March 28, 2024

Shaping Wi-Fi’s future: the wireless-mobile convergence

According to recent news, Comcast is being sued because it is taking advantage of users’ resources to build up its own nationwide Wi-Fi network. Since mid-2013 the company has been updating consumers’ routers by installing new firmware that makes the router partially devoted to the “home-user” network and partially devoted to the “mobile-user” network (a Comcast service named Xfinity WiFi). In fact, the same network infrastructure offers two different kinds of connection: the first one covers a comparatively restricted (local) area and stays under the relative control of the private end-user; the second kind of connection is “shared” between Comcast customers and covers a wider area, compatible with the range of national mobile carriers. In other words: the last mile of data transmission is being made mostly by a group of home based routers (or access points) that offers two different Internet connection services, the local “private” network and the metropolitan “shared” network.

Integrating several wireless Internet access points into a seamless shared network isn’t new technology. Lots of Wireless Community Networks (WCNs) have been doing that in a collaborative and non-profit fashion by using Wi-Fi and Mesh technologies. De Filippi and Tréguer (2015) teach us about the citizen-centric value of these communities and, more than that, about the political struggle they’re facing to produce relative autonomy against commercial Internet Service Providers (ISPs). There are many examples of WCNs in which there are no clear boundaries between Internet users and providers or between users’ equipment and network infrastructure, configured in a peer-to-peer-like architecture. This operation mode is legitimated by the community-driven approach once the network is deployed by and for the community; the network infrastructure is understood and managed as commons.

On the other hand, the picture drastically changes once a huge corporation like Comcast starts blurring the user-provider boundary by default and, apparently, without properly informing its customers. At the end of the day the lawsuit against Comcast represents a dispute that is related to the ownership and control over a specific slice of the data transfer infrastructure. With the risk of oversimplifying, we may say that the question is: who’s the owner and the controller of the wireless router?

Cutting the Network

Anthropology can teach us that ownership and property relations are absolutely decisive in Cutting the Network (Strathern, 1996). Property relations very often are the responsible for establishing the limits of a network; Strathern argues, for instance, that intellectual property rights cut down large knowledge networks. Here, specifically in regard to the XFinity WiFi case, the connection between ownership and the network’s range is quite literal: the “new network” did not affect Comcast customers who use their own wireless routers at home. Comcast has extended its Wi-Fi network using only the wireless routers that the company leases to its customers. It seems that the company considers it fair to take advantage of its own infrastructure even when it is placed into customers’ houses. However, the plaintiffs in the lawsuit raise interesting concerns by arguing, for example, that the vagueness of the contract of service does not allow one to say for sure whether Comcast is allowed to do that. The formal claim also highlights the fact that Comcast is making extra profit by sharing the costs of extending its own Wi-Fi hotspot network with a group of customers. To give a very concrete argument: the new “shared” network raises router power consumption, and Comcast customers are paying the bill. Besides, there are consumers claiming that the quality of service of the home-user (or “private”) network is getting worse since the bandwidth dedicated to each home-based access point is now being shared with other users. Finally, there are critical security issues since the “public” connection to the wireless router might represent a threat to the home-network privacy.

Dissatisfied consumers have tried to disable the “public” network with no success. There’s a long discussion on the case in the social network Reddit; a top comment reads:

[Comcast’s] website says this to turn it off:

We encourage all subscribers to keep this feature enabled as it allows more people to enjoy the benefits of XFINITY WiFi, but you will always have the ability to disable the XFINITY WiFi feature on your Wireless Gateway. Visit My Account at https://customer.comcast.com/, click on “Users & Preferences” and then select “Manage XFINITY WiFi” or call 1-800-XFINITY.

To which you’ll get a message:

We’re sorry. Something unexpected happened. Please try again later

And then you’ll call Comcast. Their techs will try to talk you into changing your mind, say they disabled it, but you check later and they didn’t.

