By Keith O’Brien, Chief Technology Officer for the Service Provider line of business at Palo Alto Networks
With each successive generation, advances in mobile technology have trained us to expect ever-faster mobile speeds and the ability of the signal to transport ever-greater loads of data.
Increased data transfer rates enabled 3G to handle larger capacities, and that generation was the first to have serious broadband capabilities. As 4G LTE rolled out, mobile signals could now support interactive multimedia, voice, and video with greater speed and efficiency.
Therefore, it’s not surprising that 5G is characterized mostly by its ability to download very large data files in the blink of an eye and to carry broader sets of data, such as HD video streaming, virtual reality applications, and massive IoT implementations.
But we’re missing a big part of the picture by looking at 5G as just a faster, stronger horse in the race. It’s not just what it does but how it does it. The software-defined architecture of 5G, including 5G security, brings forward use cases that were not previously imaginable.
5G security has a tremendous enabling role to play in this new mobile generation. Cybersecurity doesn’t have to be an afterthought, a risk to be mitigated, and a barrier to be overcome. Done right, building cybersecurity into the 5G architecture can be simple enough to reduce operational complexity, intelligent to provide contextual security outcomes, and predictive to swiftly respond to known and unknown threats in real time.
In a sense, the closer analogy for the advent of 5G is the jump that was made from 1G to 2G. Just as the move toward a digital, cellular network — Global System for Mobile Communications (GSM) — made possible an expanded scope of voice and data capabilities, so too the move to software-defined network functionality in 5G is revolutionizing what mobile networks can do for users, enterprises, and the mobile network operators (MNO) that serve them.
Here are just a few of the new mobile capabilities:
- Network slices, organized by vertical industry functions, make it possible to create custom private 5G networks with specific service characteristics.
- Multi-access Edge Computing (MEC), with its distributed support for low latency and capacity for rapid delivery of massive data amounts, enables mission critical enterprise applications and creates richer experiences across remote devices.
- Dense, small cell architecture, using femtocells, picocells, and microcells, play a critical role in densifying the network so that devices and applications that require a low latency, ultra-reliable connection can be supported.
Simply put, there is a lot more to mobile technology than a phone call these days.
And with that expansion of capabilities comes both greater opportunities for products and services delivered by 5G, as well as greater risks from exploitation of the architecture and technology. Here are some signs that we’re not paying close enough attention to how 5G is changing the game.
- Lateral movements and network complexity — hacking has never had it so good.
The distributed 5G network no longer has a clear perimeter. The assets and workloads of service providers, enterprises, and CSPs are intertwined. Only through visibility and control across the whole system can service providers and enterprises detect breaches, lateral movement, and stop the kill chains.
The US government has sounded the alarm on lateral movements, particularly in cloud-based platforms: “Communications paths that rely on insecure authentication mechanisms, or that are not locked down sufficiently by policy, could be used as a lateral movement path by an attacker, allowing the attacker to change between planes or pivot to gain privileges on another set of isolated network resources.[1]”
The CISA/NSA joint guidance statement then gives several recommendations for mitigating damage caused by such breaches, among them, “Sophisticated analytics, based on machine learning and artificial intelligence, can help detect adversarial activity within the cloud and provide the 5G stakeholder with the means to detect malicious use of customer cloud resources.”
Protecting this complex, distributed network environment needs a platform approach with ML-powered threat detection that secures the key 5G interfaces under a single umbrella — no matter if they are distributed across private and public telco clouds and data centers.
- Speed is a two-edged sword — taming the beast with automation and AI.
By now it is clear that the expanded surface area of a 5G network — including MEC, network slices, and instantaneous service orchestration — creates territory where speed is both expected and routine. However, highly distributed cloud and edge environments, as well as a proliferation of IoT and user-owned devices also add up to a 5G security environment where threats evolve faster, move more quickly, get more opportunities to attack, and can potentially cause more damage. Additionally, the complexity of modern mobile network architectures enhances the risk of misconfiguration.
These complex, distributed environments are making it very difficult for MNOs to manage networks or to implement changes without the use of automation and artificial intelligence that can tame the speed.
