Jump to content


Popular Content

Showing content with the highest reputation since 01/26/2012 in Blog Entries

  1. 90 points
    by Christina Herron Sprint 4G Rollout Updates Thursday, February 14, 2013 - 3:10 AM MST Travel back with me a few weeks ago. It’s a cold Saturday afternoon, and I am standing and staring into a gated area near a cell phone tower. My date for this outing is frantically taking pictures and discussing the exciting upgrades to the equipment which have occurred since our last visit. I have no idea what he is talking about, all the mechanical stuff looks exactly like it did a few weeks ago. To me, it happens to look exactly like the 3 other sites we visited today. Not wanting to look totally stupid, I just smile and nod. Yep, this is what I do for fun for my 17th wedding anniversary. I am officially married to a cell phone dork. Since May 2011, I have had many a romantic dinner which usually includes at least a drive-by various cell towers. I have smiled and nodded through countless conversations which are dominated by the words, “LTE”, “Wi-Max” and “backhaul”. I still have no idea what any of this means. I have taken vacations based on the availability of 4G coverage. This past summer our family ended up in Waco, Texas. Not exactly the place one would call a cosmopolitan vacation destination. At least we got to go to the Baylor football season opener. We ran tests in between plays, and my husband would constantly grab my phone to look at an Engineering screen. I have changed hotel and restaurant reservations based on cell tower locations. My kids can spot a cell tower from 5 miles away. Family outings generally include a visit to one. My oldest son understands which panel is which on the cell tower. He also understands the secret language of cell phone technology which my husband speaks. I just continue to smile and nod. Many friends ask me in wonder why I allow my date nights to be interrupted by visits to cell phone towers and discussions about things I don’t understand. The answer is simple; I have the best husband in the world. S4GRU is my husband’s passion. He loves all things Sprint, cell phone, and the technology that goes with it. He is passionate about the site and getting the information to its members. He gives unselfishly to it while still honoring all his work, home and community obligations. When he isn’t working on this site; he is being an awesome dad, devoted husband, faithful employee and a good citizen. The only sacrifice I make is that I occasionally have to smile and nod, so I don’t look like I have no idea what is going on. He sacrifices his sleep, his free time and sometimes his sanity to make sure we are all happy. I couldn’t ask for more than that. Today is Valentine’s Day, the day of romance. I am looking forward to a romantic dinner at my favorite restaurant and maybe a box of Godiva chocolate. There are two things I am pretty sure of. There will be at least one stop at a cell tower during the night, and I will probably be doing a lot of smiling and nodding.
  2. 70 points
    by Robert Herron Sprint 4G Rollout Updates Monday, November 11, 2013 - 11:55 PM MST S4GRU Staff and Members have been anxiously waiting for over a year for smartphones that would support all three of Sprint’s LTE bands. Since April 2012, Sprint LTE devices have been limited to only one band. Band 25 (1900MHz in the PCS Band). Subsequently, Sprint closed down the Nextel network and picked up another LTE band (Band 26). Also, Sprint purchased Clearwire and picked up another LTE band that they had started to use (Band 41). So Sprint now has use of three LTE bands which will allow it to provide more capacity, better maximum LTE speeds and coverage. With now three LTE bands, Sprint needs Triband LTE phones. S4GRU and our members really became excited in Mid 2013 when we learned through sources that the first Triband LTE smartphones would be out in late September. Shortly after that, we learned that the upcoming LG G2 would be able to support all three Sprint LTE bands via FCC reports. Rejoicing and happy tears filled all of us wireless nerds in anticipation of the first Sprint Triband LTE devices. Then we learned through a source that Sprint Triband LTE devices would not support SVLTE (Simultaneous Voice and LTE). To some this was a setback, but the hard core wireless enthusiast was not distracted. We all wanted Triband LTE at any cost. Besides, nerds don’t spend much time on the phone talking with people. We waited and waited, but nothing released. A few more Triband LTE devices came through the FCC, including the Nexus 5, Samsung Mega 6.3 and the Samsung Galaxy S4 Mini. Then we found out that the Sprint versions of these device would not launch at the same time as their competitor counterparts. Many of our members were screaming about the delays. And we knew there had to be good reason. We just didn’t know what. The dam finally broke with Google’s release of the Nexus 5 on Halloween. And Sprint finally broke down and released the LG G2 about a week later. S4GRU and dozens of our core members quickly got their hands on their new Triband Nexus 5’s and G2’s and all was happy. For a few minutes. Until they tried to use Sprint’s LTE network where they used to on previous LTE devices. Some of our members reported that both the G2 and N5 had extremely strong LTE signals in Band 25. The best they have ever encountered. However, there was a very vocal group who were reporting that they could not stay connected to LTE for more than a few seconds. Something was very wrong. We tried to troubleshoot and figure out the problem with our members. But there were no clear common denominators among the problems that we could ascertain. We could not figure it out. And then we received heads up from internal memos within Sprint as to the problem. Sprint Triband LTE devices use Circuit Switched Fallback (CSFB) on the network. Sprint Triband LTE phones dropped SVLTE for eCSFB/CSFB Up until these new Triband devices, previous Sprint LTE devices supported simultaneous voice and LTE (SVLTE). It could do so with two separate transmission paths from the antennas to the chipset. Voice/texting could run via 1xRTT on one transmission path. LTE could run a separate path, allowing data and voice to be used simultaneously. In contrast, Sprint Triband LTE devices do not support two separate transmission paths. They have one path, shared by voice/SMS and data. We were alerted to this months in advance. However, we did not realize that the network would have to run on Circuit Switched Fallback in order for this to work and what the ramifications of this would be. S4GRU was told by a source this past summer that Sprint and the OEM’s came to the conclusion that these new Triband LTE devices could not use SVLTE in the conventional way they used to, and it would require a lot of engineering, testing and cost to even attempt such a design change. It was decided to release Triband LTE devices without SVLTE. It may seem that the only drawback for doing that is Sprint Triband LTE devices would not be able to run simultaneous LTE data while on a phone call or when actively transmitting a text. But there is another. And it’s why many early adopters of these new Triband LTE smartphones no longer are being able to connect to Sprint LTE in many places that they used to. How it works In previous Sprint LTE phones, when a device was in Sprint LTE coverage it would park in both the LTE and CDMA Sprint networks at the same time. When a voice call came in, it would just go straight through to the device. And signal to the LTE network would be maintained the whole time while the call was active. In contrast, a Sprint Triband LTE device can only stay on one technology at a time. CDMA or LTE, not both. So when a Sprint LTE Triband device is in Sprint LTE coverage it parks only in LTE. And doing so means it cannot transmit calls without Circuit Switched Fallback (CSFB) on the network side. CSFB and eCSFB (Enhanced Circuit Switched Fallback) are network controls that will allow a single mode/single path network to operate in two modes, both CDMA and LTE. Here is how it works in the simplest way I can describe. When your Triband LTE device has an LTE signal, it cannot receive or make calls on its own. It is just using LTE data happily. However, what if someone calls you? How does it get through the CDMA network to your device? Via CSFB. When the Sprint network tries to forward a call to your device but cannot see it via CDMA, it then checks for an LTE connection to your device. If it sees one, it tells your device to disconnect from LTE for a moment and reconnect to CDMA. Your device then jumps over to take the call on Sprint CDMA and the LTE session is interrupted. This happens very fast and seamlessly. Except for the loss of data availability. If you receive a text, the Sprint network is able to route it to your device via LTE. Circuit Switched Fallback is a great solution to the issue of Sprint Triband LTE smartphones. But the problem here is that the Sprint network is being upgraded in Network Vision, and not all Sprint parts of the Sprint network can currently support CSFB. And it affects all Sprint Triband LTE phones, not just the Nexus 5 and LG G2. Why it’s not working and impacting LTE in some places As everyone reading this article probably knows, Sprint is in the middle of a network modernization program nationwide called Network Vision. It upgrades every piece of network hardware, site equipment, radios, software and network backhaul to every one of Sprint’s nearly 40,000 CDMA sites. And much of Sprint’s legacy network either doesn’t support Circuit Switched Fallback or doesn’t support it in cases where the legacy network equipment is by a different manufacturer than the new Network Vision equipment. The problem that these early adopters of Sprint Triband LTE devices are encountering is that when their phones connect to the Sprint network they try to connect to LTE. And when it cannot see the CDMA network through CSFB, it then reverts back to Sprint CDMA and stays there. It does this in order to preserve device connectivity for the user to Sprint voice capability. When forcing these devices into LTE Only mode, the LTE works very well ruling out a device problem. They just are unable to use LTE in default mode without being able to have access to CSFB on the Sprint