So long, and thanks for all the fish

Starting this week and for the foreseeable future I will be busy doing real wireless work, and will no longer have the time needed to keep this blog going. There is a lot of “wireless fiction” out there, caused by widespread ignorance about the basic principles of wireless communications. The only real solution is to get more people educated about this great technology, what it can do, as well as its limitations. However, based on my experience with writing this blog, this does not seem to be a useful way of trying to accomplish this goal. It has been a fascinating experience, but I prefer to spend my limited time doing other things.

Thanks to all the readers of the blog, and to all those who sent me e-mails. I can still be reached by e-mail at I wish everyone the best of luck. May you all find truth and happiness.

The xMax rollout

Latest news about the xMax commercial rollout:

Hmmmm … part III

The old story

And now a smooth transition to the new story 4/25/2006.

It’s deja vu all over again.

The long and the short of it

In an earlier post (xMax and its deployment – part II) I mentioned that if one puts a Wi-Fi access point on a tall tower it could be received at long ranges, and suggested the possibility that a dual Wi-Fi / xMax phone may function as well (or as poorly) using Wi-Fi as it will using xMax. This was in the context of my discussion of potential difficulties in evaluating xMax performance in an actual deployment (which, as we all know, will happen any day now).

This resulted in a flurry of e-mail discussing the “well known fact” that Wi-Fi has a very short range of a few hundred feet at best. As a followup here are some recent articles which these readers may find interesting.

What I do not understand is why Intel, who are pushing very hard the WiMax story, are at the same time working on a cheap alternative, albeit one which will only work in some situations. But that is another story.

A trip down memory lane

Pre – xG Technology


  • 3/2000
  • 6/2000
  • 9/2000
  • 12/2000 ” … Jeff Bobier, director of operations, said, “We could not be more pleased with the level of excitement that our Wish wireless product has generated. Our client partners are delighted with the progress of the deployments in Florida and across the US.””
  • 2/2001
  • 1/2002
  • 2/2003 “… Karason said he found out later that the whole demonstration was fraudulent, “using illegal power, illegal equipment, preloaded stuff on computers that we were told was downloaded over the system, etc., and in fact the Icelandic system did not work at all, as the people running it found out after the presentation — when the ‘magic dust’ had disappeared.””
  • 4/2004 “Surviving Smear” (contradicts some of the information stated in the 2/2003 item)

Finally, an interesting article from the early days of xG Technology 4/2003. Note that

“A small startup claiming breakthroughs in phase-coherent filter modulation will be demonstrating its signal-processing techniques in analog pager bands later this year, as a prelude to offering its technology for a range of wireless and wireline networks.”  “Bobier previously presented details of his work at the Telecommunications Industry Association’s Ventures 2002 show last July. At the time, Bobier claimed that analog paging channels would be able to transmit data at speeds of up to 150 Mbits/sec. He also began work last summer in reducing a typical TV transmission channel for coaxial cable from 6 MHz to 30 kHz.”

In other words he is talking about a spectral efficiency of 200, even bigger than that of Ultra Spectral Modulation (USM) who were going to demonstrate spectral efficiency of “only” 100! (Such a system would require an SNR greater than 600 dB.) Unlike Walker of VMSK fame and the USM folks he apparently managed to get rid of this particular delusion (xMax has a spectral efficiency on the order of 1 which is of course feasible). So he is learning, but still has some fundamental misconceptions about signal and systems, filters, and communication theory as will be immediately recognized by any competent communication engineer reading his patents.

