Mobility

iMAT: MIMO without multiple antennas

iMAT (Isolated Mode Antenna Technology) is a new antenna technology that improves antenna gain and receiver sensitivity in a MIMO environment and it does it with only one antenna.

Antenna systems are not normally a preeminent topic when it comes to discussions about wireless equipment. That attitude is completely understandable; antennas are just not that interesting to most people. To some, including myself, RF propagation and antenna technology are very worthy discussion topics. My fascination started 40 years ago when I earned my first amateur radio operator’s license. I still remember the excitement of talking to as many fellow amateurs as I could on an old home built CW transmitter. Not having a great deal of money for new equipment, I focused on improving my antenna farm. It did not take long for me to notice the difference a good antenna made. Poorly designed or poorly built antennas—I learned the hard way—can literally destroy transmitter RF circuits. Whereas a well built antenna afforded me the opportunity to communicate with other amateurs anywhere in the world.

Antenna systems finally get some respect

Wireless equipment manufacturers will always have the challenge of trying to equal or surpass the capabilities of wired voice or data networks. Helping to close that gap are recent innovations using multiple—same frequency—antenna systems. Researchers and equipment manufacturers now understand that using MIMO antenna technology means greater receiver gain, increased data rates, larger network throughput, and improved reliability through antenna diversity. It all sounds good, but as that saying goes “nothing’s ever free” applies here as certain challenges surface when using MIMO antenna systems.

EM coupling and isolation

One of the more important challenges facing RF engineers developing MIMO antenna systems is how to negate the effects of a phenomenon called electromagnetic (EM) coupling. EM coupling according to Wikipedia occurs when a change in current flow through one device induces current flow in the other device. The two devices may be physically contained in a single unit, as in the primary and secondary sides of a transformer, or may be separated as in the antennas on a transmitter and a receiver. One other related term that appears in conversations about EM coupling is EM isolation which is the opposite of coupling. EM isolation is generally achieved in two ways, either by increasing physical distance between RF radiators or by creating a barrier with RF shielding between the two radiators.

It is important to understand that intentional EM coupling is a good thing when it is occurring between multiple antenna/radio systems that are supposed to communicate. On the flip side, unintentional EM coupling between two or more antenna/radio systems not meant to be communicating is detrimental to all of the involved systems. The receiving radio has to deal with additional in-band interference and the transmitting radio loses efficiency as some of its valuable RF output is not reaching the appropriate antenna/radio combination. Figure 1 represents the two extremes of coupling and isolation.

figure-1.jpg

Figure 1. Antennas on the left have coupled near-radiation fields and highly correlated inputs, coupling energy from one into the other. The ideal situation is shown for the antennas at right where little radiation is coupled between antennas, and the correlation coefficient is zero. Courtesy of SkyCross.

Knowing this, one has to wonder if it is such a good idea to have all those antennas so close together on a single radio. Wouldn’t most of the RF energy from one antenna just radiate to the antenna only a few centimeters away? This is certainly one area of concern and compromise, as it forces MIMO equipment developers to delicately balance improved data rates and signal reception against the loss of RF signal performance due to unintentional EM coupling.

Metamaterials to the rescue

A recent post of mine “Metamaterial: the unnatural composite is transforming wireless technology” describes how antennas made from metamaterial are revising RF propagation theory. Metamaterial composite antennas are physically smaller yet electrically more efficient. This creates a win-win situation as it allows more antennas to be located in the same physical space, yet they are less likely to be affected by unintentional EM coupling. Metamaterial antennas can also be constructed to focus the RF wave front away from nearby antennas which also reduces the effect of unintentional EM coupling.

Finally getting to iMAT

Over the past few years I have been following a very interesting company called SkyCross. What initially aroused my curiosity was SkyCross being picked by Samsung to produce antenna systems for the popular Blackjack and Blackjack II cell phones. Being selected is fairly significant especially since SkyCross has only been around since 2000. The company’s forte is described by Chris Morton, CEO:

“SkyCross enables manufacturers to make any wireless device a reality. Whether you're designing a low-end, cost-conservative product or a high-end device packed with features, SkyCross has the right solution for you.”

The newest project at SkyCross is iMAT (Isolated Mode Antenna Technology) which is the topic of this post and what can be called a radical departure from typical multiple antenna methodology. An iMAT antenna system is a single antenna with multiple feed points that has performance characteristics similar to a MIMO antenna system. The significance of this can be seen immediately. Having only one antenna mitigates the power-robbing effects of unintentional EM coupling. It also gives equipment and PC board designers more freedom as there is no need to reserve space for several antennas and their supporting PCB components.