network. How and when is this problem going away? The good news is that most of the Sprint network is capable of supporting CSFB in some form or another now. Some markets are not having any problems at all, like Ft. Wayne/South Bend, Puerto Rico/Virgin Islands, most of Chicago and Indianapolis. eCSFB is complete or very close to complete in these markets. Upgrades to the Sprint network are being handled nationwide by three different OEM’s. Samsung, Alcatel/Lucent and Ericsson. They are in various stages of deployment and are currently impacted differently by region. In places where CSFB is in place and operational, there are no problems with using LTE on a Sprint Triband device. And Sprint and their OEM’s are scrambling to get CSFB operational in all the other places. Some of the existing networks are capable of supporting CSFB and Sprint is working to get software upgrades in place for these networks to get it operational on them. However, some of the Sprint network has unsupported equipment from Motorola and these cannot be upgraded and will need to be replaced with their new Network Vision equipment to allow LTE and voice to work together via CSFB. Currently, just over 60% of Sprint sites have their sites upgraded to new Network Vision 3G standards which allow Circuit Switched Fallback capability. However, not all 60% of these sites are currently allowing LTE to work on a Triband device. These all should be capable of using LTE on a Triband device now, or in the next few weeks. Many of these markets will need to have their MSC Switch Center’s software upgraded too for CSFB to work. Beyond this, Sprint also has another 10% of their sites that have LTE operational but not the 3G upgrades that support CSFB. These 3,000 sites currently have Sprint LTE live, but it cannot be used by Triband devices without CSFB active. But there is hope for these locations. These sites do already have all the hardware needed to install upgraded 3G that will work with CSFB on the network. Sprint is scrambling with their OEM’s to get 3G up and running on these sites as soon as possible. Many have been upgraded recently and they will continue to be upgraded over the next weeks and months. I was told by an unnamed Sprint source that half of these will be CSFB capable in a month and the other half will be between 2-3 months additional beyond that. Sprint should be in a position that in the next 3 months that their entire LTE network will be CSFB capable and this will go away. As each site gets CSFB capable, Sprint LTE Triband device owners will be able to connect to LTE. And some S4GRU Members have already experienced this and are now reporting some sites reappearing to be used by their Triband LTE devices. This is likely do to a recent enabling of CSFB at the connected site. What about the last 30%? The last 30% of the Sprint network is not currently affected by this problem because they have yet to be upgraded with Network Vision or LTE. These sites are in various stages of being upgraded. In internal correspondence, Sprint says they will now take into account CSFB availability before launching new markets. Network Vision deployment will continue as normal, but OEM’s will now try to launch LTE and CDMA upgrades together at each site whenever possible and install CSFB capability at the network level for all the remaining sites. In cases where they cannot happen together, Sprint will continue to allow the LTE site to go live. But the site will only be discoverable initially to Sprint SVLTE devices. But by the time Sprint is ready to launch the whole market, CSFB will need to be operating before they issue the Press Release so customer expectations are met for all LTE device holders. Conclusion The bottom line here is that there are thousands of Sprint Triband LTE early adopters that are currently not able to connect to LTE sites that do not have a CDMA network connection that support Circuit Switched Fallback. But the problem is temporary, and improvements will go live every day around the nation reducing the number of affected sites. It will get better and better every day. However, we do not know how different markets will fare and when. It will be highly variable. There are many advantages of being an early adopter. However in this instance, for those who are very dependent on their new found Sprint LTE service, this may be too big of a burden to bear. These folks will need to use a Sprint single LTE band device until CSFB is working in their area or, as some have threatened, use another wireless carrier. At S4GRU, we believe that knowledge is power. This is the explanation of what’s going on, and what is being done about it. Now use the info to determine what’s best for you. Most of our members will likely just endure it and then reap the rewards once CSFB can be brought online in their area. A parting point in all this is Sprint is promising some advantages to a single transmission path with Circuit Switched Fallback. Sprint says in their memo that Sprint Triband LTE devices with CSFB will have improved battery life and better edge of cell radio performance. We’ll be glad to enjoy those benefits when they are fully realized. EDIT: Since the initial publishing of this article, it was discovered that Triband LTE devices were capable of sending/receiving texts via LTE. It is only voice calls that require Triband LTE devices to shunt back to the CDMA network via CSFB. The article has been edited to make this clarification. Initial LTE devices were data only (like USB dongles and MiFis), then LTE devices with voice/text services use either SVLTE or CSFB. Finally, Voice over LTE (VoLTE) will be enabled in the coming years that will allow simultaneous voice and data without need of falling back to 3G/CDMA networks. But VoLTE is still at least 18 months or more from being instituted on a large scale. Sprint Internal Memo regarding Circuit Switched Fallback issues:
  3. 66 points
    by Robert Herron Sprint 4G Rollout Updates Tuesday, July 17, 2012 - 5:01 PM MDT I am taking a moment and breaking from our normal Network Vision news and educational pieces to write an editorial. I try not to bloviate, but I feel like I am at a breaking point here. This weekend was a moment of great joy for many of us Sprint Network Enthusiasts as 4G LTE started going live at hundreds of sites across the country in a few select markets. However, our geek party was frequently interrupted by incessant whining. One of the chief frequent whines I heard around S4GRU was related to LTE coverage. And it still is populating our forum posts, my e-mail box and our social media sites. You folks need to wake up and get a grip. The world is not ending. S4GRU has been out there building expectations among our members and readers from the beginning. We have written hundreds of articles on Network Vision/LTE deployment. Anyone who actually reads our content knows that Sprint is targeting 40% market completion at market launch. 40%. That means 60% of the sites within a market do not have LTE at the time a market is planned to launch. Sprint did not quite even get to 40% with these July 15th markets, but proceeded any way based on demand from customers wanting access. You hypocrites! You know, I find it very interesting that so many people were pushing Sprint to stop blocking LTE connections. There was a large battle cry from most Sprint LTE device holders in active deployment areas for Sprint to stop blocking completed LTE sites. "Let me use my LTE, darn it!" This was heard over and over again. We even were championing for Sprint to open up their LTE network at completed sites for customers to use. Finally, Sprint does exactly that. Instead of rejoicing, there was whining en masse. "I live in the San Antonio market. And the block where I live behind the Piggly Wiggly doesn't have LTE right this very second. Sprint sucks. I'm leaving!" Really? Are people that messed up??? Many of you should be ashamed of your self-centered ridiculous tantrums that you posted, publicly embarrassing yourself. You act as if Sprint actually went through maps and hand picked who would win and who would lose in early deployments. This is far from reality. Early access or comprehensive coverage? Pick one, you can't have both The question I have for you folks is this... Should Sprint have waited until these markets were 100% complete later this year to allow the completed LTE sites to be used, or should they open up the markets now where at least the completed sites can be used? This is a no-brainer! Open them up now and every additional site that goes live every week, as they are complete! These markets that have launched are not done. They are still active deployment zones and additional sites will come live every week until completed. And we will update the progress here at S4GRU. This ridiculous moaning and complaining will just make it more likely that Sprint will not allow other markets to go live early. If all they hear from their customers are the whiny bunch, then they will think their customers don't want LTE until it is completely ready, with no bugs and completely 100% deployed. You may not like it, but the complainers are speaking for all of us. Time to stand up and go on the offensive If you want to continue to have access to the LTE network early, then you need to stand up and start posting out there the counter story. The tech sites, blogs and forums are being inundated with these people speaking on your behalf. Complaining about all the problems of an early launch and early access to LTE. You may even have to go to the Sprint Community Forums and help defend the intelligent decision to open up LTE early. The counter point needs to get out there. It's time for the Wireless Nerds to take our rightful place. We want access to the network early. We would rather live with a few bugs and limited coverage than to not have access to Sprint's LTE network at all. Sprint needs to continue opening up their LTE network even in more places where they can. And they will be afraid to do that in the next markets if we don't take a stand. I know this editorial may be a little over the top for some of you. But I am mad as hell, and I'm not going to let the whiners speak over our voices any more. We are Sprint, not them! Signed, Robert Leader of the Nerds EDIT: Changed the two references to bitching. I violated my own rules.
