Speaking Engagements & Private Workshops - Get Dean Bubley to present or chair your event

Need an experienced, provocative & influential telecoms keynote speaker, moderator/chair or workshop facilitator?
To see recent presentations, and discuss Dean Bubley's appearance at a specific event, click here

Friday, March 15, 2019

Wi-Fi dominates in the home, but the industry cannot be complacent about the challenges of 5G

I recently gave a keynote speech at the Wi-Fi Now conference in Shanghai (link), which was co-located with the Wi-Fi Alliance's Asia regional members' meeting. My presentation (link) covered the forthcoming opportunities and challenges for Wi-Fi technology and solutions.



My key points were that of a mixed, nuanced, good/bad story:
  • The position of Wi-Fi in the home is extremely strong. The growth of whole-home Wi-Fi using extenders and mesh-network technology, cloud management platforms and the imminent arrival of Wi-Fi6 (formerly 802.11ax) for more capacity and performance, gives it an unassailable lead for most domestic applications and devices.
  • 4G and 5G pose almost no threat to Wi-Fi dominance in the home. While smartphone users may spend slightly more time on cellular if they have large/unlimited data plans, this is more than offset by greater use of Wi-Fi-only devices (TVs, PCs, smart speakers & other IoT products). Few homes or people will go smartphone-only, especially in developed economies. Furthermore, cellular networks lack the in-home positioning and intelligence layer of newer Wi-Fi solutions. Fixed-wireless 5G will almost always need Wi-Fi inside, connected to an external/window antenna. Even in the (few) places 5G FWA can substitute for fixed/cable broadband, it will still mostly deliver last-metre connections via Wi-Fi.
  • A small number of future consumer IoT devices may have "direct to cellular" connections, but these will be the minority, typically restricted to out-of-home products like pet trackers or smart watches. 5G/eSIM-based PCs, wearables & tablets will remain a tiny % of the total device universe. Don't expect 5G-enabled TVs or 4G/NB-IoT washing machines. For low-power smart home products, alternative connections like Bluetooth and ZigBee will remain important - not everything can use Wi-Fi. There might be some LPWAN devices using SigFox, LoRa or perhap Wi-Fi's own HaLow technology, although that is still moving only slowly.
  • In "carpeted enterprise" offices, Wi-Fi will remain critical for many users, alongside wired ethernet for desktops, servers and other non-moving equipment. While there will be a growing need for better indoor 4G/5G coverage (especially in higher frequencies), that cannot replace normal LAN technology. Users will expect both Wi-Fi and cellular to have reliable indoor coverage, capacity and security. The Venn diagram of use-cases only has a narrow overlap. That said, delivering good Wi-Fi indoors is not easy - and like cellular, it will need more "deep fibre", especially for higher frequency bands in future, like 6GHz or 60GHz.
  • Industrial Wi-Fi faces more challenges. While for many applications, industrial-grade Wi-Fi and meshes is widely used, the growth of IoT, robotics and realtime automation/cloud will start to make wireless connectivity more business-critical, and in some cases safety-critical. Unlicensed spectrum, and limited coverage/mobility support may make Wi-Fi's role prone to substitution by public and private 4G/5G networks in some cases. There appears to be more work being done to make spectrum available, as well as integrate with manufacturing / process machinery on the cellular side. There doesn't appear to be as much of a cohesive ecosystem - I don't see a Wi-Fi equivalent of the 5G Alliance for Connected Industries and Automation, for instance. 
  • Limited focus on verticals and testbeds. While I'm skeptical of mobile operators' roles in pushing 5G solutions on various sectors, there is no doubt that the cellular industry is working hard. There are countless "5G for Sector X" associations and collaborations, multiple university and government 5G research institutes, and promotional efforts galore. Why is there no equivalent "University of ABC Wi-Fi Innovation Lab", of "Wi-Fi Healthcare Alliance", where vendors and developers can experiment with new use-cases and create more public "buzz"? The industry - and its major vendors - need to step up and increase funding for this sort of thing. Broadcom, Qualcomm, Cisco, Google, Intel, HPE, Amazon, Arris etc - I'm looking at you.
  • Public Wi-Fi faces growing risks. While I'm writing this article on my laptop in a cafe, most of my fellow coffee-drinkers are using smartphones. This location has a simple one-click access to Wi-Fi, but other places often have crass and privacy-invasive attempts to "monetise" Wi-Fi, with extra login personal data demanded, or attempts to get users to connect with FB or Twitter accounts. The growing number of large/unlimited data-plans, coupled to reduced roaming fees for travellers, and even tethering for PCs, makes the relative pain of accessing 4G vs. public Wi-Fi very different to a few years ago. Venues wouldn't force people to give email addresses or social logins to use amenities like elevators or bathroooms; they shouldn't for Wi-Fi either. The industry should campaign against cumbersome logins. It doesn't need full automation like PassPoint or HotSpot 2.0 - just "frictionless" access. One click on a splash-screen is fine.
