenger whose sole job is to enter a destination and the car will get them there quickly, safely and without human intervention.
A level five vehicle requires hundreds of on-vehicle sensors, such as cameras and radar, to operate safely. It also needs detailed maps — down to the inch — which must be continuously updated. Sensors will collect information on weather and road conditions, accidents, obstacles, pedestrian movement and objects surrounding the car. These smart cars will rely not only on vehicle sensors as wells as those embedded in road infrastructure itself for time-sensitive information for accident prevention and route planning, as well as non-time-sensitive information such as driver assistance, Infotainment, gaming, personal assistance, and user device integration. As the decision-making process moves from inside the car to one that is increasingly decentralized — communication from vehicle to vehicle, vehicle to infrastructure and vehicle to network — high capacity, low-latency network connectivity is key. This places a whole new set of demands on wireless networks.
Morgan Stanley predicts that by 2050, an automated car will produce data at a rate of 40 TB/hour. To put it in perspective, that’s equivalent to 3,300 years of iPhone usage. In terms of wireless usage, that same smart car will consume anywhere from 50 to 500 GB per month. Multiply that by the number of smart cars on the road, and it’s clear that the amount of data being produced will be staggering.
In addition to handling high volumes of data, 5G networks will need to transmit data in real-time. On the road, latency is not just a nice-to-have — it could be a matter of life and death.
How do we facilitate capacity and speed? Enter edge computing. Pushing decision making bits close to the edge is essential to making these latency-sensitive applications work. Rather than transmitting data to the cloud or a data centre, it will need to be processed at the edge of the network, closer to the end user and in real time for communications to the car. Milliseconds matter.
What’s exciting for Zayo is in the midst of spectrum changes, massive MIMO, carrier aggregation, network slicing and edge compute, fibre is still the foundation. As a global fibre provider, Zayo is focused on developing networks that will be 5-10 times as dense as today and designed with latency in mind.