Thursday, August 4, 2011

Could DIDO Revolutionize Rural Wireless Broadband?

Earlier this week, I came across a fascinating technical white paper by Steve Perlman (President and CEO, Rearden Companies) and Antonio Forenza (Principle Scientist, Rearden Companies), and I decided it would be the perfect opportunity to finally write about a technical subject on Rural TeleCommentary. Normally I don’t get especially excited about technical papers, but this one really blew my mind. Distributed-Input-Distributed-Output(DIDO) Wireless Technology: A New Approach to Multiuser Wireless literally turned upside-down almost everything I thought I knew about wireless limitations. Now, I don’t have enough “field expertise” to make a determination about whether or not DIDO is possible, but I am still fascinated with topic and I think there will be some interesting developments in the near future in terms of testing and early deployment. If the predictions in Perlman’s paper do come to fruition, there may be a complete transformation of the wireless industry, where wireless broadband could actually become an equal competitor to FTTH. 

This is important to me because I am constantly thinking about how the broadband ecosystem is going to look in 5, 10, 15 years. When I think back 5, 10, and 15 years and contemplate how far technology has come in just half of my lifetime, it really gets me excited about the future. I have firmly argued since the day the National Broadband Plan was released that the plan’s fundamental flaw is the 4/1 Mbps broadband definition (which the Gang of 6 ILECs is now hoping to reduce to 4 Mbps/768 kbps).  I totally understand that it costs a lot of money to deploy 100 Mbps broadband to everyone, but I do not think there should be a static definition for broadband because the principle of regulatory lag dictates that the FCC will never be able to keep up with consumers’ constantly evolving expectations and definitions of broadband. I also believe that a static regulatory definition of broadband will stifle innovation. I like to envision a future in 10 years or so where nobody is restricted to the unimaginable world of broadband-enabled content and applications by capacity and speed constraints. Could DIDO help achieve this “perfect world” of super-fast, high-capacity broadband? The authors of the paper seem to think so, and they “believe that DIDO wireless will completely transform the world of communications and far more.” What I found to be particularly interesting is that DIDO apparently works at distances up to 250 miles in rural areas.

According to the paper, “Distributed-Input-Distributed-Output (DIDO) wireless technology is a breakthrough approach that allows each user to use the full data rate of shared spectrum simultaneously with all other users, by eliminating interference between users sharing the same spectrum. With conventional wireless technologies the data rate available per user drops as more users share the same spectrum to avoid interference, but with DIDO, the data rate per user remains steady at the full data rate of the spectrum as more users are added” (pg. 1).

It took me awhile to wrap my brain around this concept. It was only a year ago that I took a wireless engineering class at CU, and at that time MIMO was all the rage and there was still an assumption that we had to behave by Shannon’s laws although I remember a discussion about how we are starting to test those theoretical boundaries. I definitely do not remember my teacher ever estimating that one day there may be a wireless technology that eliminates interference between users and allows multiple users to all utilize the full data rate of the spectrum. That is basically like a wireless optical fiber for each user. As many of you know, I am a dedicated member of Team FTTH, but DIDO could actually make me consider switching my allegiance. 

Perlman provides an interesting background about the evolution of wireless and the allocation of spectrum to meet constantly increasing numbers of users and high data rate requirements, but he seems to believe that even with all the technological advances in MIMO and beamforming, wireless is still not up to par for applications like video and gaming. There are simply too many wireless users trying to do too many things with not enough spectrum. Perlman explains, “As data capacity needs continue to grow exponentially, and applications like video and videogames require high reliability, it is unclear whether there will be enough usable spectrum to meet user needs. And, once all of the usable spectrum has been allocated to consumer devices, it is effectively impossible to recover that spectrum as new technologies evolve that can use it more efficiently. We run the risk of allocating all of the usable spectrum, but still not having enough” (pg. 3). Just look at broadcasting—the fight to get some of that spectrum released has no end in sight and is years overdue. I thought the point that once the spectrum is allocated, it may be impossible to recover it if a more efficient use comes along in the future is very reflective of today’s spectrum crisis. This is one of the main reasons why I have consistently been so hesitant to get on board with wireless broadband—spectrum is a limited resource that is tightly controlled and often controlled by the absolute least efficient user, and the amount of time and effort it takes to unravel that web could deter innovation. I had a lot of hope for the White Spaces a few years ago, and I’ve all but forgotten about them now.  One thing is for sure: consumers have an insatiable appetite for wireless broadband, and there must be some combination of new technology and effective and fast-moving spectrum policy to meet the explosive demands—if not, wireless broadband will be second chair to FTTH. 