Automation is the key for building 5G networks that have security built in from the moment they are built or the moment new services are configured.
Additionally, there is the concept of proportional response in managing threats to consider. Attacks rapidly and automatically evolve, and attackers use machines to automatically morph attacks. Also, the threat actors utilize artificial intelligence (AI) to automate and obfuscate the attacks, and thus similar techniques are needed to defend.
Automation and AI should be at the core of 5G security to analyze vast amounts of telemetry, proactively assist in intelligently stopping attacks and threats and recommend security policies.
- 5G innovation and architecture are driving new use cases into unexpected quarters.
For example, the ultra-dense, low latency world of small cell architecture (femtocells, picocells, and microcells) is growing exponentially. Recent industry reports project small cell deployment to grow by 800% over the next decade[2]. Some of the largest growth in the US will come in suburban areas.
The advent of mobile slice services, with their customizable network features and service level agreements (SLA), are already making waves within specific vertical industry use cases, such as automation of field services in upstream energy enterprises, exciting new potential for real-time clinical analysis in healthcare provider settings, and for ultra-reliable, very low latency data services for automated vehicles. The potential has been barely considered thus far.
Meanwhile, U.S. fixed wireless access (FWA) where mobile is used as the primary form of connectivity is expected to grow at 16.1% CAGR through 2026[3]. This growth will be driven by the expansion of 5G networks, particularly in rural areas, coupled with increasing demand for broadband internet in the post-Covid-19 era.
The challenge, of course, will be to meet the demand for 5G use cases in areas well outside of the densely populated core city areas.
- CSPs aren’t picking up on enterprise appetite for 5g security-as-a-service demands.
A 2020 study[4] by BearingPoint conducted with analyst firm Omdia found that only one in five (21%) of early enterprise 5G deals is being led by communication service providers (CSPs), while secondary CSP suppliers such as hardware suppliers and systems integrators are leading 40% and large enterprises themselves are leading about a third.
Another BearingPoint survey of 250 executives[5] from large enterprises and small/medium businesses (SMBs) found a predominant appetite for all-encompassing platform solutions. Some 62% of enterprises and 68% of SMBs want to buy a 5G solution that is “ready to go.” This, of course, means solutions that act on and deliver outcomes that are designed from the beginning to be secure and to keep privileged data private.
- Skills are not keeping up with needs for 5G security.
Perhaps the biggest flashing signal that 5G security needs more attention comes from a GSMA study last year that reported[6] 48% of surveyed operators said that not having enough knowledge or tools to discover and solve security vulnerabilities was their greatest challenge. This is further exacerbated for 39% of surveyed operators who indicated that they were faced with a limited pool of security experts. Operators want to focus on security, but to do so they will need to partner to build up their offerings.
There are tremendous opportunities and appetite for the capabilities that 5G mobile technology can deliver. We can open new markets for private networks, industry-centric slice solutions, small cell deployment, and MEC applications. But it’s equally clear that all of us in the mobile industry, whether MNOs, network equipment providers, integrators, OEMs, cybersecurity software vendors or cloud players, must work together to develop the simpler platform solutions that the markets are seeking if we are to make good on the promise of 5G.
About Keith O’Brien:
Keith serves as the Chief Technology Officer for the Service Provider line of business at Palo Alto Networks and has over 25 years of leadership experience in the cyber security industry. His specialties include secure network design and architecture, application security, devsecops and service provider security.
[1] Security Guidance For 5g Cloud Infrastructures, October 2021. US National Security Agency (NSA) and Cybersecurity and Infrastructure Security Agency (CISA).
[2] The Suburban Migration: New Mapping Analysis Reveals Surprising U.S. Small Cell Growth, Altman Solon. July 2022.
[3] US FWA residential subscribership will grow at 16.1% CAGR through 2026, Global Data Technology. July 2021.
[4] DTWS: CSPs losing big on 5G enterprise deals, BearingPoint and Omdia. October 2020.
[5] DTWS: CSPs losing big on 5G enterprise deals, BearingPoint. October 2020.
[6] Securing private networks in the 5G era, GSMA. June 2021.
