Background
I'm increasingly finding myself drawn into discussions of
#geopolitics and how it relates to #telecoms. This goes well beyond
normal regulatory and policymaking involvement, as it means that rules -
and opportunities and risks - are driven by much larger "big picture"
strategic global trends, including the war in Ukraine.
As well as predicting strategic shifts, there are also lessons to be
learned from events at a local, tactical level which have wider
ramifications. Often, there will be trade-offs against normal telecoms
preoccupations with revenue growth, theoretical "efficiency" of spectrum
or network use, standardisation, competition and consumer welfare.
This is the first of what will probably be a regular set of articles
on this broader theme. Here, I'm focusing on the Ukraine war, in the
context some of the other geopolitical factors that I think are
important. I'm specifically thinking about what they may mean for the types
of network technology that are used, deployed and developed in future.
This has implications for #5G, #6G, #satellite networks, #WiFi, #FTTX
and much more, including the cloud/edge domains that support much of it.
Ukraine and other geopolitical issues
This article especially drills into how the conflict in Ukraine has
manifested in terms of telecoms and connectivity, and attempts to
extrapolate to some early recommendations for policymakers more broadly.
I'm acutely consicous of the ongoing devastation and hideous war
crimes being perpetrated there - I hope this isn't too early to try to
analyse the narrow field of networking dispassionately, while conflict
still rages.
For context, as well as Ukraine, other geopolitical issues impacting telecoms include:
- US / West vs. China tensions, from trade wars to broader
restrictions on the use of Huawei and other vendors' equipment, as well
as sanctions on the export of components.
- Impact of the pandemic
on supply chains, plus the greater strategic and political importance
of resilient telecom networks and devices in the past two years.
- The
politics of post-pandemic recovery, industrial strategy and stimulus
funds. Does this go to broadband deployment, themes such as Open RAN,
national networks, smart cities/infrastructure, satellite networks... or
somewhere else?
- Tensions within the US, and between US and
Europe over the role and dominance of "Big Tech". Personal data,
monopoly behaviour, censorship or regional sovereignty etc. This mostly
doesn't touch networks today, but maybe cloud-native will draw
attention.
- Semiconductor supply-chain challenges and the geopolitical fragility of Taiwan's chip-fabrication sector.
- How telecoms (and cloud) fits within Net Zero strategies, either as a consumer of energy, or as an enabler of green solutions.
- Cyber
threats from nation-state actors, criminal cartels and terrorist-linked
groups - especially aimed at critical infrastructure and
health/government/finance systems.
In other words, there's a lot going on. It will impact 5G, 6G
development, vendor landscapes, cloud - and also other areas such as
spectrum policy and Internet governance.
Network diversity as a focus
I've written and spoken before about the importance of "network
diversity" and the dangers of technology monocultures, including
over-reliance on particular standards (eg 5G) or particular business
models (eg national MNOs) as some sort of universal platform. It is now
clear that it is more important than ever.
The analogy I made with agriculture, or ecological biodiversity, is proving to be robust.
(Previous work includes this article from 2020 about private enterprise networks, or my 2017 presentation
keynote on future disruptions, at Ofcom's spectrum conference. (The
blue/yellow image of wheat fields, repeated here in this post, was
chosen long before it became so resonant as the Ukrainian flag). I've
also covered the shift towards Open RAN and telecoms supplier
diversification – including a long report I submitted to the UK
Government's Diversification Task Force last year - see this post and download the report).
A key takeout from my Open RAN report was that demand diversity is as important as creating more supply
choices in a given product domain. Having many classes of network
operator and owner – for instance national MNOs, enterprise private
4G/5G, towercos, industrial MNOs and neutral hosts – tends to pull
through multiple options for supply in terms of both vendor diversity and technology diversity. They have different requirements, different investment criteria and different operational models.
In Ukraine, the "demands" for connectivity are arising from an even
more broad set of sources, including improvised communications for
refugees, drones and military personnel.
