Cellular WiFi Handoffs

If you use anybody except Verizon you may have noticed that your cellphone has become adept at handing your cellular connections to a local WiFi network. Like most people I keep my smartphone connected to WiFi when I’m at home to save from exhausting my cellular data cap. I have AT&T cellular service and I’ve noticed over the last year that when I’m out of the house that my phone often logs onto other WiFi networks. I can understand AT&T sending me to their own AT&T hotspots, but often I’m logged on to networks I can’t identify.

When I lived in Florida I was a Comcast customers and so when I was out of the house my phone logged onto Comcast hotspots. Even today my phone still does this, even though I’m no longer a Comcast customer and I assume there is a cookie on the phone that identifies me as a Comcast customer. I understand these logins, because after I the first time I logged onto a Comcast hotspot my phone assumed that any other Comcast hotspot is an acceptable network. This is something I voluntarily signed up for.

But today I find my phone automatically logged onto a number of hotspots in airports and hotels which I definitely have not authorized. I contrast this with using my laptop in an airport or hotel. With the laptop I always have to go through some sort of greeting screen, and even if it’s a free connection I usually have to sign on to some terms of service. But my phone just automatically grabs WiFi in many airports, even those I haven’t visited in many years. I have to assume that AT&T has some sort of arrangement with these WiFi networks.

I usually notice that I’m on WiFi when my phone gets so sluggish it barely works. WiFi is still notoriously slow in crowed public places. Once I realize I’m on a WiFi network I didn’t authorize I turn the WiFi off on my phone and revert to cellular data. Every security article I’ve ever read says to be cautious when using public WiFi and so I’d prefer not to use these connections unless I have no other option.

There was a major effort made a few years back to create a seamless WiFi network for just this purpose. The WiFi Alliance created a protocol called Hotspot 2.0 that is being marketed under the name of Passpoint. The purpose of this effort was to allow cellular users to automatically connect and roam between a wide variety of hotspots without having to ever log in. Their ultimate goal was to enable WiFi calling that could hand off between hotspots in the same way that cellular phones hand-off between cell sites.

It’s obvious that AT&T and other cellular carriers have implemented at least some aspects of Hotspot 2.0. In the original vision of Hotspot 2.0 customers were to be given the option of authorizing their participation in the Passpoint network. But AT&T has never asked my permission to log me onto WiFi hotspots (unless it was buried in my terms of service). AT&T has clearly decided that they want to use these WiFi handoffs in a busy environment like an airport to protect their cellular networks from being swamped.

It’s interesting that Verizon is not doing this. I think one reason for this is that they don’t want to give up control of their customers. Verizon foresees a huge future revenue stream from mining customer data and I’m guessing they don’t want their customer to be shuttled to a WiFi network controlled by somebody else, where they can’t track customer behavior. Verizon is instead pushing forward with the implementation of LTE-U where they can direct some data traffic into the WiFi bands, but all under their own control. While LTE-U uses WiFi frequency, it is not a hotspot technology and is as hard to intercept or hack as any other cellular traffic.

Most new cellphones now come with the Passpoint technology baked into the chipset. I think we can expect that more and more of our cellular data connections will be shuttled to hotspots without notifying us. Most people are not going to be bothered by this because it will reduce usage on their cellular data plans. I’m just not nuts about being handed off to networks without some sort of notification so that I can change my settings if I don’t want to use the selected network. I guess this is just another example of how cellular companies do what they want and don’t generally ask for customer permission.

A Business Case for WiFi Hotspots

Wi-FiLately I have been asked a number of times if there is a business case to be made for providing a large outdoor WiFi hotspot network. Today I will look at the two issues that answer that question:  1) the hardware available today and;  2) the revenue opportunities.

Hardware Issues. The WiFi industry is currently in a state of what I call ‘between’. This often happens when a new standard is being introduced. There have been existing hotspots on the market for many years. But the whole industry is moving towards implementing Hotspot 2.0, which is a standard that allows for roaming between hotspots the same way that cellphones roam between cell towers. But since the coverage distance of a hotspot is far less – around 250 feet at most from a hotspot – roaming is even more of an issue for WiFi.

With Hotspot 2.0 fully implemented, a customer can automatically log in when walking within range of a hotspot. But more importantly they will maintain whatever they are doing  (such as a web session or IP phone call) without interruption as they move to a new hotspot (as long as they don’t hit a dead area). But the units on the market today can best be characterized as pre-Hotspot 2.0 and they do not yet include all of the features needed to fully support roaming. This means any units you buy today are going to need an upgrade eventually to a standard that is not yet fully defined.

The units on the market today are also very expensive compared to older hotspots. The manufacturers are concentrating on high-capacity hotspots that can handle as many as 500 simultaneous users. These are complicated hotspots with multiple antennae and cost as much as ten times as the old simple hotspots. But these are what are selling and they are made for stadiums, event centers, busy shopping districts or places where there will to be a lot people. But a citywide deployment doesn’t need many hotspots with that huge capacity, but rather much cheaper and lower capacity units that also do Hotspot 2.0.

Revenue Opportunities. The revenue opportunities for an outdoor WiFi network are not clear. I don’t know of any hotspot networks that have been able to pay for themselves. But there may be new revenue opportunities coming that could improve the picture.