Aren’t there any competent communication engineers on the xG team who could teach him this basic stuff? Or are they carefully hiring people who will not challenge the founder? Surely when they tested their systems he must have realized that the performance claims they have been making all along are based on serious conceptual errors (as xG’s own BER curve shows), not unlike the errors leading him to try and develop a system with spectral efficiency of 200! There is so much to learn in order to put together a working wireless network, and so little time …

As the stomach turns

Hichens Harrison and Co., brokers to xG Technology, published earlier this year a report modestly titled “xG Technology – Flash of Genius” (xG Flash Signal – get it?). The report includes many interesting statements such as “These numbers, taken in conjunction with our model, imply that the company will reach profitability in Q4 this year.”, “Skepticism surrounding these large claims is evaporating in the industry.”, “enable them to offer an xMax™ 4G service by 2008! This is at least 3 years ahead of any rival, none of whom have demonstrated that they have an economically viable 4G solution.”, “the company is on timeline to begin commercial rollout in April”. Right …

But my real interest was in the “technical” part of the report. After all we are talking about a super special revolutionary technology which will change the wireless world as we know it! If you expected some insightful analysis you will be sorely disappointed. The analysis consists of faithfully quoting the material fed by xG Technology. We once again read that

“xG, more specifically its resident inventor, Joe Bobier, has discovered and patented a process called “single cycle modulation”. Single cycle modulation is where individual sinusoidal (a type of wave) cycles of RF energy are modified to carry one “bit” of information. Rival technologies use tens to hundreds of thousands of waves to convey one “bit” of information, so the efficiency of xG’s discovery is of a stunning order of magnitude.”

Stunning indeed! I am stunned that this technical nonsense keeps appearing unchallenged.

“With each additional cycle requiring commensurate power output, it can be easily seen that vast efficiency gains are achieved and this is particularly important for mobile wireless communications, because of the implications for infrastructure requirement. However, the savings in battery power are not an insignificant factor. In addition, low power output “clutters” the spectrum much less, allowing for a greater number of simultaneous users of a network, i.e., enhanced operational efficiency and voice quality gains.”

Oh yeah!

“Hence, xMax™ –
• Can transmit substantially more information at long ranges than other technologies operating at similar power, in a “real world” context.
• Yields significantly greater range at similar levels of power.
• Yields the same range for significantly reduced power.
• Leads to much greater battery life for devices using the technology, such as laptops and – of course – mobile phones. (Initially, the technology will be deployed at similar power levels as rival technologies. At these levels, xMax™’s range is far superior, leading to reduced infrastructure costs. As denser coverage is built into specific areas, power can then be turned down very substantially.)”

Amen and hallelujah!

We also find a table which shows that xMax has a range cell adjustable from 5 to 48 km for 50MBPS peak rate. Adjustable no less – all you need to do is to turn the range knob on the xMax device and the range will magically increase.

Now who is responsible for this technical discussion? Looking at the front and back pages of the report we find that Robyn Harte-Bunting is the estimable technology analyst (I love the name – very P.G. Wodehouse). I was curious as to what Mr. Robyn Harte-Bunting’s qualifications are to pass judgment on this advanced intricate technology. Thanks to a kind reader I found out the answer! It appears that Robyn Harte-Bunting was previously a senior executive at Private Media Group, an adult entertainment company based in Spain. Now this makes perfect sense. After all, we know what all this broadband stuff is really used for, so who is better placed to evaluate a new wireless technology than an ex adult entertainment executive?

Tune in next week to another episode of as the stomach turns

A bedtime story

Gather around children and I will tell you a story about a nice little wireless company called aeroTelesis. Let me read you from a their SEC filing which tells about aeroTelesis “an international telecommunications company which intends to provide next generation telecommunications technologies and services.”

“The Company’s core technology platform is a licensed modulation method known as Ultra Spectral Modulation (USM). USM is a technology that significantly increases spectral efficiency in wireless applications and provides for high-speed and high-capacity networks at substantially lower cost relative to existing wireless technologies. USM is designed to avoid bottlenecks by providing data transmission channels with higher quality and throughput rates than those of conventional modulation techniques.”

“The Company intends to launch its first commercial service, VoIP network services, in the second half of 2004. It plans to deploy USM technology for use in satellite communications in 2005. aeroTelesis eventually plans to deploy wireless networks for mobile voice and data services utilizing USM in the manner that GSM (Global System for Mobile Communications) and CDMA (Code Division Multiple Access) have been utilized to deploy the majority of wireless networks existing today.”

