Close Message
None

Entities Mentioned



  1. Parsing The Need for Speed

    By Josh Adelson

    As mobile broadband becomes more popular, people are asking the natural question, just how “broad” is it?  That is, how fast is the connection?

    Recently Mike Hibberd of Informa reported on the results of a study that gave a sobering answer: not very fast.  Test firm Epitiro found that speeds achieved on HSPA and HSPA+ networks were about 75% below the operators’ advertised rates.  There will likely be debate about this study, but taken on its face it begs the question, how fast does mobile broadband need to be?

    To answer this, it’s helpful to reflect back on a dozen years experience in wired broadband services.  Before broadband, there were analog modems with published speeds like 28.8 kbps.  Cable and DSL represented a step function, initially launching at speeds of half a megabit or so.  Those numbers caught everyone’s attention and even more so did the dramatically different experience.  Since those early years, cable and DSL have steadily grown faster, but today the actual numbers are largely ignored.  Verizon’s FIOS web page shouts that its service is “blistering” and even “mind-bending”, but you need to read much smaller print to learn that data rate is an impressive 50 Mbps.  From the user’s perspective, we rarely even know whether we are getting 2, 6 or 10 megabits.  We only care that our web pages load fast and our videos play without halting.

    Perhaps the reason wired broadband services have drifted away from their fixation on speed measurement is that, while the speeds do vary, they are consistently high enough to ensure a satisfying user experience.  This predictably good experience is the foundation for so many of the services that people now enjoy.

    Wireless services have a harder time reaching this consistency threshold.  Unlike the signals that traverse fiber and copper cables, radio waves are severely challenged by distances, by hills and walls, and by the inherent scarcity of spectrum that results in chopping up the bandwidth.  Femtocells will help bridge this consistency gap.  In Airvana’s recent home user study, the result that grabbed the headlines was “5x data speed vs. macro network”, but just as important was the fact that the femtocells achieved greater than 1 mbps 95% of the time, while the macro network (measured indoors) achieved this speed only 5% of the time.

    That femtocells can provide wireline-like consistency of experience is no accident.  The femtocell uses the wireline network for all but the last few meters of its connection.  This strategy makes sense in light of the high penetration of wired broadband service, and of operators’ propensity to market mobile broadband service as an add-on, rather than a substitute, to wired broadband.  See for example Ovum Research (“Mobile Broadband is Complementary to Fixed”), and comments by Virgin Mobile.  Femtocells also eliminate the need to subdivide spectrum, since each cell is shared by just a few users.

    When mobile broadband services can take full advantage of the consistency enabled by femtocells, network speed may be considered a topic too obscure for all but the most technical of industry journals.

    Bookmark or Share this article


    Related Articles

Login to comment.

 
HomeAboutBlogAirvana Login Register Powered by HiveFire