The war in Ukraine & telecoms
There have been numerous articles published which highlight the
surprising resilience and importance of Ukrainian telecoms during the
war so far. Bringing together and synthesising multiple sources, this
has highlighted a number of important issues around network
connectivity:
- The original “survivability” concept of IP networks seems to
have been demonstrated convincingly. Whether used for ISPs’ Internet
access, or internal backhaul and transport for public fixed and mobile
networks, the ability for diverse and resilient routing paths seems to
have mostly been successful.
- Public national mobile networks -
mostly 4G in Ukraine's case - have proven essential in many ways,
whether that has been for reporting information about enemy combatants'
locations and activities, obtaining advice from government authorities,
or dealing with the evacuation as refugees. (I'm not sure if subway
stations used as shelters have underground cellular coverage, or if
there is WiFi). Authorities also seem to have had success in getting
citizens to self-censor, to avoid disclosing sensitive details to their
enemies.
- Reportedly the Russian forces haven't generally
targeted telecoms infrastructure on a widescale basis. This was partly
because they have been using commerical mobile networks themselves.
However, because roaming was disabled, Russian military use of their
encrypted handsets and SIMs on public 3G/4G networks seems to have
failed. Two articles here and here
give good insight, and also suggests there may be network surveillance
backdoors which Russia may have exploited. There have also been reports
of stingrays ("fake" base stations used for interception of calls /
identity) being deployed. It also appears that some towns and cities -
notably the destroyed city of Mariupol - have been mostly knocked
offline, partly because the electrical grid was attacked first.
- Ukraine’s
competitive telecoms market has probably helped its resilience. There
is a highly fragmented fixed ISP landscape, with very inexpensive
connections. There are over a dozen public peering-points across the
country. There are three main MNOs, with many users having SIMs from 2+
operators. (This is a good overview article - https://ukraineworld.org/articles/ukraine-explained/key-facts-about-ukraines-telecom-industry). It seems they have enabled some form of national roaming to allow subscribers to attach to each others' networks.
- WiFi hotspots (likely with mobile backhaul) have been used by NGOs evacuating refugees by buses.
- Although
it is still only being used at a small scale, the LEO satellite
terminals from SpaceX’s StarLink seem to be an important contributor to
connectivity – not least as a backup option. Realistically, satellite
isn’t appropriate for millions of individual homes – and especially not
personal vehicles and smartphones – but is an important part of the
overall network-diversity landscape. Various commentators have suggested
it is useful as a backup for critical infrastructure connectivity, as
well as for mobile units such as special forces.
- Another satellite broadband provider, Viasat, apparently suffered a cyberattack at the start of the war (link here),
which knocked various modem users offline (or even "bricked" the
devies), reportedly including Ukrainian government organisations.
Investigations haven't officially named Russia, but a coincidence seems
improbable. This attack also impacted users outside Ukraine.
- Various
peer-to-peer apps using Bluetooth or WiFi allow direct connections
between phones, even if wide area connections are down (see link)
- There
have been some concerning reports about the impact of GPS jammers on
the operation of cellular networks, which may use it as a source of
“timing synchronisation” to operate properly, especially for TDD radio
bands. While this has long been a risk for individual cell-sites from
low-power transmitters, the use of deliberate electronic warfare tools
could potentially point to broader vulnerabilities in future.
- There
has been wide use of commercial drones like the DJI Mavic-3 for
surveillance (video and thermal imaging), or modified to deliver
improvised weaponry. These use WiFi to connect to controllers on the
ground, as well as a proprietary video transmission protocols (called
O3+) which apparently has range of up to 15km using unlicensed spectrum.
Some of the "Aerorozvidka" units reportedly then use StarLink terminals
to connect back to command sites to coordinate artillery attacks (link).
In short, it seems that Ukraine has been well served by having lots
of connectivity options - probably including some additional military
systems that aren't widely discussed. It has benefited from multiple
fixed, cellular and satellite networks, with potential for interconnect,
plus inventive "quick fixes" after failures and collaboration between
providers. It is exploiting licensed and unlicensed spectrum, with
cellular, Wi-Fi and other technologies.