Researchers at SkyCross use a rather complicated metric called the correlation coefficient to determine the ability of a multiple antenna system to produce independently received signals. Complete signal overlap which is bad, results in a channel correlation coefficient of one. The graphs in Figure 2 show that the conventional two antenna system’s coefficient is more than double the coefficient of the iMAT antenna system.

figure-2.jpg

Figure 2. Illustration of the advantages of the iMAT single antenna over multiple individual antennas showing enhanced efficiency and reduced inter-element coupling. Courtesy of SkyCross.

How is this significant?

Using a single antenna to achieve multiple antenna results has to excite equipment developers. It gives them the chance to achieve MIMO performance levels in devices that have space or size limitations. As I understand there are no real frequency restrictions, so iMAT will support all legacy RF protocols and be available for use with HSxPA, LTE, WiMAX, and 802.11x

I had the good fortune to ask Mr. Paul Tornatta, managing director for SkyCross USA and VP of Operations a few questions about iMAT. I was really curious to learn if my Blackjack phones used iMAT technology. Mr. Tornatta replied:

“Regarding the Backjack and Blackjack II; these designs do not use iMAT, as the design work was completed some time ago. We are currently working on Smartphone implementations, but will not make a specific announcement for several months.”

I then made mention that iMAT must only work if the multiple feeds were on the same frequency, but was pleasantly surprised when Mr. Tornatta corrected my error:

“The comment about iMAT only working for multiple multi-feeds at the same frequency is not quite right. Certainly a key feature is the ability to run multiple feed points from the same multi-band feed structure, at the same frequency, for diversity and MIMO applications, but another key feature is the ability to create isolation nulls to other bands as well.

An example is GPS. In many of our diversity designs, we include GPS on the Diversity antenna port. In this case we are able to create an isolation-null to prevent the radiation from the primary antenna port from effecting the GPS receive function.

We are also doing work using iMAT to isolate transmit and receive functions normally associated with a single antenna (for GSM). By creating isolation between transmit and receive frequencies, we can relieve the filter specification for down-stream components with a potential cost saving. So in this case, the two ports of the antenna structure are at different frequencies.”

To their good fortune, SkyCross readily admits to discovering the underlying technology behind iMAT while trying to solve a completely different issue. As I understand it, an engineer was trying to place several antennas on a miniature circuit board and accidentally created a path between two adjacent antennas. When running some initial tests the engineer determined that the two antennas had a high degree of isolation. After some development SkyCross realized that they now knew how to develop a multiple feed single antenna with efficient RF characteristics.

Final thoughts

It appears that iMAT technology has a very bright future. The next step would appear to be determining how to simultaneously transmit on different frequency bands or even different protocols.

About

Information is my field...Writing is my passion...Coupling the two is my mission.

25 comments
seanferd
seanferd

Your articles which I have recently read have made up for three months of relatively lame science magazines. You have renewed my interests in antenna physics and technologies. I remember that when my interest in antennae was first sparked: An article in a paper (The Electron) had mentioned that no one knew how antennae *really* worked at some fundamental level. So, thank you for your in-depth articles. I'll need to read all of these several more times...

w2ktechman
w2ktechman

I'll have to admit Michael, that this is an interesting read. Thank You (again) for bringing some quality reading material. As for questions, the first ones that I had, you already covered. But, how will multiple antenna's in close relation, be able to deal with each other? As in an 802.11 environment or even a heavy cell phone environment.

dawgit
dawgit

...At the moment. Will this lead to a way for reduction in power (transmisition power) to cover the same area? I can see were it could help to eleminate cross talk interference. Now you've made to to back to the books. I've got to do some reading. As I said, more questions. Good post BTW. I enjoy that kind of stuff. Thanks. -d (I guess that's why it's called the TechRep. :^0 )

Michael Kassner
Michael Kassner

iMAT appears to be a well-thought out technology. Do you see any flaws or situations where it would not be the best choice?