There are two traditional WiFi revenue opportunities. One is to sell access to the WiFi network by the hour, by the day or by the month – traditional ISP services. There are customers in any town who would prefer WiFi to more expensive cellular data if you can create good enough coverage. You can sell this to individuals or in bulk to large employers in a town that have employees who work outside. The other traditional revenue opportunity it to sell dedicated hotspots to restaurants and other businesses that want to offer a branded hotspot for their customers. This will require that you (or somebody) provide a broadband connection to that customer to feed the hotspot.

There are two revenue opportunities on the horizon today. The first is to offer WiFi phones. These phones are being offered today in two ways. First, there is the WiFi-only phone like Cablevision is offering and that only works on WiFi. Cablevision prices this at $9.95 per month for an existing cable customer and it’s nearly all margin. But there are several wireless resellers (and now also Google) who sell WiFi phones that will roam to cellular when WiFi is not available.

The primary issue with copying this business plan is that the companies doing it have all created a proprietary system that works only on a specific phone. That is not something easy for a smaller company to work out. There are some cheap Chinese WiFi-only phones available, but if you choose them you are competing against people’s preferences to use an iPhone or a Samsung Galaxy by forcing them to your handset choice. This is not likely to be very popular until it becomes an app that will work on any phone.

The other new revenue opportunity is to sell wholesale WiFi access to others. I know Cisco has been touting this opportunity for several years. But I have yet to hear of anybody who has been able to monetize the idea. The cellular companies love it when customers use their phones on WiFi, but that’s a far cry from them being willing to buy time on your network on their customer’s behalf.

My conclusion of all of this is that it looks a tough business case today to build a citywide WiFi network. Right now the network hotspots are too expensive for a mass deployment. But there are vendors working on lower-cost hotspots. It also makes sense to wait until Hotspot 2.0 is fully fleshed-out and functional rather than buy a network with undefined future upgrade costs. And on the revenue side, while it sounds interesting to sell bulk WiFi, I have a hard time recommending this as a business plan unless you have presold some large customers like a utility or other carrier to buy bulk access to your new network. I have always been leery of ‘build-it-and-they-will-come’ business plans and I could recommend this only if there is a clear path to monetize it.

Expanding Public WiFi

Wi-FiComcast began the process last week of turning home WiFi routers into public hotspots. They announced that they were turning up 50,000 home routers in Houston, and that this was going to be followed up nationwide with millions of home routers being opened up to allow access to anybody with a Comcast password or anybody willing to buy bandwidth by the hour.

Comcast says that this is being implemented by opening up a second channel in each router so that external users won’t be using the same bandwidth as paying customers. Comcast promises this won’t degrade the bandwidth purchased by customers. Interestingly, they are going to match the bandwidth from each public channel to match the home bandwidth that has purchased.

I must say as a Comcast customer that this feels both good and also a bit scary. It certainly would be convenient when walking around my town to be able to be connected to Comcast WiFi and not use cellphone data. And it certainly could make it convenient for me to go outside and still be able to work on my laptop or tablet. So for someone like me who is always connected this sounds promising.

But as the owner of a Comcast router of my own I am somewhat worried by the security aspects of this. There is a nagging part of my brain that tells me that even if this is done on separate channels that there are people smart enough to hack this. So I worry that this could give somebody access into what I am doing inside my own home on my own network. I hope I am wrong about this, but it seems a lot easier to think somebody could hack me when starting inside my router rather than having to start outside of it. Comcast does offer the option, for now, of turning off the second public channel of your router. I’m not sure what they’ll do if everybody chooses that option.

One thing to remember is that this is not Hotspot 2.0 which is a suite of technologies that is going to let people automatically connect to WiFi routers as they move from place to place. That new technology is supposed to come with new security features that will make it safer to be on a public WiFi router. But Comcast is still deploying current WiFi technology, and a user just has to log on one time to any Comcast hotspot and they will then automatically log on to other hotspots with the same password and ID.

Certainly as I move around town on Comcast hotspots I am going to use the same security measure that I would use at a Starbucks. I won’t log into financial institutions or make credit card purchases. Those are common sense security measures to take when sharing a hotpot with people you don’t know. But over the last few days I read a lot about hotspot security and there are a lot more dangers out there. A smart hacker can get into your computer and dig out whatever data you have stored including passwords to accounts and other damaging data. So this is the scary side of using Comcast hotspots or allowing my home router to become one of them.

I also now have to worry that I am giving Comcast the same sort of data about my whereabouts that I give to the cell phone companies. Comcast will be able to follow me as I move around and the knowledge of when and where I go has to be worth something in terms of profiling me.

Why would Comcast do this? They began deploying public hotspots in areas where they are having significant competition with Verizon FiOS. For example, it’s been reported that you can go almost anywhere on the Jersey shore and stay connected to Comcast. So in those kinds of markets it is a feature and a service that they think gives them a competitive edge.

But I see less advantage from deploying this in the average suburban neighborhood. It makes a lot of sense in downtown areas, even in small towns, where WiFi can be deployed where people shop and dine and congregate. But a WiFi signal doesn’t propagate very far from any one hotspot and so in suburban areas one can imagine your cell phone gaining and losing WiFi access as you take a walk. I shudder to think about what that is going to do to the battery on my cell phone as it constantly searches and adds and drops WiFi connections.

The big beneficiaries of this are the wireless companies and one can speculate that Comcast has figured out a way to charge them something for WiFi offload of cellular data. If not they are missing an opportunity. I know that Cisco and other manufacturers have been talking up WiFi offload as a new business line, but I have not yet heard of any specific deal being struck anywhere for this as a revenue generating service.