  • The Wi-Fi brand is over-protected. Technically, Wi-Fi is a brand that can only be used by products or SPs that get certification from Wi-Fi Alliance. The technology is based on IEEE 802.11 standards. While the guarantee of interoperabilty and "it works anywhere" is great, this poses a problem. There's a lot of cool stuff being done with 802.11 that isn't standardised and fully certified. I've seen licensed-spectrum versions (potentially more suitable for industrial markets). There's some great innovation with 60GHz 802.11, formerly called WiGig. Some vendors have proprietary tweaks, like Blu Wireless allowing gigabit transmission to fast-moving vehicles with handoffs in the UK AutoAir testbed (link). These innovators can't use the term "Wi-Fi", so instead they call their products "5G", adding to the noise and hype - and its impact on regulators and policymakers who then think that 5G deserves special treatment (and spectrum). In my view, the Wi-Fi industry is undermining its own importance, especially if they want to create a case for 6GHz, or other future bands. Most people in politics don't grasp Wi-Fi's level of economic and societal contributions, so to me this seems to be an own-goal. Maybe WFA should create a new category called something like Wi-FiX for "experimental" technologies to capture this extra goodwill.
  • Wi-Fi combines well with other technologies. I'm seeing a growing amount of important intersections between Wi-Fi and other domains. AI, for instance, is being used to manage Wi-Fi fleets by major service providers, as well as cloud companies. Juniper just acquired Mist Systems for its AI-enhanced Wi-Fi solution (link). It's being integrated with both consumer and enterprise IoT. Amazon just acquired eero, a home mesh-Wi-Fi specialist. I'd be unsurprised to see some sort of Alexa-eero hybrids in future (link). And I've written before about the Wi-Fi + blockchain opportunities, including those of my friends at AmmbrTech (link) and assorted others. There's some cool stuff using Wi-Fi for motion-detection as well.
  • The Wi-Fi industry needs to be emphatic - and fast - about creating versions designed to work in all spectrum bands, licensed, unlicensed and shared. There are use-cases for all of these, especially with moves to opening up CBRS and c-Band for more innovative use-cases. We see cellular technologies adopting unlicensed variants like 4G-LAA and MulteFire. Wi-Fi should make the opposite & equivalent move.
  • Yes, 5G and Wi-Fi will work together. Absoutely, I see many reasons to integrate 4G, 5G and Wi-Fi in various guises, both at a network level and service level. We will see MNOs that have big Wi-Fi footprints. We'll see 5G FWA with Wi-Fi indoors. We see dual-connected home broadband gateways with both fibre and cellular modems. We have offload, onload, Wi-Fi first MVNOs, Wi-Fi calling and and non-3GPP access to 4G and 5G core networks. This trend will continue. Yet I still see ignorant references to Wi-Fi being "part of 5G", or "killed by 5G". It is neither; sometimes the technologies will be complementary, and sometimes competitive or substitutive. But in all cases, they compete for the oxygen of publicity, attention and policymaker focus. The Wi-Fi industry needs to shout louder to the media and governments.
  • This is Wi-Fi's 20th Anniversary year as a consumer brand, notably. Maybe Apple (which launched its original AirPort in 1999) might pull a "One more thing..." surprise this year...
In summary: I'm very positive about Wi-Fi, especially with the capabilities of mesh, cloud and Wi-Fi6 for the home. But I'm also concerned that the industry isn't being sufficiently ambitious. Yes, in the US there are a lot of positive signs from the FCC about 6GHz The recent European emphasis on Wi-Fi derived DSRC for vehicles is another win. But while those could be big successes, and keep the industry busy for a long time, it's not enough. 



In industrial use-cases, Wi-Fi faces significant challenges. For public hotspots, some venues' ludicrous sign-up demands are self-defeating and harm overall public perception of the technology. HaLow isn't getting enough attention in the LPWA space. The industry needs to market itself more loudly, more globally, and to a wider audience. It needs to create more space for innovators and developers, with collaboration forums and easier access to documents - and a willingness to extend the brand's goodwill, even to those that aren't doing something fully-standardised.

The Wi-Fi industry deserves congratulations on 20 years and $2trillion of economic value. But it needs to double-down on its scope and ambitions, to make sure it will be in strong position at its 40th as well.


Watch out soon for a podcast covering this article - my SoundCloud account is here (link)

And if this is an area where you'd be interested in my input for advisory work, speaking/presentations, or other engagements, please contact me via information AT disruptive-analysis dot com

I will also shortly be publishing a long-form research report on the Consumer Wi-Fi sector, as part of my Network Futures subscription stream with STL Partners (link). It looks at some of the issues raised here in more detail, and focuses in particular on the implications and role of telecom operators and other broadband providers. I'll update this blog post, and also put out details on my Twitter account (link) when it's available.