Seybold's Take: Should Vodafone migrate DSL customers to LTE?

Andrew SeyboldVodafone Germany announced that it will attempt to migrate its DSL (wired) customers to its LTE wireless broadband network. The move is designed to save Vodafone Germany up to €500 million a year that it now pays to Telkom Deutschland for using its fixed network (as reported in FierceWirelessEurope on Monday, August 22, 2011).

This story got me thinking about the roll-out of LTE worldwide, the need for more spectrum, and how this will all fit together. In the United States, network operators are saying that because of increased demand for wireless broadband services they need more spectrum. The Federal Communications Commission (FCC) agrees with them and in its broadband report to Congress, the FCC stated it would "find" 500 MHz of spectrum for broadband services over the next 10 years. There is no doubt in my mind that more spectrum is needed for wireless broadband and I have to wonder if moving wired customers onto a wireless broadband network is such a great idea.

Wireless broadband demand is increasing by more than 50 percent year-over-year, even with the spectral efficiencies of LTE. Is there enough bandwidth on a given network to support those who are currently connected from their home and/or office via a DSL connection? LTE, like all cellular network systems, is designed to handle mobile traffic as opposed to fixed-location traffic. When cell sites are designed they are designed for a specific and well-calculated load factor that is based on mobile devices moving in and out of the cell or cell sector on a regular basis. DSL customers, on the other hand, operate from a fixed location.

What this means is that the load on a specific cell sector is nearly constant, that is, customers use the same cell sector most of the time. If they are heavy users of data, Internet, and streaming video, that particular cell sector will have less capacity for mobile users entering that sector. DSL is inherently always on and always connected while mobile customers are more likely to use their LTE devices in spurts. All of this points to having to change some design parameters for cell sectors that have DSL-like fixed connections.

As an example, if a cell sector has a data capacity of 30 Mbps and within that cell sector there are 5 new fixed-site customers, and each of these customers is streaming a video at the same time, assuming the video stream is 2 Mbps (some video is less but HD can use up to 5 Mbps), then one-third of that cell sector's capacity is being used by fixed users. The design of an LTE system must take into account peak loading and while voice services tend to peak before and after work hours, fixed usage for data is more constant and around the clock. Typical cellular systems are designed to operate at about two-thirds capacity leaving additional capacity for those entering a cell so they aren't disconnected or cut off when moving from one cell to another. The reality is that these five fixed-use customers can be using one-half of the bandwidth and capacity allocated within that cell sector.

This is certainly not an insurmountable problem, but it will require some of the engineering as well as operational parameters to be adjusted to take this new type of fixed user into account. I have to wonder how many applications and devices will be coming online that will be fixed in nature and therefore take a bite out of a cell sector's total capacity. I have not been able to determine how many users on the Verizon LTE system are fixed-location users, but some of the applications and devices, including surveillance cameras, are fixed-location users. If surveillance cameras are on all of the time, and they usually are, and a given business is using 5 or 6 of these cameras, the impact on the LTE cell sector could be significant. Surveillance cameras, on the air 24/7, change the amount of bandwidth needed depending on whether they are looking at an area with no movement or an area where there is movement, people walking, etc. Therefore, the bandwidth requirement for these cameras changes constantly, but in calculating their load on a cell sector, you have to figure the worst-case bandwidth use AND you have to know they have been installed on the network.

Vodafone will know where each DSL customer is located. However, there is no way to determine other fixed-location usage ahead of time. The company that installs cameras does not have to register with the network operator, only subscribe to the service. Perhaps tiered pricing will help keep these types of uses to a minimum, and if these users work with the network operator for special rate plans, at least the network operator's sales organization will know but it is not clear that this information will be sent up the chain to the network engineers who design these systems.

One way of off-loading some of this bandwidth would be to install femtocells at a location where cameras are to be used and then backhaul the signal to the network for distribution, but this will take coordination. Today, for every camera in service in a fixed location, an uplink and a downlink are consumed-the uplink from the camera to the network and the downlink back to the monitoring location that could be within the same location or at a remote site. If it is the same location, the amount of streaming video traffic will essentially be doubled for each camera in service.

LTE is the most efficient wireless broadband technology available today. It is evolving and network capacities will increase over time, but the real question is whether or not they will increase fast enough to keep up with the demand. Even if the FCC does "find" more spectrum for broadband wireless services it will be 3-5 years before network operators can build it out and deploy it. Wireless spectrum is a shared resource, and it is limited on a cell sector basis. Using what we have wisely is the best thing we can do today. I have to wonder if fixed-location LTE is the best and highest use of our constrained spectrum resources.

Andrew M. Seybold is an authority on technology and trends shaping the world of wireless mobility. A respected analyst, consultant, commentator, author and active participant in industry trade organizations, his views have influenced strategies and shaped initiatives for telecom, mobile computing and wireless industry leaders worldwide.