Could this be the partner Ed was referring to?
By: Steven J Crowley 1/23/2013
Google’s “confidential” test might be a super-dense LTE network using Clearwire’s spectrum
Google filed an application at the FCC last week seeking permission to conduct testing of an experimental radio system. Portions of the application and accompanying exhibits have been designated confidential and are thus not available to the public. Even the request for confidential treatment has been redacted. Let’s try to infer what’s happening from the information available.
An exhibit says that Google intends to construct a radio network in the vicinity of the Google corporate campus in Mountain View, California. Google plans to test up to 50 base stations and 200 user devices. Base stations will be indoors and outdoors, with the range of each 100-200 meters, and 500-1000 meters, respectively. Both directional and non-directional antennas will be used. The experiment is to take place within a two-mile radius, so this is a quite dense network, which could have very high capacity for carrying data.
The frequencies requested are 2524-2546 and 2567-2625 MHz. These are bands allocated to the Educational Broadband Service (EBS) and the Broadband Radio Service (BRS), which are used by Clearwire for its mobile broadband service. (Google owned a stake in Clearwire, which it sold last year.) A cursory check of the FCC’s database (the accuracy of which varies) indicates that Clearwire, in the Mountain View area, might be leasing at least some of this spectrum from Stanford University.
The output power of the devices is designated on the application form as “not applicable,” which doesn’t make sense. The power is a fundamental quantity that should be disclosed so others may independently #$%$ the potential for interference from the experiment to their services. FCC staff should ask Google to supply this information. Because the area covered by the experiment is small, and there are many base stations, we can guess that the power of each is relatively low, on the order of a few watts or less (with device power even lower).
As with the missing power information, the information supplied for the emission designator is incomplete on the form (the “necessary bandwidth” is missing). The code supplied, “F9W,” is sometimes used for LTE, and it would be reasonable to expect LTE is being used in this test.
We don’t know yet exactly what Google is testing here. It might be devices it created. I suspect, though, that this is a test of a network architecture or service, using existing equipment. Google has been lobbying the FCC to approve the agency’s proposed shared-spectrum small-cell service in the 3550-3650 MHz band, and these test results might be relevant there.
This appears to be Google’s first experimental radio application using mobile broadband bands. Prior to this, Google’s requests for experimental radio authority have generally been confined to frequencies used for unlicensed devices (e.g., the 2.4 GHz band, 5 GHz band, and the 76-77 GHz band). There has also been experimentation on UHF TV frequencies for white-space device testing.
Spectrum bands are specified in several ways as to what technology can be used. That 'specification' tends not to be so specific that it requires one type of the technology standard, ie. using LTE vs. WiMAX or different types of 3G. That tends to be the operator's joint decisions... it generally has to be common for operators in the particular segment of the band so that interference is not created and the spectrum gets used more efficiently. So, it looks like Google will use LTE technology in 2.6GHz as that is the most widely accepted and most advanced standard. However, there remain segments of spectrum where HSPA continues to be deployed by T-Mobile, for example, because it is extension of current use not ready for conversion and HSPA/HSPA+ is still a new, although pre-True 4G standard that delivers improvement over prior 2G-3G.
Besides the coordination of how standards get adopted to make sure operators will tend to work well together in the same or nearby bands, the FCC regulates power levels, guard bands that separate adjacent bands, and many cases of more finite requirements, such as areas where use must be limited. They regulate things like tower height and co-use situations that can vary depending on where and what band is deployed.
An advantage of the higher frequency bands is that signals do not travel as far and through things as well. That sounds like a mistake.. 'Advantage?', yes because when trying to pack small cells closer together the problem or signals not traveling as well is turned more towards being a benefit as higher order MIMO-AAS and future developments in MU-MIMO/Co-MIMO and multiple tier and beam forming methods are used. This is where Google can be of significant service: in showing off some of the advanced network device topologies, interference mitigation, and subscriber devices that work as part of the network rather than just consumers. The advanced network technologies that have gone into LTE-Advanced will be a real great play field for 'innovations in the network' and the end-to-end delivery that takes in we little idiot users. Google can play this role because they can do a lot of cool and showy stuff without spending all that much. And they are not an equipment supplier, (they do sell Google/MOTO subscriber devices), so they appear more agnostic to the industry. And they tend to be looked upon to set the pace in a way many local-regional-national government groups can buy into.. including the regulator, FCC.