  4. 51 points
    by Travis Griggs Sprint 4G Rollout Updates Wednesday, January 30, 2013 - 10:00 AM MST A PRL file is a Preferred Roaming List. In simple terms, it tells the device how to scan for various wireless cell systems, which ones are native, and which priority to use them in. If there isn't a native Sprint signal available, the PRL defines which roaming partners to scan for, which ones should be used, and in what order of preference to scan for them in. Contrary to belief and what some Sprint reps may tell say, a PRL is not a list of cell sites. You do not need a new PRL update to receive service from a new cell site. Nor will a PRL update result in faster Sprint EVDO (3G) speeds either. Of course there are a few exceptions to these rules with roaming agreements and/or Network Vision in the picture, but we will explain that later. PRL updates have nothing do with 4G WiMax coverage either. On some 4G LTE chipsets such as Qualcomm, the PRL determines if LTE is enabled for the geographic region you are in. So how does a PRL really work? Before I can start to explain the inner workings of a PRL, there are few terms for reference: A PRL is broken down into a three tier system: GEO - Geographic areas (regions), they are commonly referred to as a GEO. SID - System IDs assigned to the various carriers. NID - Network IDs are assigned by carriers to break a SID up. Common wireless bands found in US CDMA PRLs: PCS Band - 1900mhz PCS band in the US (A block, B block, etc) - Band Class 1 or 25 Cellular band - 850mhz cellular band in the US (A and B side) - Band Class 0 SMR band - 800mhz band used previously by Nextel. CDMA 1xA is in active deployment - Band Class 10 Other terms: Channel – assigned frequency within a band (200, 476, 350, etc) Negative (Neg) Network – SID/NID is prohibited (only 911 calls allowed) Preferred (Pref) Network – SID/NID is allowed for acquisition and usage Preferred Only PRL - only the SIDs specified in the PRL are allowed for acquisition When a device is powered up for the very first time, the phone will start at the top of the PRL and start searching through the list of SIDs for a native Sprint signal. This usually happens very quickly. Once your phone acquires a SID in your GEO, the devices will stay within the GEO for any additional searching for SIDs before it goes out looking in other GEOs again. This gives your phone a quicker response time of finding another SID when it needs to. If you have ever noticed it takes a little longer to find a signal when the flight attendant states you may now use your wireless devices, this is your phone searching the last known GEO, the devices then gives up and starts searching the other GEOs until it finds one to acquire. The SID/NID records within the GEO have their various priorities and channel/band scans assigned to them. A SID is the regional number assigned to wireless system. A NID is used by a cellular carrier to break up a large SID into smaller pieces for further localizing scans/rules. For instance a SID that has two large metro areas could have a NID of 51 for one area and 52 for the other area. The record would be listed as 4159/51 and 4159/52. If Sprint needs to apply different rules and/or acquisition channels to either NID it will put a record for each one. If no local rules are needed, the NID is listed as 65535 to encompass all NIDs within the one SID. In the PRL analysis reports, any NID of 65535 is suppressed as it is not needed. It may sound confusing at times but it is a simple three tiered system; GEO area, SID, then NID. In the PRL example above there are 5 SIDs assigned to Geo #4. The first two have a roaming indicator of 0, meaning a native Sprint signal. 22411 and 4159 have a priority of 1. These two SIDs do not necessarily have a preference in which either is used since they are the same priority but the device will scan for 22411 first. If 4159 is acquired, the device will not actively seek another network to use. During various sleep periods and/or timers the device could scan/acquire 22411 though. Once the device finds itself without a usable signal from 4159 or 22411, the scan will proceed into the next priority group. The next priority group of 2 has SID 4279 and a roaming indicator presented to the user. The device will acquire 4279 and notify the network carrier of its presence. The device will actively and aggressively continue to search for a non-roaming signal. Due to this continued scanning this may cause the radio chipset to not enter into the power saving sleep modes causing increased battery usage. As long as SID 4279 is available, the device will not search for SID 4160 with the priority of 3. 85 is a NEG network meaning your phone is not allowed to use this network for any reason other than 911 calls. What happens when Sprint installs a new cell site? I will say it again and again. You do not need a PRL update to use a new cell site, you do not need a PRL update to use a new cell site. Many Sprint reps will swear up and down that a PRL update is required to use new cell sites. This is incorrect! Many Airaves are activated and deactivated everyday but yet we don't see new PRL updates for these everyday. Using the example above, the phone is attached to Sprint 4159/51 using the same cell sites that were active on the previous day. Today the Sprint crews activated a new cell site to extend coverage a few more miles down the highway. Sprint will configure this cell site with the same licensed channels for the area and also configure it as a 4159/51 site. The devices in this area will use this new site without ever needing any type of PRL update. I've only scratched the surface of the various inner workings of the PRL file. Stay tuned for part 2 of this article. The next article will take a more in-depth look on EVDO records, MCC/MNC records for LTE, 800mhz SMR for Network Vision, and much more.
