19 Comments

Summary:

The demand for mobile broadband will surpass the spectrum currently available some time in mid-2013 according to an analyst. Without new spectrum, the demand for bandwidth consuming services will lead to a crappy user experience, or heavy-handed pricing and limitations on mobile application from carriers.

The demand for mobile broadband will surpass the spectrum available to meet it in mid-2013, according to Peter Rysavy, a wireless analyst. In a report on the looming spectrum crisis that was sponsored by Research in Motion for the Mobile World Congress in Barcelona, Rysavy explains how the demand for bandwidth-consuming services used by more and more people will lead to a crappy user experience, or heavy-handed pricing (GigaOM Pro, sub req’d) and limitations on mobile application from carriers absent new spectrum allocations.

He begins with data showing the increasing demand on mobile networks and lays out how much bandwidth a variety of services need, from up to 12 kbps for voice calls to 1-2 Mbps to stream HD YouTube videos. After illustrating the capacity crunch, he starts to tie it to spectrum. In many cities, carriers have 55-90 MHz of capacity, only some of which is allocated to data services. Even today, some providers such as AT&T have indicated they’re using up to half of their spectrum resources in heavily populated markets.

But more spectrum is only part of the issue. Operators have options ranging from more cell sites (either towers or even femtocells) to data offloading (sub req’d) to next-generation radio technologies such as Long Term Evolution to even antenna optimization. I detail many of these measures as well as AT&T’s spectrum shortage in a GigaOM Pro report published today called Everybody Hertz: the Looming Spectrum Crisis.

The Federal Communication Commission and Congress are playing a role in the spectrum issue, but it would be insane to think that handing over more airwaves will be enough, or that it will happen quickly. Useful spectrum for mobile broadband isn’t an infinite resource, so everyone from the developers building more resource-aware mobile applications to the folks in Washington allocating the spectrum and dictating regulations around mobile broadband will have to work together in order to make sure our desires for the mobile web are met.

  1. Is that top chart Gigabytes of spectrum demanded per user per month? In the U.S.? If so, he’s got the value for this year pegged somewhere below 100MB. Also, those curves there look pretty pedestrian — the demand is exponential, and the supply is linear, ignoring the effects of moore’s law and other factors driving down capex.

    See: http://fastnetnews.com/a-wireless-cloud/61-w/2573-absolutely-no-wireless-spectrum-shortage-in-2010

    Share
    1. Lilly, the chart is measuring Average Demand Per User Versus Average Capacity Per User.

      Share
  2. This will be a good post to remember when Verizon and AT&T launch their LTE services — using just about 20-30 MHz of spectrum depth. How long will that last?

    Share
  3. Why can’t we wait till we get home to watch HD YouTube videos or TV shows. Gotta use those 50″ plasmas for something.

    Share
  4. Spectrum is the real scarce resource for mobile services. LTE without additional spectrum will not solve the coming crunch. Many analysts and pundits in the press overlook the physics of the mobile space.

    Share
  5. ATT has actively fought municipal wifi for years. What looks like competition amoung the carriers is actually not-so-well- cloaked collusion. The FCC may need to step in to prevent egregious 3G and 4G bandwith demand surcharges. WTF – I can’t tether a jailed iPhone but you can Sling on ATT?

    Share
  6. It’s important that any discussions begin from the assumption: that Gray-Hoverman antennas and “white space” lower the cost of community-based local broadband infrastructure for mesh networks to almost zero. Once every community has local broadband network infrastructure in place, competition for affordable Internet access becomes a reality. Any business “hot spot” becomes a candidate for local/regional ISP, much in the same way we now have local/regional radio stations.

    Share
  7. While it’s true that the supply of spectrum is finite, and the likelihood of the FCC being able to allocate more spectrum for mobile broadband use will probably take many years, there are alternatives. One solution to the problem of spectrum scarcity is to help everyone more efficiently use existing spectrum, otherwise known as creating “universal spectrum access”.

    By creating a single point for anyone to locate and access available spectrum, both licensed and unlicensed, it will provide wireless users with a variety of spectrum allocation solutions, such as buying, selling, leasing, time-sharing, partitioning, or disaggregating spectrum, or even utilizing white space for various applications. The result, better utilization of bandwidth and enhanced optimization of network performance, while increasing the amount of spectrum available.

    Share
  8. The report is very interesting & well worth reading.

    http://www.rysavy.com/Articles/2010_02_Rysavy_Mobile_Broadband_Capacity_Constraints.pdf

    However, the situation is actually rather worse (or rather more immediate) than you suggest.

    Those graphs are average across the country, but as Peter says:
    “Of particular significance is that usage from cell site to cell site is not uniform, and varies based on the location of users and their behavior. Even if a network has sufficient “average” capacity, these variations will result in some cells being overused and some cells being underused.”

    This situation is happening now.

    Vodafone recently reported that 7% of its cellsites are saturated, and people in Manhattan or San Francisco are also seeing that problem.

    Some of the solution will be technology (femtocells, better use of WiFi, more spectrum, better signal processing) but I do suspect economics need to be part of it too (differential pricing).

    We accept that in other domains (first class costs more than coach; travel at Thankgiving costs more than other times; Feex costs more than USPS) so why shouldn’t wireless do this?
     

    Share
  9. Spectrum crunch (for mobile) is inevitable as more content/services are expanded to mobile with proliferation of smartphones and people’s insatiable want for more bandwidth/more content at acceptable quality of experience. Carriers will make network/content investment as long as there is a profitable opportunity. Consumers will ultimately decide what price is acceptable for all those services. I am amazed by how much more I spend on communication/pay TV today versus 10 years ago. My monthly home communication (entertainment – pay TV) bill has increased about 5x over 10 year period.

    Share
  10. [...] Spectrum Shortage Will Strike in 2013 See All Articles » Broadband Fans, We Have an Innovation Problem [...]

    Share

Comments have been disabled for this post