If you are confused with allÂ the hype around LTE then you may be in good company. AT&Tâ€™s Chairman and CEO, Randall Stephenson, whose company is rolling out LTE right now, wasÂ reported by mocoNews.net to have made some comments on LTE speeds that, if correct, have him firmly in the same â€˜confusion campâ€™ as the rest of us.
When outlining how his company would overcome issues of poor coverage areas or holes in the LTE network he stated that AT&T would upgrade its entire 3G footprint to support HSPA+ technology that can deliver 2 to 3 Mbps. Then, it will layer on LTE, which can offer 7 to 10 Mbps. If you canâ€™t deliver broad ubiquitous coverage of 7 to 10 Mbps, then falling back to 700 kilobytes per second â€œis an unacceptable experience,â€ he said. â€œTo fall back to two to three [Mbps] is really important.â€
If thatâ€™s the best speed Stephenson can expect from his multi-million dollar LTE network and that he has to rollout HSPA+ as a comfort â€˜fallbackâ€™ then why bother with LTE at all?
According to the engineers,Â targets for LTE should include download rates of 100Mbps, and upload rates of 50Mbps for every 20MHz of spectrum. Telstra is already achieving speeds on its HSPA+ network ofÂ 42Mbps and trials of its LTE network producedÂ 100Mbps over a 75Km test cell.
Of course, these top speeds are hypothetical when it comes to the man in the street who is more concerned with reasonable coverage at any speed than none at all. Yet, even the average speeds on most HSPA+ networks exceed those that Stephenson is quoting for his LTE network.
LTEâ€™s advantages are not all speed related, however. It introduces technology such asÂ OFDM (Orthogonal Frequency Division Multiplex), MIMO (Multiple Input Multiple Output) and SAE (System Architecture Evolution) that make more efficient use of spectrum (and we all know how important that is).
Stephensonâ€™s theory also works assuming we see dual 3G/LTE chipsets and mobile devices commercially available. These are currently predicted only for the latter part of 2011. I might be missing the point here, but does anybody really need a 100Mbps handset.
Looks to me like LTE is all about mobile computing fed by dongles and devices like iPads and notebooks. Not sure if thatâ€™s the market that will generate revenue commensurate with the bandwidth take-up. One wonders if the headlong rush into LTE has been driven by the marketing people with apparent scant regard for costs/benefits analysis or returns on investment.
Whether driven by the fear of being left behind, or the need to keep shareholders happy, CEOs like Mr Stephenson continue to make big investments in future technology they themselves seem unsure off. Our industry has always had the â€˜luxuryâ€™ of strong cash flows and high revenues to fund the big technology rollouts but the signs are that this may not be sustainable. Not all network investments have returned the expected increases in revenue to justify them, especially from data traffic.
What does it matter, those spending big today will probably not be around when the real figures are tallied up? However, you’d think somebody in that position and on that salary would at least know how fast his network is planning to run, at the very least!
I also asked myself the same questions when LTE started, but then I realized that LTE is not about speed but about cost of operations of SPs. AT&T loses today for every new iPhone subscriber with unlimited data plan because 3G architecture was not designed to support massive data transfers. Assuming the world (at least part of it) is moving to high data usage with unlimited plans or at least plans that support current pricing, SPs do not have a choice and must upgrade the network else they lose on every new subscriber.