  5. 47 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Thursday, December 19, 2013 - 2:12 PM MST As most of our S4GRU readers are aware, Sprint is pursuing a three pronged approach to LTE. Tri band 25/26/41 LTE 1900/800/2600 -- the first two bands operating as FDD in Sprint and Nextel PCS 1900 MHz and SMR 800 MHz spectrum, the last operating as TDD in Clearwire BRS/EBS 2600 MHz spectrum. A year and a half ago in the early days of Network Vision, S4GRU was the first web site to offer a peek at a live Sprint LTE downlink carrier. We did likewise in running tri band hotspot field tests upon the emergence of Sprint/Clearwire TD-LTE 2600 in Denver this past summer. But the missing piece in the tri band LTE strategy has been the 800 MHz spectrum and the decisive propagation advantages it brings to the table. Then, three weeks ago, S4GRU was able to start sourcing inside info on a few but growing number of band 26 site acceptances around the country. And today, S4GRU presents an exclusive first look at a live Sprint LTE 800 carrier. From the spectrum analyzer RF sweep, we can see that this site has achieved SMR 800 MHz deployment completion. On the left is the 1.25 MHz FDD CDMA1X 800 downlink carrier at band class 10 channel assignment 476, which equates to center frequency 862.9 MHz. On this site, CDMA1X 800 was deployed earlier this year just prior to the Nextel iDEN 800 shutdown. But LTE 800 did not follow -- until now. On the right is the newborn 5 MHz FDD LTE 800 downlink carrier. Temporarily, connections to the LTE 800 carrier are not yet allowed, so an exact EARFCN cannot be determined. But frequency domain analysis suggests a downlink EARFCN 8763, which equates to center frequency 866.3 MHz and is smack dab in the middle of the EARFCN 8761-8765 range that I predicted in one of my engineering screen articles earlier this year. In our Premier sponsors section, S4GRU continues to track ongoing band 26 LTE 800 site acceptances, which should accelerate rapidly over the next several weeks and months. Most progress thus far is in the Chicago, Houston, Kansas, Jacksonville, and North Wisconsin markets. However, LTE 800 will continue to sprout up across the Sprint network -- outside of those areas encumbered by IBEZ restrictions with Canada and Mexico. Below is today's snapshot of the evolving LTE 800 site map. Source: author's field test, S4GRU map data
  6. 39 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Monday, July 16, 2012 - 1:40 AM MDT As Sprint LTE 1900 has become live and discoverable in numerous markets over this past weekend, some of our readers, especially those who are using Android 4.0 ICS based ROMs, have expressed concern at the seemingly low signal levels that they have encountered. For example, see this screenshot from an HTC EVO 4G LTE (under Settings > About > Network): Note the -102 dBm signal level. If this were measuring CDMA1X or EV-DO, then, yes, -102 dBm would be nearing the margin of usable signal. But -102 dBm is actually relatively healthy LTE signal level. To understand why, we need to learn the differences between two types of signal measurement: Received Signal Strength Indicator (RSSI) and Reference Signal Received Power (RSRP). First, an LTE downlink is divided into subcarriers. A 5 MHz bandwidth downlink, which is the configuration that Sprint is deploying, contains 300 subcarriers. And of those subcarriers, one in three carry LTE reference signals. In other words, of the 300 subcarriers, 100 transmit periodic reference signals. To illustrate, I captured this power vs frequency sweep with a spectrum analyzer. The LTE downlink graph comes from a Sprint site in the Kansas City area in late April, well before Sprint stopped blocking devices from live LTE sites. So, the sector depicted here exhibits no data traffic; it is transmitting only the periodic reference signals on 100 subcarriers, which you can clearly count in the graph: Now, RSSI is the more traditional metric that has long been used to display signal strength for GSM, CDMA1X, etc., and it integrates all of the RF power within the channel passband. In other words, for LTE, RSSI measurement bandwidth is all active subcarriers. If we take the above RF sweep of a Sprint 5 MHz bandwidth downlink, RSSI measures the RF power effectively of what is highlighted in yellow: RSRP, on the other hand, is an LTE specific metric that averages the RF power in all of the reference signals in the passband. Remember those aforementioned and depicted 100 subcarriers that contain reference signals? To calculate RSRP, the power in each one of those subcarriers is averaged. As such, RSRP measurement bandwidth is the equivalent of only a single subcarrier. And using our graph once more, RSRP measures the RF power effectively of what is highlighted in red: Since the logarithmic ratio of 100 subcarriers to one subcarrier is 20 dB (e.g. 10 × log 100 = 20), RSSI tends to measure about 20 dB higher than does RSRP. Or, to put it another way, RSRP measures about 20 dB lower than what we are accustomed to observing for a given signal level. Thus, that superficially weak -102 dBm RSRP signal level that we saw previously would actually be roughly -82 dBm if it were converted to RSSI. To conclude, here are a few takeaways about RSSI and RSRP as signal strength measurement techniques for LTE: RSSI varies with LTE downlink bandwidth. For example, even if all other factors were equal, VZW 10 MHz LTE bandwidth RSSI would measure 3 dB greater than would Sprint 5 MHz LTE bandwidth RSSI. But that does not actually translate to stronger signal to the end user. RSSI varies with LTE subcarrier activity -- the greater the data transfer activity, the higher the RSSI. But, again, that does not actually translate to stronger signal to the end user. RSRP does a better job of measuring signal power from a specific sector while potentially excluding noise and interference from other sectors. RSRP levels for usable signal typically range from about -75 dBm close in to an LTE cell site to -120 dBm at the edge of LTE coverage. Sources: 3GPP, author's graphs
  7. 38 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Friday, April 26, 2013 - 6:29 AM MDT A significant piece of S4GRU's educational mission is helping our readers understand what goes on behind the scenes and underneath the hood in the operation of a wireless network. This often requires getting readers to access internal engineering (or debug) screens on their handsets to view numbers and metrics, such as PN offset, Ec/Io, cell identity, etc., as we track the progress of Sprint's Network Vision deployment around the country. So, S4GRU staff thought it long overdue to publish a tutorial on what all of those engineering screen numbers and metrics actually mean. And in this first part of what will hopefully be a long running series, we will examine frequencies, namely center frequencies. First, let us kick things off with CDMA2000 (e.g. CDMA1X/EV-DO). CDMA2000 is divided into band classes. Those band classes basically represent spectrum bands of operation. Some common CDMA2000 band classes familiar to Sprint users include: band class 0 (Cellular 850 MHz), band class 1 (PCS 1900 MHz), band class 10 (SMR 800 MHz), and band class 15 (AWS 2100+1700 MHz). Then, each band class is further divided into carrier channels. These carrier channel numbers represent the actual RF locations -- center frequencies -- of the carrier channels that we use for voice and data services. To illustrate, see the EV-DO engineering screenshot below, specifically the "Channel Number" and "Band Class" fields: Taking into account the band class and carrier channel number, we can use the following formulas to calculate both uplink and downlink center frequencies: uplink center frequency (MHz) = 1850 + (0.05 × carrier channel) downlink center frequency (MHz) = 1930 + (0.05 × carrier channel) In other words, the spacing in between potential carrier channel assignments in band class 1 is 0.05 MHz (or 50 kHz). And the band class 1 range of carrier channel numbers extends from 0-1199. So, using our formulas, the band class 1 carrier channel 100 in the included screenshot has an uplink center frequency of 1855 MHz, a downlink center frequency of 1935 MHz. This FDD paired set of center frequencies falls toward the lower end of the PCS A block 30 MHz license, which is 1850-1865 MHz x 1930-1945 MHz. Next, we can shift over to the 3GPP (e.g. LTE) side, which does things a bit differently. 