Australia needs to take the lead on 5G again
17 Sep 2020|

In what seems like another age, back in August 2018 in the midst of yet another change in prime ministers, Australia made a world-first decision to exclude high-risk vendors from its 5G networks. Others, such as the European Union, are still struggling with their 5G decisions, but as I argue in my new ASPI report, Ensuring a trusted 5G ecosystem of vendors and technology, we can’t afford to sit back and reassure ourselves that we have long since resolved all our 5G vendor risk problems.

Although the marketing promise of 5G, as with most new technology, runs ahead of reality, and 5G rollouts have been delayed by Covid-19 and related impacts in many countries, ‘true 5G’ isn’t far away. Telstra announced in May that it had enabled its network for standalone 5G, which is what provides transformational features like ultra-low latency (as opposed to the current capability to download 4K movies to your mobile phone, which, while it may be useful in these days of social isolation, is unlikely to fundamentally change the world).

Opinions differ on what the ‘killer applications’ will be for the 5G future, but it is certain that within a few years these networks will be a key part of mission-critical and safety-critical systems. The resilience of the network will be vital to our national security—not just the privacy of data sent across it, but the availability and integrity of systems will be critical. Trust in the ecosystem of vendors that supply and support these systems will be essential.

The problem is that the market for 5G equipment vendors that we consider to be a reasonable level of risk is highly concentrated, with Nokia and Ericsson at the forefront. From a long-term resilience point of view, the dangers are clear. What happens if one vendor fails or becomes untrusted? What about the security risks if we use just one vendor’s kit everywhere and a vulnerability emerges? After all, when flaws have been found in Apple’s iPhone software, they seem to expose almost every device in the world at the same time. Talking to regulators and telecommunications operators, it’s clear that a number of factors are holding back opportunities for new entrants that could help improve resilience.

It was never supposed to be like this. Telecommunication networks were founded on international standards for interoperability that should allow multiple vendors to work together. However, the process for setting international standards has fallen behind the pace of technology development, and also become politicised—witness some of the recent debates about ‘new IP network’ proposals from China that would put in place new mechanisms for control and centralisation of networks, and the fear US companies have of interacting with Chinese company representatives on international standards bodies due to strict interpretation of export control rules.

This frustration has led to some industry groups setting up their own ad hoc consortiums such as the open radio access network (RAN) movement, but they struggle to gain critical mass and effective funding in many cases. However, Nokia’s recent commitment to include open RAN interfaces on all its products may be a sign of momentum growing behind this movement. This is the time when Australia needs to engage with standards setting and embrace open RAN and similar projects so that we can make a difference.

Another problem is the unwillingness of network providers to integrate multiple vendors. A combination of lack of system engineering expertise and aversion to risk means it often seems easiest to buy an end-to-end solution from one vendor, but in the long term that’s probably neither the most cost-effective nor the most secure option. Redressing this balance will require a combination of carrot (such as setting up central integration lab facilities) and stick (mandating vendor diversity).

One of the big changes of 5G was supposed to be virtualisation. All the complicated processing would be done in software, even complex processing of radio signals, and the underlying hardware would become standardised and commoditised. If this model can be realised, the real innovation and value will come from developing software, not hardware, which would save on the lead time and massive capital costs to establish hardware design and manufacturing capabilities.

In Australia, we have innovative technology companies that should be well placed to develop new remote-sensing and remote-control applications, and we have ready access to vast, sparsely populated areas to test and prove them. Realising the vision of an open marketplace will not only improve the security and resilience of our communications infrastructure, but also provide new industry growth opportunities as we seek to rebuild the economy post-Covid-19.

The time for action is now. True 5G is starting to be rolled out, telecommunications operators will soon make decisions on their network architectures and vendor partnerships, and tens of billions of dollars will start to be spent on infrastructure. The actions that Australia needs to take to address the marketplace challenges would, of course, be easier and more effective to implement by building an international consensus, but that would take too long given the current distractions on top of existing divisions even among our Five Eyes allies.

In August 2018, we were ready to take bold steps and lead the way for like-minded countries. Now is our opportunity to do the same again.