Now Ultra Spectral Modulation is a fascinating technology described for example in US patent 20070002973 where we read that “[0026] The novel narrowband digital modulation technique of the present invention, hereinafter referred to as Ultra Spectral Modulation (“USM”) technique, comprises a return-to-zero modulation technique that uses abrupt phase changes to represent incoming binary symbols in a modulated signal, herein referred to as a “USM-modulated signal.” In one embodiment, the abrupt phase changes occur mid-pulse, i.e., in the middle of a bit period, after an integer number of cycles of the carrier signal.”

“[0027] The USM technique represents all the information in the message signal with the abrupt phase shifts occurring in the USM-modulated signal. That is, all the information conveyed in the message signal may be recovered by knowing where the phase shifts occur or by preserving the positions of the phase shifts during transmission. As a result, transmission may be accomplished by transmitting only a narrow band of frequencies required for identifying the phase shifts, i.e., by transmitting only a portion of one or both sidebands in the USM-modulated signal within a narrow band of frequencies.”

“[0028] The present invention also provides a high-Q, low-tolerance sophisticated digital filter that is able to preserve the positions of the phase shifts in a very narrow band of frequencies. Filtering the USM-modulated signal with the digital filter designed according to the principles and embodiments of the present invention and described herein below produces a filtered signal with a time response that accurately identifies where the abrupt phase shifts in the USM-modulated signal occur.”

“[0036] Advantageously, the communications systems and methods of the present invention enable data rates exceeding 5 Mbps to be delivered through frequency channels as narrow as 50 KHz under a variety of channel conditions.”

In other words USM involves modulating an integer number of cycles, filtering the sidebands to a narrow bandwidth and using a very special filter to recover the signal at the receiver. They are able to achive a spectral efficiency of 100 bits/sec/Hz. Now isn’t this simply wonderful!

The technology was actually developed by Photron Technologies and later licensed by aeroTelesis “Photron Technologies Ltd. today announced that their ground breaking modulation technology USM, is in the final stages of implementation utilizing FPGAs (Field Programmable Gate Arrays) as the hardware platform. This final stage completes the transition from simulations to hardware functionality while confirming the successful application of a unique digital filter developed during 2005. The completion of the hardware demonstration will potentially show USM to be the first communications technology to achieve data transmission rates of 100bps/Hz. The phenomenal breakthrough rates provided by Photron’s Market Changing Technology will allow wireless data transmission rates of 5Mbs through a narrow 50Kz channel. Current wireless technologies require channel bandwidths that are 25x larger.” We also read that

“aeroTelesis Inc. and Photron Technologies Ltd. today announced they have started development on prototype hardware for Photron’s Ultra Spectral Modulation(USM)(TM) technology. The prototype hardware will be implemented in Field-Programmable Gate Arrays (FPGAs), with an Application-Specific Integrated Circuit (ASIC) design to follow later in the year. Developed by Photron and licensed exclusively to aeroTelesis, USM redefines the next generation of wireless communications by providing wireless data transmission at rates of 100bps/Hz (bits per second). Current technologies produce rates of only 4 bps/Hz or less.”

“Successful end-to-end MATLAB(R) simulations demonstrating USM’s ability to transmit data at 5Mbps (megabits per second) through narrow channels of 50KHz has established the feasibility of the core technology. The successful simulations were the foundation for the all important milestone of completing the critical hardware architecture necessary for the prototype hardware design.”

You can read about the many accomplishments of USM in many other articles such as here and here. So what happened to this wonderful technology you ask? Sadly not much. If you look at the history of aeroTelesis you see that on 22 October 2004 they were trading at $8.03 with a market cap of approximately $800 Million, but today they are trading at a faction of a cent with a market cap of $690,000. What went wrong you ask? Well, it is getting late and it is past your bedtime, so I will tell you the rest on another day.

What is it Jimmy? You say this story reminds you of xMax? You mean the remarkable performance claims, the modulation of an integer number of cycles, the very special filter, the VoIP network, the MATLAB simulation and so on? Well, I know that TechWorld, VONMag and some others have tried to make a connection, but they are wrong, absolutely wrong. There is no similarity whatsoever. None.