This is not the only report from Comcast customers saying that they have followed all required steps to disable the “public” network but it did not work out. In face of this situation, another comment argues that the only ultimate solution is replacing Comcast’s wireless router with a private one, so that the end-user can control the device’s setup. The ownership over a strategic component of the network infrastructure, the wireless router, apparently assures relative autonomy to the end-user, at least to avoid broadcasting the additional “shared” network. The lawsuit is a formal way of concretizing the conflict over the ownership and the boundaries of Internet network infrastructure. In fact, the formal complaint against the cable operator Comcast is just a facet of a much wider conflict: Comcast is stepping in the mobile broadband market and directly competing with cellular network operators. Coming from other direction, mobile carriers are adopting Wi-Fi as complementary solution to cellular networks. In short, ISPs and mobile carriers (among others market players, such as equipment manufacturers) are competing and cooperating to define the technological standards and the business model that will rule the mobile broadband market.

Wi-Fi’s next generation: 802.11ax – High Efficiency Wireless LAN.

I recently learned that the Wi-Fi’s next generation (IEEE802.11ax) is being designed to increase the level of integration and synergy with cellular networks, especially in regard to the LTE (4G) network. I was advised about that while doing fieldwork research within a standards development organization (SDO), the IEEE-SA.

The movement of cellular network operators towards Wi-Fi technology is quite easy to understand once one considers at least two facts: first, Wi-Fi works on unlicensed bands which are free of charge – it’s largely known that mobile carriers normally spend billions of dollars in spectrum auctions for licensed bands.

The second fact is that 4G/LTE networks do not work properly in areas with a high density of users. Despite the advances in relation to 3G systems, 4G/LTE network does not provide enough throughput capacity to satisfy the always increasing user demand. Mobile carriers already make use of Wi-Fi hotspots to offload traffic from cellular networks. But, Wi-Fi technology wasn’t initially designed for this purpose and also fails to work properly in the dense area situation. That’s why some network operators (especially Orange and China Mobile) have advocated for the IEEE802 group (responsible for developing Wi-Fi standards) to propose the High Efficiency Wireless LAN (HEW) project. To put it simply, we may say that carrier’s approach to 802.11 was something like that: hey, why don’t you make 802.11 look more like LTE?

Carriers were complaining about 802.11’s performance in dense environments and requiring changes and new features that turn Wi-Fi very close to Wireless Advanced standards (4G requirements as defined by ITU). Carriers are also concerned with producing handover mechanisms between Wi-Fi and LTE (4G) networks; those mechanisms must help switch the user’s connection between networks by defining which connection to choose and how long to maintain the connection. Initially, carriers’ requirements were too broad and it took about a year to narrow down the scope and write the HEW project. The project ended up focusing on solving the dense environments situation and increasing the throughput capacity. One must keep in mind that IEEE802.11 is the Working Group (WG) responsible for writing Wi-Fi’s standards for MAC and PHY layers. Along with IEEE802 there are other entities (like IETF, 3GPP, WBA and Wi-Fi Alliance) writing pieces of the ultimate solution that will integrate Wi-Fi and cellular networks; 3GPP, for instance, is writing down the handover mechanisms and, at the same time, has proposed developing its own network standard working on unlicensed spectrum, the LTE-U.

The technopolitics of wireless-mobile convergence.

The mobile telephony business has evolved into the mobile broadband business; as cause and consequence cellular networks are converging with broadband networks. The only two existing cellular 4G standards (LTE-Advanced and Wireless MAN-Advanced) are all IP systems. Wireless and mobile both designate that the end user’s connection to the network is made through an air interface. However, the prime difference is that the former was created within the computer networks evolution while the latter comes from cellular telephony networks. This convergence process is not new but, from now on, the merge between wireless and mobile must become an object of public attention. It’s an important technopolitical issue for at least two good reasons.