Apparently, DIDO has been tested with 10 simultaneous users in the same area, but Perlman hopes to increase that number to 100 or even 1000—therefore achieving 100X and 1000X the Shannon Limit. I found the following description of DIDO to be very illustrative:

“Distributed-Input, Distributed-Output (DIDO) wireless technology is a new approach to multiuser wireless that allows the number and density of users in the same area to be steadily increased without additional users reducing the data rate of others. In other words, the shared spectrum capacity is not subject to Shannon’s Law: as more users in a given area share the same wireless spectrum, the data rate per user does not decline. As a result, regardless of how many users are in a given area, each user is able to use the entire Shannon Limit of the channel, despite sharing the same spectrum” (pg. 4). 

I recommend that you read the paper yourself to learn more about how DIDO actually works, as there are some interesting illustrations that accompany the description. Perlman compares DIDO to a simple Wi-Fi configuration, noting that in urban Wi-Fi networks it is not unusual for users to only get about 4% of the available data rate. Well, DIDO is here to save the day, if the Wi-Fi access points are replaced with DIDO access points in this hypothetical scenario. Unlike Wi-Fi, data is processed through a “DIDO Date Center,” which “processes the data, modulates it into a radio signal waveform and sends the waveform [to the access point], which simply sends the waveform to its antenna and transmits it as a radio signal” (pg. 9). From what I gather, the DIDO Data Center does all the heavy lifting as more users are added and there is no risk of a stronger signal overpowering a weaker signal when signals are transmitted at the same time. Perlman explains, “something rather remarkable happens: the sum of the radio signals at each computer’s location results in a clean modulated waveform carrying the data intended for that particular computer…And here’s the really amazing part: what each user receives is what they would have received if they had the channel to themselves, without another user sharing the same spectrum. There is no interference from the other user. Each user is able to utilize the full Shannon Limit of the channel” (pg. 10). Wow… Is he serious?! DIDO just sounds awesome. The authors call DIDO a “cloud wireless system,” which seems appropriate, as “all of the intelligence of the DIDO system is in a DIDO Data Center, which then communicates to all of the users at once through all of the APs at once” (pg. 12). 

The paper explains that DIDO is able to crush the limitations of Shannon’s Law because, “DIDO is a general solution that creates an independent channel for each user, even in densely-populated areas. Since each user has an independent channel, Shannon’s Law does not apply, despite the fact that all users are sharing the same spectrum” (pg. 13). The authors also provide the following effective visualization for DIDO: “A DIDO channel can be roughly visualized as a 3D sphere that surrounds the antenna of a user device. We call it ‘area of coherence.’ Within that sphere, the channel for that particular user exists. Outside of that sphere, it does not. So, if there are 10 users in a DIDO network, there are 10 spheres, one around each user device. As a user device moves, the sphere moves with it” (pg. 14).

So, what is the relationship between DIDO and rural wireless broadband? More importantly, are there any business opportunities for RLECs in DIDO? I think that the answer to both of these questions remains to be seen, depending on whether or not this product becomes available to the mass market (the authors explain that it was designed for the mass market, and DIDO was tested at frequencies from 1 MHz to 1GHz).  The authors also argue that DIDO systems are highly reliable and do not have the dreaded dead zones—something that is especially prolific in rural and rugged areas. I think the most important aspect of DIDO for rural broadband providers to consider is the potential range of 250 miles (“DIDO Rural,” pg. 15) if HF frequencies in 3-7 MHz are utilized. The authors note that the HF frequencies have the limitation of being very narrow, but “DIDO would be an ideal technology to overcome this limitation” (pg. 15).  The key point in the paper for rural broadband providers to pay attention to is: “DIDO-NVIS provides a very inexpensive and efficient way to deliver broadband to rural areas, or remote areas in the wilderness” (pg. 17).

I do not recommend that RLECs abandon their FTTH projects and put all their bets on DIDO, but I do think this technology is worth keeping a close eye on while it continues to develop. I can see it being very useful in extremely rural areas, such as the areas slated for receiving satellite broadband within the new USF framework. Could DIDO eliminate the need for satellite broadband altogether? Are there any business opportunities for the RLECs who are lucky enough to hold spectrum assets to manage private DIDO networks for businesses, farmers and industries? With RUS’s Jonathan Adelstein’s visit to a rural Iowa farm last week, I started thinking about the agriculture possibilities enabled by high-speed broadband. I definitely see DIDO being really useful for “precision farming” and other high data-rate needs on the farm. It could enable innovations like remote crop video monitoring, so farmers could literally watch their corn grow from a beach in the Caribbean. 

I will definitely be keeping an eye on new developments in DIDO, and if anyone has any good resources about it, or any studies that refute these claims (I like to hear the other side of arguments too), please send them my way. This article by Dean Takahashi was also very interesting. 

I will be on vacation next week until the 16th, so I hope everyone keeps plugging away on USF Reform. The FCC released its Public Notice seeking comment on the ABC Plan, and comments are due August 24 with reply comments on August 31. This is the FCC being extremely aggressive about the expected October deadline, as expected. Good luck with your comments; I look forward to reading them!

Cassandra Heyne

1 comment:

  1. DIDO live demo shown last week... http://www.youtube.com/watch?v=5bO0tjAdOIw

    ReplyDelete