The FCC sometimes provides experimental licenses that allow more leeway in what can be tried on a basis of oversight. In this case, Google will have to live within the norms for the spectrum... but I wouldn't be surprised to see some things like approaches to multiple-band use or use of the higher frequency bands also be tried that are outside of the norms.. are experimental in regulatory treatment. For the most part, Google will likely not be trying to press the envelop similar to the White Spaces effort because the 2.6GHz spectrum is so large an expanse of possibilities in itself and when combined with other bands and modes using the current regulatory framework.
Before buying into the "Wowsie wow, wow, wow" of tech announcements, investors need to consider the what's and wherefore's of commercial market applications. Google is pursuing leading edge developments, however, when, and on what scale and whom/what companies will benefit has to be pinned down to more than the woosie-woo statements. What companies, for example, have benefitted from Google's much talked about fiber optic pilot projects? By how much has that cuased the stocks of whatever companies to rise? Its very easy for a fellow idiot to post "Woosie, wooo, woopie! Google, the big kahoona is deploying pilot projects!, the agreed to be acquired Clearwire (CLWR) will be $30 and my Grandma will become president!"
The Google projects represent commercial stage technology and market rediness which is a very important thing to note. However, the how, how much, and who is going to benefit questions are not trivial. Just ask all the investors in CLWR over the past four years as the stock went from ~20 to where it is now... Woosie, woo, whompass
hello, team rep
When CLWR, down in the 1.20's you might remember what you were saying --perhaps you could have given better advice then...
Politely said your logic/credibility/insights were nothing more then noise.
Additional reading indicates they are looking for same Station Location approval(s) for:
Mountain View California
Los Angeles California
New York New York
Interesting they are for such a wide geographical area and on same frequencies.
Found this on FCC document;
Please explain in the area below why an STA is necessary:
On January 26, 2012, Google Inc. (?Google?) was granted special temporary authority (STA) to test a next-generation personal communications device (Call Sign WF9XKU). The STA was extended to December 15, 2012, see File No. 0448-EX-ST-2012, in order to ensure that Google could continue testing until receipt of equipment authorization, which Google anticipated receiving in the fourth quarter of this year. Google now expects to receive equipment authorization early in the first quarter of 2013, having made minor changes to the device. Consequently, Google requests a brief extension of the STA, until January 31, 2013, so that it may continue operations pending receipt of authorization. Google also requests approval to make minor changes to the number of devices currently approved at two locations, without changing the total number of devices.
Will the antenna extend more than 6 meters above the ground, or if mounted on an existing building, will it extend more than 6 meters above the building, or will the proposed antenna be mounted on an existing structure other than a building? No
(a) Overall height above ground to tip of antenna in meters:
(b) Elevation of ground at antenna site above mean sea level in meters:
(c) Distance to nearest aircraft landing area in kilometers:
(d) List any natural formations of existing man-made structures (hills, trees, water tanks, towers, etc.) which, in the opinion of the applicant, would tend to shield the antenna from aircraft:
Action Frequency Station Class Output Power/ERP Mean Peak Frequency Tolerance (+/-) Emission Designator Modulating Signal
New 2400.00000000-2483.00000000 MHz MO N/A 1.000000 W P F1D
Action Frequency Station Class Output Power/ERP Mean Peak Frequency Tolerance (+/-) Emission Designator Modulating Signal
New 2400.00000000-2483.00000000 MHz MO N/A 1.000000 W P W7D
A system devised using the current state of the art technology such as IMT-Advanced/LTE-Advanced, while having to comply with power limits would not necessarily have power ratings strictly prescribed for network and subscriber devices. These can make use of adaptive power control that is modulated based on how they are used and what other devices are in use at the current time in the surrounding environment.
Several of the schemes for highly dense networks include enhanced self-configuration and self-healing methods. For instance, a network might be deployed that includes a number of types of units designed for different physical locations and purposes that vary in the range, bandwidths of spectrum they use, the overall throughput, numbers of antennas/arrays and level of MIMO, and numbers of users they can support. Although 'self-configuring', these would be deployed based on how they can be deployed and the type of usage expected. The concept is pretty simple, deploy outdoor units designed for pole, wall, embedded, roof-top, etc. locations. The practice still requires the planning, purchasing, and skilled labor deployment in many cases. its possible for indoor units to be similar to Wi-Fi devices or ceiling and wall mounted units or for units to come built into light fixtures, etc. Or that integrated type deployment can be put in the appendix of the reports as being among future goals for the network approach.