First, carriers are going to monetize the unlicensed spectrum that is widely understood as a common and shared resource. The intensive use of that spectrum’s share might interfere negatively with the functioning of other technologies that coexist in it. Engineers and policy makers should take that into account, especially having in mind that the unlicensed spectrum has a truly public dimension and must support the simultaneous working of distinct technologies. Second, mobile carriers are used to exercising a high degree of control and concentrated power. The cellular architecture is designed in a way that the service provider has “absolute” control over the network infrastructure including users’ devices. At the cellular network management level, identifying and localizing each device is essential, for example, to decrease signal interference, to apply billing methods and to maintain an expected level of quality of service. The same does not happen with Wi-Fi network architecture. In addition, carriers are used to exerting dominance over the whole market since they’ve paid a lot of money for a licensed share of the spectrum; they exercise great influence over device manufactures by approving (or not) the equipment to work on that spectrum band, or even by telling the manufactures what chip they can put into the device. As soon as carriers are moving into a position in which they have paid nothing for the spectrum, the picture changes and they can’t have the same degree of control, since other people are allowed to deploy technology operating at that band. The simultaneous working of two technologies in the same spectrum share might be harmful to both of them; for instance, they might simultaneously cancel each other’s signal.

To finish up this post, let’s go back to the beginning: the lawsuit claims that Comcast is practicing unfair competition by deploying a nationwide wireless/mobile network based on a pre-existing infrastructure and end-users’ resources. Comcast is trying one way to navigate the crossroad that puts together ISPs and cellular operators. We’re in the early stage of a huge competition to achieve dominance and control over the mobile broadband market. The hybrid architecture mixing Wi-Fi, cellular, and other upcoming technologies is still being shaped, and there are many questions to be answered about its mode of operation and topology. Mobile carriers will try to impose their centralized approach over the upcoming network infrastructure, including Wi-Fi and unlicensed spectrum. Regulators, academics, and policymakers must pay attention to protect public interests during this upheaval in the market.

Comments

  1. John Laprise says

    This misses the strategy being employed by global internet companies to provide ubiquitous broadband without relying in terrestrial architecture and largely sidestepping regulatory supervision.

    https://t.co/BrC2wpHPKL

    • Hi John, thanks for your comment. I’ve been tracking “non terrestrial architecture” and, by now, it seems to me more like a complementary infrastructure especially related to regions where the internet access is poor. In terms of market share, it cannot compete in equal feet with “terrestrial” (cable, wi-fi or cellular). But, anyway, it poses interesting questions related to regulation, and to the competition/cooperation between Telecom and Internet companies.

  2. Does anyone else think that the obvious solution in the US is for Comcast et al to simply force other users out of the bands they want? After all, those other uses are interfering with the telephone network… (Yes, I’m being sarcastic. Yes, I think we’ll start hearing statements to this effect if we haven’t already.)

    • Hi Paul, probably there will be statements in this direction, and they’ll certainly argue that it’s an important measure to foster economic growth by providing better broadband service.

  3. Steve Loughran says

    The “BT Fon” system in the UK does something similar, but they are (a) open about it and (b) offer you roaming wifi if you don’t opt out of it.

    1. The “fon” system started out as something co-operative added to linksys WG base stations
    2. After BT acquired the company, they put it into their “smarthub routers”
    3. In exchange for not disabling in your account (web settings at bt.com), you gain the right to use free roaming wifi access round the country -aided by apps for android and IOs devices.
    4. Even if you swap out the base station for your own, you can retain roaming access: They don’t check.
    5. The guest wifi is throttled.
    6. non-members of the network can join, they just pay by the hour

    This is actually a nice arrangement: more than once I’ve ended up roaming attached to a wifi unit near a hotel rather than pay hotel fees. The company offers you something “free roaming” in exchange for you not opting out of a service implemented in their hubs. The company gains from (a) bandwidth sold to visitors and (b) “stickier” customers who would lose the service if they went with a competitor.

    I don’t notice a bandwidth problem, but then BT provide 80 mbit/s down, 18 up, for $80/month which comcast probably don’t.