In other words, network diversity is working properly. There appears
to be no single point of failure, despite deliberate attacks by
invading forces and hackers. Connectivity is far from perfect, but it
has held up remarkably well. Perhaps the full range of electronic
warfare options hasn't been used - but given the geographical size of
Ukraine and the inability of Russia forces to maintain supply-lines to
distant units, that is also unsurprising.
Another set of issues that I haven't really examined are around
connectivity within sanctions-hit Russia. Maybe it will have to develop
more local network equipment manufacturers - if they can get the
necessary silicon and other components. It probably will not wish to
over-rely on Huawei & ZTE any more than some Western countries have
been happy with Nokia and Ericsson as primary options. More problematic
may be fixed-Internet routers, servers, WiFi APs and other
Western-dominated products. I can't say I'm sympathetic, and I certainly
don't want to offer suggestions. Let's see what happens.
Recommendations for policymakers, industry bodies and regulators
So what are the implications of all this? Hopefully, few other
countries face a similar invasion by a large and hostile army. But
preparedness is wise, especially for countries with unfriendly
neighbours and territorial disputes. And even for everywhere else, the
risks of cyberattacks, terrorism, natural disasters - or even just
software bugs or human error - are still significant.
I should stress that I'm not a cybersecurity or critical
infrastructure specialist. But I can read across from other trends I'm
seeing in telecoms, and in particular I'm doing a lot of work on "path
dependency" where small, innocent-seeming actions end up having
long-term strategic impacts and can lock-in technology trajectories.
My initial set of considerations and recommendations:
- As a general principle, divergence in technology should be
considered at least as positively than convergence. It maintains
optionality, fosters innovation and reduces single-point-of-failure
risks.
- National networks and telcos (fixed and mobile) are
essential - but cannot do everything. They also need to cooperate during
emergencies - a spirit of collaboration which seems to have worked well
during the pandemic in many countries.
- Normal ideas about
cyber-resilience and security may not extend to the impact of full-scale
military electronic warfare units, as well as more "typical" online
hacking and malware attacks.
- Having separate "air-gapped"
networks available makes sense not just for critical communications
(military, utilities etc) but for more general use. It isn't inefficient
- it's insurance. There may be implications here for network-sharing in
some instances.
- Thought needs to be given to emergency
fallbacks and improvised work-arounds, for instance in the event of mass
power outages or sabotage. This is particularly important for
software/cloud-based networks, which may be less "fixable" in the field.
Can a 5G network be "bodged"? (that's "MacGyvred" to my US friends)? As a sidenote - how have electric vehicles fared in Ukraine?
- Unlicensed
spectrum and "permissionless communications" is hugely important during
emergency situations. Yes, it doesn't have control or lawful intercept.
But that's entirely acceptable in extreme circumstances.
- Linkages
between technologies, access networks and control/identity planes
should generally be via gateways that can be closed, controlled or
removed if necessary. If one is attacked, the rest should be firewalled
off from it. For the same reason "seamless" should be a red-flag word
for cross-tech / cross-network roaming. Seams are important. They offer
control and the ability to partition if necessary. "Frictionless" is OK,
as long as friction can be re-imposed if needed.
- Governments should be extremely
cautious of telcos extending 3GPP control mechanisms – especially the
core network and slicing – to fixed broadband infrastructure. Fixed
broadband is absolutely critical, and complex software dependencies may
trade off fine-grained control vs. resilience - and offer additional
threat surfaces.
- Democratising and improving satellite
communications looks like an ever more wise move, for all sorts of
reasons. It's not a panacea, but it's certainly "air-gapped" as above.
3GPP-based "non-terrestrial" networks, eg based on drones or balloons,
also has potential - but will ideally be able to work independently of
terrestrial networks if needed.
- I haven't heard much about LPWAN and LoRa-type networks, but I can imagine that being useful in emergency situations too.
- Sanctions,
trade wars and supply-chain issues are highly unpredictable in terms of
intended and unintended consequences. Technology diversity helps
mitigate this, alongside supplier diversity in any one network domain.
