IT Jobs
Super-fast wireless breaks cover
Mystery about how it works.
By Peter Judge | Techworld
Published: 17:43 GMT, 06 March 06
Another novel wireless modulation scheme has broken cover - but it is saying even less about its technology than its peers.
Singapore-based Photron Technologies has announced it is making chips for what it calls Ultra Spectral Modulation (USM) technology, which it claims can allow data transmission at extremely high rates for a given bandwidth, and at very low power.
Photron claims its technology can achieve rates of 100bit/s per Hz of bandwidth allocated, and that it is packaging it in demonstration chipsets using FPGAs (Field Programmable Gate Arrays) in the hardware platform. Photron claims this makes it around 25 times more efficient than the leading 3G technology in the US, EV-DO.
These claims are strikingly similar to those of xG Technology, whose xMax technology appeared, in a vendor-controlled demonstration last year, to achieve 7.4Mbit/s per MHz per Watt, through technology it described as "single cycle modulation".
Readers will have noticed that, even if these claims were demonstrated under verifiable conditions, they would be hard to compare since the units quoted are different.
Photron's site has a White Paper, by Photron's "technology adviser", Larry Garcia, who is also "senior technical advisor" for Aerotelesis, a Los Angeles company that promises to implement USM. The paper doesn't explain USM, unfortunately, but hints that it uses "the correct mix" of phase modulation (PM) and amplitude modulation, (AM), and an "ideal digital filter".
The other uncertainty in any wireless technology is range, and Photron doesn't address this at all. While it compares its technology with long-range 3G technology such as EV-DO, others have commented that it may have a much shorter range.
Add your commentComments
Phil | Published: 01:29 GMT, 07 August 2007
It would really be nice if communications journalists could learn to use the Shannon Channel Capacity formula to screen out absurd claims like these. A spectral efficiency of 100 b/s/Hz with a typical receiver and noise level would require a *received* energy per bit of nearly 15 kilowatt-hours! Such claims can be quickly dismissed out of hand. Photron and USM are closely related to Harold Walker's "VMSK" nonsense. Although they seem to work on the bench, that simply proves they are much wider than claimed. Every one of these supposedly "ultra narrowband" signals actually consists of a wideband, weak data signal plus a strong, information-free spectral line that serves as a distraction and wastes most of the transmitter power. These schemes are totally worthless, but they keep reappearing and confusing the unwary and wasting investor money.