3GPP sets forth bands, instead of band classes, but otherwise, the functions of bands and band classes are the same. In the US, common 3GPP bands for LTE include: band 4 (AWS 2100+1700 MHz), band 13 (Upper 700 MHz), and band 17 (Lower 700 MHz). But we are most interested in band 25 (PCS 1900 MHz + G block), the band in which Sprint is initially deploying LTE. As with carrier channel numbers in CDMA2000 band classes, 3GPP bands are subdivided into Evolved Absolute Radio Frequency Channel Numbers (EARFCNs). And like carrier channel numbers, EARFCNs indicate center frequencies. However, EARFCNs do so separately for uplink and downlink, as LTE allows for different pairings of uplink and downlink via carrier aggregation. Now, see the LTE engineering screenshot below for its "Band," "UL channel," and "DL channel" fields: Per band 25, we can enter the "UL/DL channels" (i.e. EARFCNs) into the following formulas to determine again both uplink and downlink center frequencies: uplink center frequency (MHz) = 1850 + [0.1 × (uplink EARFCN - 26040)] downlink center frequency (MHz) = 1930 + [0.1 × (downlink EARFCN - 8040)] In this case, spacing between EARFCNs is 0.1 MHz (or 100 kHz). Additionally, the uplink EARFCN range is 26040-26689, the downlink EARFCN range 8040-8689, both for band 25. And in the end, EARFCN 26665 in the included screenshot has an uplink center frequency of 1912.5 MHz, while EARFCN 8665 has a downlink center frequency of 1992.5 MHz. This is an FDD paired set of center frequencies, not a carrier aggregated set, and it resides exactly in the middle of the PCS G block 10 MHz license, which is 1910-1915 MHz x 1990-1995 MHz. In part two, we will take a similar look at center frequencies in the PCS 1900 MHz band's lower frequency cousins, SMR 800 MHz and Cellular 850 MHz. So, stay tuned. Sources: 3GPP, 3GPP2
  8. 38 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Friday, February 22, 2013 - 2:55 PM MST Update: Many hands on reviews of the HTC One are emerging this week. Courtesy of Engadget, we can report that the Sprint variant is one of the very first Sprint LTE handsets to include a removable micro-SIM. Removable SIM cards have long been part of the Network Vision roadmap for 2013, so it looks like that time may have arrived. S4GRU welcomes you to the first major Sprint handset announcement of 2013. Earlier this week, the upcoming HTC One was revealed at an event in New York City. Not to be confused with last year's HTC One X, the HTC One is the new flagship of the line and will be offered by dozens of carriers around the world, including AT&T and T-Mobile in the US. Last year, Sprint got essentially a customized version of the HTC One X in the HTC EVO 4G LTE. This year, however, another EVO handset is not currently in the offing, and Sprint is joining its fellow carriers in standardizing around a universal HTC One platform. The only notable customization is for Sprint's specific CDMA2000 band classes and LTE band. And that Sprint variant had its authorization documents uploaded to the FCC OET (Office of Engineering and Technology) database earlier today. If you have followed our series of articles on the EVO LTE, Samsung Galaxy S3, Motorola Photon Q 4G, LG Optimus G, and Samsung Galaxy Note 2, then you know what is at hand. Here is an RF focused breakdown of the HTC One coming to Sprint: CDMA1X + EV-DO band classes 0, 1, 10 (i.e. CDMA1X + EV-DO 850/1900/800) LTE band class 25 (i.e. LTE 1900; PCS A-G blocks) LTE 5/10 MHz FDD carrier bandwidth LTE UE category 3 802.11a/b/g/n/ac Wi-Fi 802.11n MCS index 7, 40 MHz carrier bandwidth 802.11ac MCS index 9, 80 MHz carrier bandwidth SVLTE support, including SVLTE and simultaneous 802.11a/b/g/n Wi-Fi tether NFC Antenna 0 max RF ERP/EIRP: 20.10 dBm (CDMA1X/EV-DO 850), 23.80 dBm (CDMA1X/EV-DO 1900), 19.23 dBm (CDMA1X/EV-DO 800), 12.30 dBm (LTE 1900) Antenna 1 max RF ERP/EIRP: 13.78 dBm (CDMA1X/EV-DO 850), 13.58 dBm (CDMA1X/EV-DO 1900), 14.27 dBm (CDMA1X/EV-DO 800), 23.63 dBm (LTE 1900) Antenna locations: (see FCC OET diagram below) Simultaneous transmission modes: (see FCC OET diagram below) As for analysis of the specs, the HTC One is the world's first handset to include the new 802.11ac Wi-Fi standard. But let us address right away another potential first that has become the so called elephant in the room. The Sprint version of the HTC One is limited to band 25 LTE 1900. It does not support either of Sprint's upcoming LTE bands -- band 26 LTE 800 and band 41 TD-LTE 2600. One or both of those bands are expected to be incorporated in new handsets sometime this year, but the HTC One will not be the first. The other notable absence is SVDO support for simultaneous CDMA1X voice + EV-DO data, though its omission is growing less and less notable as time goes on. SVDO requires separate RF paths for CDMA1X and EV-DO. The first few Sprint LTE handsets did support SVDO, utilizing separate paths for CDMA1X and EV-DO/LTE. But the last nine Sprint LTE handsets have foregone SVDO, combining CDMA1X/EV-DO on a single path, so SVDO was likely just a temporary measure or a fringe benefit of the Qualcomm MSM8960 chipset and will not be a common Sprint handset feature going forward. In its press release earlier this week, Sprint calls its HTC One an "international" smartphone, and that could be interpreted to mean world roaming capabilities. The FCC authorization documents show no evidence of this, but they are not required to do so, since the FCC is a US authority. What is lacking, though, is any GSM 850/1900 or W-CDMA 850/1900. So, if the HTC One is world roaming capable, it will most likely be limited to GSM 900/1800 and band 1 W-CDMA 2100+1900. Since the HTC One is really the de facto successor to the EVO LTE, a little bit of comparison would be in order. In our RF rundown article on the EVO LTE last spring, we stated that it "does not look to be a stellar RF performer" based on its low to moderate ERP/EIRP figures. And our prediction proved quite prescient, as the EVO LTE has not been noted for its performance with weak signals. The good news is that, on paper, the HTC One looks to be a notable improvement in this regard. First, the dual antenna system is optimized for CDMA1X/EV-DO on antenna 0 and LTE on antenna 1. But as long as only one antenna is in use (i.e. SVLTE is not active), the dual antennas can be switched at will to combat an RF fade at one antenna but not the other. Second, LTE max EIRP has been increased by 4 dB over that of the EVO LTE. Furthermore, LTE EIRP has been maximized around the 1912.5 MHz center frequency, 5 MHz FDD carrier bandwidth configuration that Sprint is currently deploying nationwide in its PCS G block spectrum. In short, the Sprint variant of the HTC One has been tweaked specifically for the Sprint LTE network. Source: FCC
  9. 35 points
    by Travis Griggs Sprint 4G Rollout Updates Monday, May 20, 2013 - 10:15 AM MDT In Part 1 of the "What is a PRL?" series we covered the the initial basics and building blocks of the PRL which covered the 1X portion of the wireless connection. I encourage you to read the article if you have not already done so. You may have seen the various claims of receiving faster or even slower speeds with the mysterious PRL update procedure that seems to randomly happen to our devices. In reality it could be possible that nothing changed in regards to EVDO at all in the PRL. After a PRL update is applied to the device, whether it is pushed from Sprint or user initiated, the CDMA radio will reset just like when airplane mode is cycled on and off. This causes the device to reacquire with the network which could change the site and/or channel the device was using previously. AJ wrote an excellent article, "Can toggling airplane mode actually improve your 3G data speeds?" explaining the EVDO acquisition process. With all that being said let's jump right in and look at a small piece of the PRL to determine how a device connects to the EVDO network. In a mock scenario, the device scanned and did not find signal for the SIDs 22443, 22430, etc but was able to acquire 4159. The device will then check to see if any data records are associated with the connection. The assn tag field for 4159 is a 5. Any records inside this one geo block are checked for the assn tag of 5. In some Sprint PRL versions, the creators have failed to place the EVDO record in the correct geo creating a type of orphaned EVDO record issue, but this