As 5G networks becomes a reality, they will enable data-intensive processes. But the impact of 5G technology will also be significant for security and network management.
Every 10 years or so, wireless broadband and digital cellular network standards and performance take a major leap forward – and every 10 years, the tech world’s expectations leap even further.
In 2019 and 2020, the major stride forward in wireless connectivity is 5G —the fifth-generation standard for wireless communications. 5G will follow (but not replace) current 4G networks with vastly increased capacity, lower latency, and faster speeds. The new standard also offers network-management capabilities, through network slicing, which enables multiple virtual networks within a single physical 5G network. With these multiple virtual networks, IT departments and providers can support business needs and data-intensive processes.
On paper, 5G is a quantum leap beyond 4G, with data transport speeds of as much as 20 Gbps (more than a 10x improvement), with submillisecond latency. Such speed will make real many long-held technology ambitions, including real-time augmented reality (AR) and distributed machine learning within the Internet of Things (IoT). While worldwide deployment is still at least a year away, several operators are already in test deployment.
In turn, 5G also boosts other key technology architectures that have become critical, including IoT connectivity and edge computing architecture. IoT-connected sensors are gathering and generating some of the large volumes of data that 5G is designed to accommodate. And edge computing will help support speed and lower latency, by eliminating the need for data transfer to go back and forth to centralized data centers or clouds. With edge computing architecture, data processing can occur at the edge, nearer to the users, data and devices that need it.
The benefits of 5G. With this triad of technologies working in concert, a staggering range of services and processes will develop—all reliant on edge computing.
Remote surgery, for example, becomes possible. With 5G, a human surgeon can control a surgical robot across a great distance, with immediate real-time interaction, including not just sight and sound, but haptic feedback. Remote vehicle piloting will sense speed bumps and wind pressure. And, of course, cars will be able to drive themselves, which will trigger the permanent and ongoing redesign of urban environments. Already highly sophisticated, drones will become even more finely controllable than ever.
Increased IoT sophistication enabled by 5G will improve detection systems for natural disasters. Healthcare will implement 24/7 remote patient monitoring. Students will be able to attend classes virtually, in any part of the world.
Reliability of connectivity will soar, and the dropped call will become a thing of the past. And battery life for mobile devices will get a break.
The downsides of 5G. There are tradeoffs in boosting speed and capacity. For 5G, the tradeoff is increased complexity.
The 4G network brought vast cell towers, springing up across the landscape in wide distribution; 5G, by contrast, will manifest with the addition of many smaller antennas, to boost data capacity and support improved performance in the devices that use it, and to boost the new frequencies that have a tougher time with physical barriers.
5G networks will also need to support multiple standards (Wi-Fi, LTE, LAA, etc.), especially for IoT support.
Another impact of 5G technology is that IoT sensors and data will likewise expand: Total deployment of IoT devices, according to Gartner, will hit some 38 billion devices just as 5G is deployed worldwide.
And edge computing will keep pace. In 2020, IDC predicts that edge infrastructure will account for 18% of IoT spending. As a result, 5G, IoT and edge computing are converging to bring new capabilities while generating massive amounts of data.
Other issues will arise, and it’s not clear how all of them will be addressed.
More bandwidth, less coverage. A consequence of increased bandwidth is more cells—and more cells mean a tighter coverage radius per cell. As 5G networks expand, this problem will be self-correcting, but in their early days, the no-dropped-calls promise won’t be met.
Another challenge is that part of the 5G frequency expansion hits in the 6 GHz range, which is already crowded. This could diminish real net transmission speeds, and may need to be sorted out over time.
Maintenance. More infrastructure means more maintenance. And it will be more complex maintenance, using hardware that is 10 years younger than what’s deployed for 4G today. Many parts of the U.S. (let alone the rest of the world) may not have enough qualified service personnel, early on, leading to outages as the networks go into service.
Security. A more complex digital ecosystem also means greater security challenges. Greater speed is a plus for users, but it’s also a tool for malicious attackers. Distributed denial of service (DDoS) attacks will likely increase, as 5G will boost IoT’s participation in real-time enterprise systems. And IoT is built on the old client/server model, with old security mechanisms. This will take time to adaptively correct.
The 5G network infrastructure will be largely virtualized with containerized workloads, which also widens attack surfaces. Security patching becomes mission critical. Similarly, software-defined networks will proliferate to accommodate new IoT devices: vehicles, industrial hardware, media widgets. This likewise increases the risks and will call for OEMs to secure their firmware.
Edge computing has arrived at the right moment. If 5G introduces new threats in IoT, edge computing can provide solutions.
Edge gateways. Part of the purpose of edge nodes is traffic control, to organize IoT traffic. By creating points of entry into physical networks, additional security can be layered onto IoT devices. Even if a device is hacked, a malicious actor is stopped in his tracks, getting no further than the gateway.
Edge analytics. Many network attacks are automated these days, and analytics—combined with AI—can detect and subsequently predict their activity. Anomalies in infrastructure performance can produce machine-learnable patterns that can then be watched for as possible attack points, making it easier to bolster network perimeter security.
Securing DNS. Domain name system (DNS) gets a serious workout from IoT, and DNS on the edge is yet another vulnerability. More than 90% of the malware used in network attacks exploits DNS as a protocol for intrusion. This, too, becomes a machine learning process, through the installation of a DNS security service to monitor and analyze DNS requests for suspicious request patterns—resulting in the blocking of malevolent domains and websites.
AI-powered endpoint management. Vendors have begun addressing the challenges of IoT proliferation with AI-based monitoring and management. These systems take into account the new IoT devices attaching to the network, analyze their connectivity and data traffic patterns, and offer risk exposure and vulnerability notifications.
The digital ecosystem is about to go through some profound changes. The edge and IoT will be beneficiaries of those changes, but at the cost of increased risk. The good news is that 5G isn’t here yet. There’s still time to get ready.
Scott Robinson is an enterprise architect and AI consultant with a 25-year history in business intelligence, analytics, and content management in the healthcare and logistics industries. He is currently CIO of the GlenMill Group, a research consortium providing new AI technology and infrastructure for enterprise applications and services.