xMax validated? Not …

It is a basic principle of science, as well as common sense, that extraordinary claims require extraordinary proof. Given the extraordinary claims made about xMax, such as: “xMax can use 1,000 to 100,000 times less power than comparable transmission technologies”, one would hope for plenty of concrete information about the performance of xMax, which would validate these claims. Only a very limited numerical information about the performance of xMax is currently available. As will be discussed next, the available information does not validate the key claim of reduced power, and raises more questions than it answers.

There have been several demonstrations where xMax was used to transmit data between two points. While such demonstrations are certainly very important in establishing that xMax does in fact work, a simple link budget analysis shows that there is nothing remarkable about the performance that is being quoted. See the discussion of the Florida demo and the WiFi experiment. Under the conditions of these experiments, the SNR at the receiver was more than sufficient for a conventional communication system to deliver the same (or better) data rates. In other words, the “good news” is that the results quoted are entirely credible. The “bad news” is that they are entirely unremarkable. A person knowledgeable about communication systems will understand that these demonstrations do not establish that xMax can operate with much less power than conventional systems. In fact the choice of experimental parameters strongly suggests that xMax performs quite similarly to conventional systems.

XG Technology has published on its web page two “technology documents”. The first is a report by MICOM Laboratory
which compared the xMax system to an unspecified 900 MHz system and concluded that xMax was able to operate with 18dB less power than the comparison system. Unfortunately this report is seriously flawed . The most glaring issue is that the report does not specify the system used for comparison, just refers to it as a “commercially available 900 MHz radio”. Given that the whole point of this report is the comparison of the two systems this omission is simply inexcusable. There are $50 radios and $5000 radios. Both work, but have very different performance characteristics!

The information given about the comparison system is completely inadequate: all we know is the operating frequency, bandwidth and peak power. Crucial information is missing, to give just two examples: (i) What was the actual bit rate at which the radio was operating, compared to the bit rate of xMax? If the comparison system operated at a higher data rate it would naturally require a higher SNR. But then this is not a fair comparison. (ii) What are the noise figures of the receiver of the two systems? Radios differ significantly in receiver sensitivity. A cheap radio could easily have a sensitivity that is 10dB (or more!) worse than a top-of-the line radio. Interestingly, the test procedure and presumably the choice of the comparison radio, were specified by XG Technology, and not MICOM labs. In any case, without adequate technical information about the comparison system, this test provides no useful information about the relative performance of xMax.

There are, of course, standard ways to evaluate the performance of any digital communication system. The most basic one is to measure the bit-error-rate (BER) of the system for different signal to noise ratios at the receiver. The BER vs. SNR (or BER vs. Eb/N0) plots for conventional communication techniques, both measured and theoretical, appear in numerous books, papers and reports. Having an independent laboratory measure the BER vs. SNR plot for xMax would immediately establish its performance relative to existing techniques in a direct and convincing manner. This is a simple and completely routine test. It should be noted that this type of performance curve provides no information about the design or the internal workings of the communication system. Thus, concerns about revealing intellectual property related to the communication system are not a legitimate reason for avoiding publication of such results. It is left as an exercise to the reader to suggest other reasons for not publishing this type of performance curve.

The second technology document has a statement by one of the advisors of the company. In page 2 of the technology document there is a table comparing xMax to GSM and CDMA. The table shows the spectral efficiency of these systems. Spectral efficiency, the ratio of the data rate to the bandwidth used, measured in units of bits-per-second per Hz, is a standard parameter used to characterize communication systems. The table shows that xMax is significantly less efficient than GSM and CDMA. Next the table shows a parameter which is not standard: bits-per-second per Hz per watt. This parameter is computed by dividing the spectral efficiency by the power of the basestation (!) and is shown to be much greater for xMax than for GSM and CDMA. This parameter is not meaningful unless all the basestations being compared operate under the same conditions (same data rates, same bit-error-rate, same bandwidth, same antennas, same pathloss, same interference/noise conditions)! But this is clearly not the case. A typical GSM/CDMA basestation is designed to operate under much harsher conditions than the conditions in the xMax demo. Cellular systems do not typically use 850 ft antenna towers and are expected to operate under heavy shadowing loss and large fading margins and in the presence of significant intra-cell and inter-cell interference, not present in the xMax line-of-sight demo. Had the basestation powers been compared under identical operating conditions the bps/Hz/watt numbers for xMax could not be significantly larger than the numbers for GSM/CDMA. In other words, the bps/Hz/watt numbers in this table do not provide a useful comparison.

By way of analogy consider a table comparing the miles-per-gallon (MPG) for station wagons made by two different automobile manufacturers. The MPG for the first type of car is measured when the car is completely unloaded with only the driver present, and is driven downhill on a modest incline so it can coast most of the way. The MPG for the second type of car is measured when the car is completely loaded with passengers and baggage, is pulling the largest allowed trailer also fully loaded, and is being driven uphill on a steep winding mountain road. Comparing these two different MPG values provides no useful information about the relative performance of the two types of cars. For a useful comparison the cars (and the basestations) must be tested under identical conditions!

Neither the demos nor the technology documents provide any kind of proof of the key claim that xMax can operate with significantly less power than comparable transmission technologies. Simple and direct tests are available to verify or falsify this key claim. The results of such tests do not seem to be available.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s