What Google might be pursuing is along the lines I've thought Clearwire should have attempted: get users, particularly businesses, educational institution campuses, office parks, government buildings, etc. to deploy the dense locations that fit into the wider area coverage grid. Efforts in fiber optic and microwave back-haul fit into the overall scheme as well - among the new back-haul units for 24, 40, 60GHz and other spectrum are those that make use of similar MIMO-AAS as has been developed for LTE... in fact often using the same core technologies that are modified for that use.
A common question of this type of experimental/trial network is 'how is it going to be commercially scaled?' How the heck do you get it to be widely adopted? ... Huge upfront investment for managed deployments, or do you 'design' the market solution for the near-building layers of the network to be partly to mainly user deployed, thus reducing cost and lacing a lot of it in the hands of those that will use it... similar to the case for WiFi deployments.
EBRS spectrum was originally set aside for use for video and BB in and immediately around educational and religious institutions campus type settings. That is still how the spectrum is licensed/sub-licensed. The technology has evolved to now make it feasible to use it by piggy-backing on the mainstream wireless technology, chip, device, and software environment.
Google has gotten involved in several demonstration efforts that have value. However, looking around today most people will find little evidence of 'Google everywhere' in converting these demonstrations into widespread use or in changing the overall direction of the industry. Google is a large, powerful and inventive organization, but it is no exception to the challenges of taking technology and taking it to become widely adopted. "End to end delivery system" is needed, not just the chunked down tech demonstrations.
The business environment is changing and Clearwire's management has been aware and trying to do something with it: Earthlink has announced several agreements in recent months, including one announced this morning, in which they deploy and manage campus-wide fiber optic and wireless BB networks as part of a 'cloud hosting' package. Much of this contract concentrates on cloud servers.
So, you can see that the 'industry' is now at a stage where the various pieces to build end-to-end solutions that take advantage of 2.6GHz have come into place. Clearwire has a partnership with Earthlink and other companies that can now use the spectrum as part of the package. Google is engaged in demonstrations of what might be described as the 'massively microcell' approach to building networks. All this points to the technology, supply ecosystem/cost factors, being in place to deliver to viable markets. However, the challenge remains how to scale that to the level of revenue Clearwire would have needed to remain whole. It is too late unless a miracle were now to come into play... as much as investors might hope, DISH has not dished out enough of a deal to overcome Sprint-SB obstacles or to make the resulting Clearwire attractive (enough).
When you get down to it, Google, Earthlink... have never pioneered a new field of communications. They have prospered by riding the waves of industry development that are otherwise underway. Google did not create the Internet, browsing, search, email, OS/Linux, or current mobile environment paradigm.. they mostly copied and innovated in part. A large part of their success is due to putting together a better, more complete, more competitive/lower cost approach. Companies like Earthlink capitalized on dial-up and then into broadband. So, the question is how can they now capitalize on wireless broadband? As a deployer of networks, a role they have never assumed except as following rather than leading development? Or can Google be a 'catalyst' in such a way that their example causes masses of others to follow? The answer is that to deliver a game changing solution you must change the end-to-end solution and not just a small part of it... or be enough of a catalyst that the pieces 'self assemble' in rapid order. That type of thing has happened.. it is how Google grew: the Internet was available on which they could introduce a better, more comprehensive search capability... shazoom! How do you do that for a situation that takes real hardware to be deployed in very diverse environments? The user has to play a role imo. The end user deployment model or, in the meantime at least, the 'network tier' level has to be enlisted. That sets the possibility for Google, Earthlink to work as part of a tiered HetNet approach at the campus/business park level That is because 'that is where the need/money is' and where the deployments can be engaged at the local organization and user level... its chunked down enough to be viable.
I tried posting the source link, but it was blocked by Yahoo. Just search for the author's blog. Isn't the BRS spectrum listed in the application similar to what Dish is rumored to be trying to purchase?