is not the case with this example. Record #279 has an assn tag match for the value of 5. The record is analyzed and it is determined that the device will use acquisition record #59 with a 0084:0AC0 subnet and no roaming indicator. If no EVDO signal can be found in the area with this search criteria, the device will fall back to using 1X for data and periodically scan for EVDO. The EVDO subnet is very similar to a SID, but since it is a 128 bit address scheme it offers more combinations than a SID. If needed, the provider could actually assign different priorities to individual sectors of one cell site using the subnet IDs. You may have already noticed multiple SIDs in this block share the assn tag of 5 along with the same acquisition records. The PRL is designed like a relational database where redundant data is shared to save space. So, how does the device know which channels to scan for EVDO? Let's look at the acquisition records of 2 and 59 used by SID 4159. The PCS band channels 50, 75, 100, 175, 200, 250, and 25 are used to scan for SID 4159 1X. These are not the only channels that you device will actually use. These are only used to acquire the initial CDMA handshake. The basestation of the site may direct the handset to rest on channel 25 but during an active phone call channel 150 might be used if the other available channels are at capacity. For EVDO, the device will scan 75, 175, 225, and 250 with a subnet of 0084:0AC0. If another carrier's EVDO signal happens to be on one of these channels it will be ignored as the subnets do not match. Just like on the 1X side and explained in AJ's EVDO article referenced above, the channel scan is only utilized for the initial EVDO handshake. The cell site may have a channel available that is not on the PRL list, which your device could end up using based on the basestation configuration. After attempting to digest all of this material you can see how the new PRL file itself is usually not why the speeds decreased or increased. If the spectrum licenses allowed for it in the area, Sprint could add an additional EVDO carrier channel of 300 to all of the neighboring sites and all of the handsets will be able to use it. The users in this area would probably see faster speeds due to this without a single PRL update. How does EVDO roaming work? In this example the device is connected to SID 4160 which is Verizon Wireless. Using the same analysis explained above, we see a data record of 5 is assigned. "Wait! I thought Sprint used data record 5 already?" This is correct. While your device is roaming on 4160 for 1X connectivity it is also scanning for Sprint EVDO. In order to save on roaming costs, Sprint has designed the PRLs this way. One negative impact on the user is additional battery drain due to this scan combined with the already active scan to find a non-roaming 1X signal. In the standard PRL for residential accounts, EVDO roaming on Verizon Wireless is not allowed. You will only find EVDO roaming on smaller regional CDMA carriers in some areas. On certain corporate accounts, Sprint configures devices with PRLs allowing Verizon Wireless EVDO roaming. While roaming, whether it be 1X or EVDO, the Sprint Terms & Conditions state: "Sprint reserves the right, without notice, to deny, terminate, modify, disconnect or suspend service if off-network usage in a month exceeds: (1) voice: 800 min. or a majority of minutes; or (2) data: 300 megabytes or a majority of kilobytes." Stay tuned for part 3 of the “What is a PRL?” series. We will cover the 800SMR SIDS, 800SMR acquisition records, and the coveted MCC/MNC LTE records shown in the PRL screen shots above.
  10. 34 points
    by Robert Herron Sprint 4G Rollout Updates Thursday, January 31, 2013 - 10:33 AM MST Today we feature text from internal correspondence that was distributed to Sprint employees regarding the state of the Network Vision deployment and addresses key points that employees often encounter with the public. It is from a Q&A session with Chad Elliott, Sprint's Director of Strategic Technology Programs. Although there aren't really points in the memo that will be surprises for S4GRU Members who follow deployment closely, it is helpful to get some sort of official documentation from Sprint that we can now point to explain what is going on. It is a good and concise reference of many key challenges that have impacted Network Vision, with some vague outlook for 2013. Some things discussed in the memo include that production is ramping up and with more launches more frequently, why smaller towns/cities seem to be being upgraded first, issues going on that are slowing down deployment in some areas, etc. Take a look at the memo below:
  11. 31 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Thursday, September 5, 2013 - 5:33 PM MDT About a month ago, our FCC OET reporter, Josh McDaniel, noted that a mystery handset, the LG D820, came and went from the FCC OET (Office of Engineering and Technology). Its authorizations were uploaded, then quickly rescinded, citing confidentiality reasons. Well, today, the LG D820 authorization documents are back. And we are looking at a 3GPP/3GPP2 handset that runs nearly the full North American wireless airlink gamut: GSM 850/1900 W-CDMA 1900/2100+1700/850 (band 2, 4, 5) CDMA1X/EV-DO 850/1900/800 (band class 0, 1, 10) LTE 2100+1700/850/700/1900/800 (band 4, 5, 17, 25, 26) TD-LTE 2600 (band 41) The only notable omission is LTE 750, VZW's currently boutique band 13 -- possibly left out for political reasons, since VZW has a strained relationship with Nexus devices, or for technical reasons, as band 13 has an inverted FDD uplink/downlink duplex. But in a nutshell, this handset looks like it could be headed to AT&T, T-Mobile, and Sprint, covering all of their bases. Here is the kicker, though. One of our moderators, Tim Yu, noted a significant resemblance between the back plate in the FCC OET filing and the back plate of a mystery Nexus device in a widely circulated photo recently from the Google campus. So, you be the judge. Based on the specs and pics, does the the LG D820 look like it could be the upcoming Nexus 5??? More to come... Source: FCC Thread: http://s4gru.com/index.php?/topic/4366-lg-d820-google-nexus-5/
  12. 30 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Thursday, March 28, 2013 - 10:10 AM MDT Update: The Samsung review embargo has been lifted overnight, and Galaxy S4 reviews are being published around the Web today. Thanks to one of our favorite, highly thorough reviewers, Brian Klug at AnandTech, we can confirm that the Galaxy S4 follows the recent HTC One in providing a removable micro-SIM. So, while two data points do not necessarily a trend make, the One and Galaxy S4 do suggest that removable SIMs for Sprint LTE handsets are here to stay. Arguably the most hotly anticipated handset of the year, rivaling even the next iPhone, the Samsung Galaxy S4 in its Sprint variant popped up in the FCC OET (Office of Engineering and Technology) database late yesterday, meaning that the next Galaxy is now authorized to operate in the US and is likely just a few weeks away from a Sprint street date. Not a revolutionary overhaul of the very successful Galaxy S3 platform of last year, the Galaxy S4 maintains a strong family resemblance to its older sibling but does generally and for Sprint specifically add a number of evolutionary enhancements, such as a larger 1080p display, world roaming capability, wireless charging cover functionality, and some transmit power increases. Thus, adding to S4GRU's long standing series of articles on the FCC OET authorizations for the HTC EVO 4G LTE, Samsung Galaxy S3, Motorola Photon Q 4G, LG Optimus G, Samsung Galaxy Note 2, and HTC One is our run through of the RF capabilities of the Galaxy S4: CDMA1X + EV-DO band classes 0, 1, 10 (i.e. CDMA1X + EV-DO 850/1900/800) LTE band 25 (i.e. LTE 1900; PCS A-G blocks) LTE 5 MHz FDD carrier bandwidth LTE UE category 3 W-CDMA bands 2, 5 (i.e. W-CDMA 1900/850) GSM 850/1900 802.11a/b/g/n/ac Wi-Fi 802.11n MCS index 7, 40 MHz carrier bandwidth 802.11ac MCS index 9, 80 MHz carrier bandwidth SVLTE support, including SVLTE and simultaneous 802.11b/g/n Wi-Fi tether RF ERP/EIRP maximum: 25.39 dBm (CDMA1X/EV-DO 850), 23.25 dBm (CDMA1X/EV-DO 1900), 24.62 dBm (CDMA1X/EV-DO 800), 22.83 dBm (LTE 1900) NFC antenna integrated into battery cover CDMA1X/EV-DO Rx antenna diversity Antenna locations: (see FCC OET diagram below) Simultaneous