- Spectrum
policy should enable enough scale economies to ensure good supply of
products (and viability of providers), but not *so* much scale that any
one option drives out alternatives.
- The role and impact of
international bodies like ITU, GSMA and 3GPP needs careful scrutiny. We
are likely to see them become even more political in future. If
necessary, there may have to be separate "non-authoritarian" and
"authoritarian" versions of some standards (and spectrum policies).
De-coupling and de-layering technologies' interdependency - especially
radio and core networks - could isolate "disagreements" in certain
layers, without undermining the whole international collaboration.
- There
should be a rudimentary basic minimum level of connectivity that uses
"old" products and standards. Maybe we need to keep a small slice of
900MHz spectrum alive for generator-powered GSM cells and a box of cheap
phones in bunkers - essentially a future variant of Ham Radio.
So to wrap up, I'm ever more convinced that Network Diversity is
essential. Not only does it foster innovation, and limit oligopoly risk,
but it also enables more options in tragic circumstances. We should
also consider the potential risks of too much sophistication and pursuit
of effiency and performance at all costs. What happens when things
break (or get deliberately broken)?
In the meantime, I'm hoping for a quick resolution to this awful war. Slava Ukraini!
Sidenote: I am currently researching the areas of “technology
lock-in” and “path dependence”. In particular, I have been investigating
the various mechanisms by which lock-in occurs and strategies for
spotting its incipience, or breaking out of it. Please get in touch with
me, if this is an area of interest for you.
Both organisations have previously published terrible studies by consultants, riddled with flawed assumptions and dodgy multiplier "fiddle factors". I’ve loudly criticised Axon and Coleago reports related to the (un)#fairshare and #6GHz #spectrum debates respectively.
So I started the ADL report with trepidation, not helped by a strange typo / editing error in the first paragraph.
But actually, the report is pretty good, and I broadly agree with both methodology and conclusions, albeit with one major caveat.
It estimates usage of home and mobile broadband on the basis of hours-per-day of active use of heavy applications such as video streaming, gaming and possible metaverse-type experiences.
I’ve used GB-per-hour myself, to model passenger data-traffic demand on trains. It makes more sense than the usual Gbps, as most applications are “bursty”. It also fits the typical heuristics of human behaviour. How many seconds a day do you spend on social media?
The central prediction of 20% growth in fixed traffic and 25% for mobile usage seems reasonable. I could argue for 25/20 rather than 20/25, but it's fine as a rough estimate.
Importantly these rates for the next few years are well within the bounds of both fixed broadband (moving to #FTTP) and mobile (on #5G) without incremental investments in extra capacity, beyond the main "generational" shift & CAPEX. And that is driven by government policy and competition, not traffic load and congestion. The report convincingly shows that nobody really needs/values more than 100Mbps for current apps, so #gigabit networks have plenty of headroom.
My main criticism is there is no analysis of mobile device traffic carried over fixed networks and #WiFi. Smartphones used at home for video, gaming or social media will be c80% on Wi-Fi, and indoor usage is c80% of the total.
The report also talks about AI pre-emptively downloading content for “infinite scrolling”, but doesn't suggest it could be smart enough to do so mostly over cheap / low-energy fixed connections. (IMO, by 2030, governments may *mandate* cellular offload via neutral-host or Wi-Fi for indoor use).
I agree with the report's assertions that VR is in an indoor/fixed application, that most #IoT traffic is a rounding-error and that #Web3 is probably irrelevant. The #metaverse scenarios seem mostly plausible.
One area I think ADL underestimates is fixed broadband for video streaming. While Netflix and YouTube are “active” viewing, historically, many people just leave broadcast TV switched on, even if nobody is in the room except the cat.
If TV really goes online-only, then that becomes a genuine “waste” of capacity, unless you can advertise to pets.
Overall - really quite good analysis, which (ironically, given the sponsors) fatally undermines the #InternetTrafficTax rhetoric.