Michael Kassner
Michael Kassner

I must return the favor, as people like yourself and others on this site that are into basic science and technology are inspiring to me as well. I remember reading about evolving algorithms now and how it will totally revolutionize scientific thinking. I get very excited when I see mechanical and electrical science topics mimic biological theory. One such topic that is way out there for me is the chaos theory. I liken it to where human kind always digs a straight ditch where as how many straight ditches do you see nature dig? I also have linked a web site and application that you maybe interested in. It has this fascinating program that I have be using for years. It allows you to gain a visual understanding of wave mechanics and what happens to that wave when you introduce certain variables. It is very cool. http://www.enzim.hu/~szia/emanim/emanim.htm As for that X-band antenna, I have seen that a few times but did not pursue understanding it as much as I should have. I just thought it was RF engineers approach to negate the polarization problems encountered when you are trying to communicate with satellites. Us amateurs have that problem with our satellite repeaters. One never knows what attitude the satellite will be in so we try to devise circular polarized antennas or something similar to the evolved antenna. Again amateurs are very receptive to anything that will enhance our ability to communicate with others. I have added two more links. One is to an antenna site, that is a bit cluttered, but is a treasure trove of information. http://www.rfcafe.com/references/design_data/antenna_design_data.htm Next I thought you might enjoy this article about an insect that I find pretty amazing. http://news.nationalgeographic.com/news/2005/04/0408_050408_cockroachrobot.html

Michael Kassner
Michael Kassner

I appreciate your comments very much. I believe that is the one topic I kind of let slide. Thanks for calling me out on it. EM coupling is very dependent on the physical distance between antenna elements, and the distance needed to negate EM coupling is determined by the RF signal's frequency/wavelength. To explain, there is an inverse relationship between frequency and wavelength. The higher the frequency the smaller the wavelength. For instance a VHF frequency is 125MHz and its wavelength is approximately 2.5m. Where as a typical 2.4GHz frequency has a wavelength of approximately .13m. What this means is that a multiple antenna system for VHF would require significantly more separation to prevent EM coupling than a multiple antenna system for 802.11b/g. Now this is where it gets a bit nebulous, experts have varying opinions as to what the minimal required separation distance is in order to eliminate EM coupling. But it is usually within a wavelength, so a cell phone approximately 5 meters away from you will not be affecting your cell phone performance. Where iMAT comes into play is when you want to have multiple antenna performance on a single device and the antennas will be within that single wavelength distance. So by using iMAT technology you can get MIMO capabilities without EM coupling losses and that is indeed significant. If I have bored you to tears or have already explained what you know, I apologize. It is just a very interesting subject.

Michael Kassner
Michael Kassner

I must apologize as I am not completely up to speed as to what Linksys has developed with that antenna system. I have made a few inquiries without any results. My humble opinion is that the flat protrusion is a directional antenna, most likely a panel antenna. The two vertical antennas are most likely vertical dipoles and considered omni-directional. Whereas the flat panel antenna is much more focused, usually less that 120 degrees. This is just speculation though and I would have to take a device apart to see what the actual active antenna parts looked like. As for using this approach to deal with EM coupling that maybe. Physical separation between antennas is one method to reduce the effects of EM coupling and having a directional antenna between two omni-directional antennas may do just that.

Neon Samurai
Neon Samurai

My primary reason was to upgrade my 54GS too something that could still support nonLinksys firmware and the 350N was top of the list. I had the same question as you though; what's that big flat antenna doing there? it pivots so it's not simply a blocking shield between both stick antenna. My theory has been that the middle is more of a broadcaster or at least meant to mix a different shapped broadcast with the other two. (now if only ddwrt would release v24 or v24 rc6.3 so I have something to muck with this weekend) I think dlink's latest offering is a router with no external antenna but something like seven or fourteen of them internaly. It's currently supposed to have the best signal strength as a result. If you can confirm the correct branding, that router might be worth looking at if your not going to be using non-vendor firmware (Tomato, DD-WRT, openWRT).

Michael Kassner
Michael Kassner

I see that we are thinking along the same lines. I can not explicitly say that it will. The SkyCross people were going to monitor this article and hopefully answer questions. They certainly are more capable than I am. If you consider EM coupling exclusively then I would say yes that a single MIMO antenna would cover a specific area more efficiently than multiple antennas. But there are other factors that play into the mix. What happens if that single antenna is at what is called a null point. I talk about that in this article. http://blogs.techrepublic.com.com/wireless/?p=143 Those are factors that give multiple antennas a distinct advantage. I believe that iMAT has huge potential because it allows one to take advantage of MIMO in situations where multiple antennas are not physically possible.

compughter2000
compughter2000

MK, this is superb reading material and is indeed VERY helpful. You make all of us more knowledgable... in RF and wireless. Keep it up as you do on Wi-Fi Planet and the CWNP forum! "aka" compughter

seanferd
seanferd

Emanim is something I'll have to play around with. It may be useful for explaining EM physics to some people, and also keep my layman's knowledge refreshed. As to the rfcafe site, I don't know if I have seen that one before, but I'll add it to my list of RF and antenna oriented sites. The cockroaches: I've seen various designs in robotics based on roaches, including the feeler-antenna approach to navigation, locomotion design, and I believe even the neurology of the roach has been studied for robotics purposes. Perhaps, along with antennae, they might include "hairs" on the robots to sense air currents. Thank you for the links!

w2ktechman
w2ktechman

this is interesting. Much I already knew though. I would expect other readers to find this to different degrees as informative and interesting as well.