transmission paths: (see FCC OET diagram below) Breaking down the RF specs, honestly, the Galaxy S4 may come across as a disappointment to many. That is primarily, though, because the reality could hardly live up to the expectations. First, the Galaxy S4 does not support band 26 LTE 800 nor band 41 TD-LTE 2600. Reports are that Sprint will not release any dual band LTE devices and will skip straight to tri band. Those devices, however, are still at least six months off, so like all Sprint LTE devices before it, the Galaxy S4 is limited to band 25 LTE 1900 on the native Sprint network. Additionally, the Galaxy S4's band 25 LTE 1900 is limited to 5 MHz FDD bandwidth. This seems to be largely a Samsung quirk, as Sprint LTE devices from other OEMs are tested and authorized for 10 MHz FDD (or greater) as well. That being said, this will likely be of no consequence, as all Sprint LTE FDD deployment for at least the next several years is apt to remain based on 5 MHz FDD carriers. Also, unlike the recent HTC One, the Galaxy S4 does not appear to be particularly optimized for the Sprint LTE network. Using the FCC OET authorization documents, we can gauge a device's RF prowess by examining its maximum transmit ERP/EIRP and at what frequency that max occurs. This is by no means a perfect simulacrum for both transmission and reception, but we can say that Galaxy S4 LTE is at its max RF wise in the traditional PCS A-F blocks, not the PCS G block 1912.5 MHz center frequency where Sprint is deploying its initial LTE carrier nationwide. Staying on ERP/EIRP discussion, the Galaxy S4 looks to be a rather strong performer on roaming CDMA1X/EV-DO 850 and the now being deployed Sprint native CDMA1X 800. Both show quite high ERP. On the flip side, the EIRP for CDMA1X/EV-DO 1900 is good, too, but oddly less than the ERP of the CDMA2000 airlinks below 1 GHz that enjoy significant propagation advantages. With most other handsets, the transmit power relationship is reversed, CDMA1X/EV-DO 1900 transmit power being greater to compensate for its greater path loss. Furthermore, ERP/EIRP was tested with both the standard battery cover and the wireless charging cover. A definite caveat, the wireless charging cover reduces ERP/EIRP by up to 6 dB. Most likely, the induction coil in the wireless charging cover absorbs some of the transmitted RF, thus reducing the radiated power. For some users, the convenience of wireless charging may outweigh the hit to wireless performance. But S4GRU cannot generally recommend wireless charging due to its inefficiency (much power is wasted as heat) and detriment to RF. As for simultaneous voice and data, the Galaxy S4 does support SVLTE but is the latest in a long line of Sprint LTE handsets now to forgo SVDO. Realistically, this comes as no great surprise, as we have not seen SVDO capability in any new handset since last summer. Either this is a limitation of the Qualcomm MDM9615 baseband modem that has become standard equipment or SVDO is no longer a strong priority as Sprint LTE coverage grows weekly. Regardless, CDMA1X and EV-DO share a transmit path (indicated in the FCC OET diagram above); hence, simultaneous CDMA1X voice and EV-DO data is not supported. As S4GRU has reported in the past, the FCC OET authorization documents are not required to disclose world phone capabilities because those bands are not in use in the US. However, the presence of GSM 850/1900 and W-CDMA bands 2, 5 (i.e. W-CDMA 1900/850) is strongly indicative of the inclusion of international roaming capabilities, too. Indeed, other outlets are reporting that all variants of the Galaxy S4 include at a minimum quad band GSM 850/900/1800/1900 and W-CDMA bands 1, 2, 5, 8 (i.e. W-CDMA 2100+1900/1900/850/900) -- the latter supporting DC-HSPA+ on the downlink and HSUPA on the uplink. While we cannot confirm these reports at this time, they certainly do seem plausible. What also remains unconfirmed at this point is the SIM situation: embedded or removable. As soon as this info comes to light, we will update the article. Source: FCC
  13. 29 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Monday, October 21, 2013 - 3:55 PM MDT Clarification: Many readers seem to be confusing the previously authorized and officially announced tri band Samsung Galaxy S4 Mini with this tri band Galaxy S4 reboot, which is full size. The two have quite different model numbers -- the Mini is SPH-L520, while as stated below, the tri band Galaxy S4 is SPH-L720T. Sprint has not yet formally acknowledged the latter, but it has passed FCC OET authorization. And S4GRU expects it to be another tri band handset available before the end of the year. To alleviate the confusion, we are planning an overview article on Sprint Spark tri band handsets, probably to coincide with the November 8 street date next week for the first group of released handsets. Six months ago, S4GRU published an article breaking down the FCC authorization documents for the Sprint variant Samsung Galaxy S4. Many were disappointed that the Galaxy S4, like all other Sprint LTE devices for the past year, was limited to single band 25 LTE 1900. It was not a Sprint tri band LTE handset -- even though band 26 LTE 800 and band 41 TD-LTE 2600 seemed to be right on the horizon because of the impending shutdown of the Nextel iDEN network and the likely approval of the SoftBank-Sprint-Clearwire transaction. Indeed, with the arrival of several Sprint tri band mobile hotspots, TD-LTE 2600 started to become available in metros around the country late this past summer. Today, we bring you another teaser article. The FCC OET database this afternoon uploaded the authorizations for this Samsung model number: SPH-L720T. Now, if you are familiar with the Galaxy S4, you know that its Sprint variant model number is SPH-L720. So, it does not take a genius to put 2 + 2 together. Or in this case, to put S4 + 2 together. Yes, Samsung has just revealed a Sprint tri band Galaxy S4. It carries very similar specs to those of the original Galaxy S4, but it adds two additional Sprint bands: band 26 LTE 800 and band 41 TD-LTE 2600. So, between the previous release of the single band Galaxy S4 and the almost inevitable upcoming release of the tri band Galaxy "S5," look for a tri band updated variant for the Sprint Galaxy S4 in the coming months, probably before the end of the year. Rest assured, this handset will be of popular interest among the faithful -- and possibly despised among those who already used a subsidized upgrade on a single band Galaxy S4 -- so we will run a full FCC OET RF breakdown in the coming days. To head off the obvious questions in the meantime, no SVDO, no SVLTE, as expected. Before we go, though, view the antenna diagram below to see the Sprint tri band LTE goodness. As always, stay tuned... Source: FCC
  14. 27 points
    by Robert Herron Sprint 4G Rollout Updates Wednesday, February 13, 2013 - 1:13 AM MST Often you may see us refer to a GMO site around S4GRU. But, what is a GMO site? GMO stands for Ground Mount Option. Or sometimes, it will be referred to as a GMR (Ground Mount RRU) site. In this article we will explain many points about the Ground Mount Option. In the most basic explanation, a Ground Mount site is one where they are doing a partial Network Vision conversion instead of a full build conversion. A full build site is one where they upgrade all the hardware at a site, including the base station equipment (RBS/MBS), install new multi-mode antenna panels on the tower, add Remote Radio Units (RRU’s, sometimes also called RRH’s), and run new fiber optic lines from the base station equipment up to the RRU’s on the tower. These are the ones most people who follow along Network Vision deployment are familiar with. However, a GMO site will install new base station equipment, with the RRU’s mounted down at the Ground Level, near the new base station cabinets. Then the existing lines running up the tower and the existing panels are reused. These are not to be confused with full build sites with Ground Mounted RRU’s. Those are not Ground Mount Option sites, because they still offer full Network Vision panels, and complete 800MHz and LTE services (where possible). They just are required to mount the RRU’s away from the panels for logistical reasons. How did Sprint determine which sites were to receive the Ground Mount Option instead of a full Network Vision rebuild? I have had the privilege of talking with several