Neon Samurai
Neon Samurai

I'll focus a flat side towards my couch on the floor above and see if signal strength improves though it'll be hardly scientific.

Michael Kassner
Michael Kassner

I think the flat panel in the middle is a directional antenna, but as I alluded to before I would not be sure unless I could take one apart and look at the active antenna elements. One technology that you may have some interest in is that of metamaterial antennas. I think that is what you are seeing when you look at the antenna system on the Dlink device. I talked about that technology in this article. http://blogs.techrepublic.com.com/wireless/?p=189 Netgear also has a very efficient wireless device using metamaterial antennas that Tim Higgins reviewed in this link. http://www.smallnetbuilder.com/content/view/30274/100/

Michael Kassner
Michael Kassner

Thank you and you know I appreciate your informative discussions on those websites as well. RF is so cool.

seanferd
seanferd

Those sites you provided links to seem quite comprehensive. I see that quite a few "calculator" programs are offered. Enthusiasts must find them quite handy. Much of this stuff is way over my head, as I am reader, but not a radio operator. Still, I find all things radio very interesting.

Michael Kassner
Michael Kassner

The antenna link is mainly interesting to amateur radio ops, I suspect you would have a great deal of enjoyment experimenting with radios as we do. You can get as sophisticated as you want or stay as simple as you want. I linked the QEX which is the hams experimental magazine. That is on the main ham website ARRL and if you have any kind of a RF question it is a good place to start. http://www.arrl.org/qex/ One last link is to a group of very wise DIY's living in my "old home town." They have a lot of neat stuff on their website. Their work with laser data links made from hand-held pointers is especially neat IMO. They also have significant quotes on their site that I appreciate. My favorite is from one of the two most innovative companies ever, both are very near and dear to my heart, the old AT&T and Xerox Parc. This quote is from Parc: "The best way to predict the future is to invent it", I thought you would like it. http://wireless-homebrew.com/index.html I also wanted to thank you for your links, they are always very informative and helpful.

The Scummy One
The Scummy One

of keeping up with things like this that I havent heard about yet or dont have a lot of time to research. This is all good, and I like following your blogs/articles. P.S. -- your enthusiasm shows, in the way you write. It actually adds interest (for me at least).

Michael Kassner
Michael Kassner

If you have any topics or areas of interest that you would like see covered, please let me know. I am by no means an expert, just very curious and super interested in RF propagation.

seanferd
seanferd

is that many times, they have to do a lot of work to find out just how the evolved circuits work. Many times, they make no apparent sense, and are much simpler than thought possible. I still wish I could find the precise link I was looking for. It was perfect.

Neon Samurai
Neon Samurai

Ha.. good, I may still dig through my boxes of Pop Sci back issues or whichever magazine it was just to reread about the cluster. Basically, You give it two starting points and it "breeds" them together resulting in the various offspring mutations. It calculates which are the best solutions then breeds those together. This continues until the most refined soltution is found generations later. The really cool part is that the near-AI cluster can add random mutation along the way so that 1 + 1 = 3 instead of just breeding the initial two together for a 1 + 1 = 2 solution.

Michael Kassner
Michael Kassner

I would very much appreciate any information you would have on that AI antenna program and the results it created. What is funny is the actual antenna in most 802.11 equipment not much more than just wire that is held in position by the plastic. So the paper clip antenna would work as well if not better. It is the RF characteristics that count.

Neon Samurai
Neon Samurai

I remember it was not Linksys but was one of the other more recognisable brands. Your link doesn't give me a shot of the chassis though so I'm not sure... wait.. page loaded finally and sure enough, that's the little monster right there. It was the Netgear not a DLink I was thinking of. I'd give metamaterial replacement antenna a go on the Linksys if such things became available. The coolest antenna I've heard about previous to these little numbs of material is the crazy "bent paperclip" developed by a computer geniouses AI cluster for use on NASA gear. The AI cluster is a cluster of machines this guy has written his own OS for specifically for developing solutions. He tells the machine what the need is and it basically goes through generations of random mutation until it ends up with the best possible design. It's a very cool cluster (think it was a Popsci or Wired article) and the antenna it designed did literaly look like a paper clip bent into a kind of trangle slash number 4 shape. (if it's of interest, I can track it down though the article is more about the near-AI than about the antenna)