Sprint and OEM employees about the Ground Mount Option the past few weeks. Every one of the 38,000+ Sprint sites in the country had a site survey visit in 2011 to establish logistics and planning for the Network Vision upgrade. Each site is broken down to three priorities, largely based on the traffic and carrier count. See the priorities below: : High Priority...site gets full Network Vision upgrade. If site cannot support RRU's and new panels, engineering is done and structure modifications will be made and the site is fully upgraded. Moderate Priority...site gets full Network Vision upgrade. If the site requires minor modifications to support RRU's and NV panels, then it gets fully upgraded. If it requires major attention with full engineering, then a ground mount solution is implemented. Low Priority...low priority sites only get a full NV upgrade with new NV panels and tower mounted RRU's if no structural modification is necessary. If anything is required at a low priority site, the Ground Mount Option is deployed. Also, some low capacity/low priority sites get GMO installs, no matter if the site can support a full install now. At the site survey time back in 2011, each survey team made a judgment call based on their review of the site whether to go full build or GMO, taking into account the priority. And there are anomalies that just do not make any sense. Some markets have no GMO sites at all. And some markets have all GMO sites, like Western Pennsylvania. Also, some site owners will not allow NV full build for various reasons. In these instances, a Ground Mount Option was selected. What are the advantages of a Ground Mount site? The biggest advantage of a GMO site is these sites are being worked on now and getting Network Vision benefits in the middle of the NV program, instead of at the end of the build out. Many 3rd Round Markets have started earlier because of GMO conversions. 1st and 2nd round markets have mostly full build sites with only a few GMO’s, or none at all. This allows some love for customers that would have been pushed off to the very end of Network Vision to see some improvements now. GMO sites are much faster to deploy with no tower work required. Most GMO sites will require minimal permitting from local authorities, or often no permitting at all. Also, GMO's require less negotiation with the site owner, as it does not materially change the site. GMO site conversions are already under way all around the country, and all of them should be completed before the end of this Summer. There are already 100’s of them with 3G upgrades in place. Ground Mount Option sites also will bring LTE much sooner at many locations. Because LTE 1900 can be run on most GMO sites if the appropriate backhaul is available and Sprint has the OEM install the appropriate number of RRU’s or RRU type. The first LTE capable GMO’s are coming online now. Alcatel Lucent has two live, one in New Bern, North Carolina and another one in the Shentel market in South Central Pennsylvania. Samsung has one live in Dayton, Minnesota. This is just the beginning. What are the cons of a Ground Mount Option site? There are a few. The first con with the Ground Mount Option, is there will not be any 800MHz service deployed. Sprint is in the process of adding CDMA 800 voice service to full build Network Vision sites. Sprint will also begin deploying LTE 800 service to full build NV sites before the end of 2013. However, GMO sites cannot support 800MHz service, as the existing tower mounted panels do not support 800MHz. In some rural areas, this is a big disappointment as customers have been waiting for 800 MHz signal propagation benefits in the boonies (like me). The second issue, is the availability of LTE. All full build sites get LTE, but some GMO sites will not be getting LTE deployments. Most GMO sites can support LTE through existing panels, so long as there are not too many CDMA carriers installed. However, some higher capacity GMO sites will not get LTE. Also, some of the most backhaul challenged sites in the Sprint network are GMO sites. They will not get LTE initially because Sprint is unable to get sufficient backhaul to the site to support LTE performance requirements, or in some instances Sprint does not want to go through the difficulty of equipping some sites that are a low priority. The last negative detriment of a GMO site is signal propagation benefits of panel mounted RRU’s. A Network Vision full build site with panel mounted RRU’s can achieve up to a 20% signal gain at 1900MHz. However, the full 20% is only realized at very tall boomer sites with little downtilt. Most sites get more like a 5% signal increase. And these GMO’s will not get that extra signal benefit. Are Ground Mount Options this way forever? Furthermore, at sites where the GMO is implemented, supposedly they will come back at the end of NV and do the engineering and structural modifications. At that time 800 service will be added when the new panels are installed, as well as LTE to sites that can secure appropriate backhaul. I have heard that in some instances (maybe a few hundred), they are using GMO's where they could not come to an agreement with the site owner. Whether financial agreement or logistical/structural. In those instances, Sprint is identifying other adjacent sites that they may move the site to at the end of NV. If no other options can be achieved, it may permanently stay a GMO and never have NV panels and 800 service. My understanding has grown tenfold in the past 2 weeks between talking to the Ericsson tech that's been on site and a long conversation I had with an OEM deployment manager. The most recent conversation I had, the source said they recently heard that more funding is being identified that could go ahead and do more work with GMO sites. Which may include converting them to full builds earlier, or at least changing out legacy panels to NV panels to add support for 800MHz. Differences between vendors Not all GMO sites are the same. Sprint is using three different vendors to deploy Network Vision. Ericsson, Samsung and Alcatel Lucent. Each of these three OEM’s have their own proprietary equipment. Different base station equipment and different RRU’s. Samsung has two types of RRU’s. 800MHz and 1900 MHz RRU’s. Each of the two Samsung RRU types can do both CDMA and LTE from the same unit, supporting up to four carriers each. At a Samsung GMO site, only one RRU is needed per sector, as the RRU can do LTE and CDMA on the same unit. However, Ericsson and Alcatel Lucent do not have it so easy. These two OEM’s cannot run CDMA and LTE on the same RRU. They need a separate RRU for CDMA and LTE on each sector. This is more work and more cost. S4GRU has been told that Ericsson is finalizing a new RRU that can handle CDMA and LTE on the same unit, but they are not in production yet. These are referred to around the forums as RRUS12. Many Ericsson GMO sites have been spotted with only a single RRU per sector. Unfortunately, these have all been RRUS11 units, which cannot support CDMA and LTE together, only in separate RRU’s. Hopefully many of these will get a second RRU still to support LTE, or maybe be switched out with an RRUS12 unit when they start to hit the streets. In closing Some of our members have been quite disappointed to learn that their site was selected for a Ground Mount Option. And I have to admit, I too initially was disappointed myself. Especially since my site is one of the GMO’s that will not receive LTE. At least, at first. The thing that we have to keep in mind is these are sites that are either very low priority or very difficult to upgrade. These were always going to be the very last sites to be touched at all, if at all. The majority of GMO sites probably wouldn’t have started until Spring/Summer 2014. For these sites to receive partial upgrades now is a very good thing. Many of us want everything, and we want it yesterday too. This is not practical though. All things considered, the Ground Mount Option is an elegant solution to the problem. Sprint just needs to push the envelope and install LTE on every one where it is physically possible. Oh and Dan, please add LTE to my GMO site (EP03AL506). It just will take two more RRUS11 units, or possibly a prototype RRUS12 unit. Just imagine the good S4GRU publicity you’d get. I will even arrange the backhaul for you! Ericsson GMO site photo. New Ericsson NV base cabinets in the back and ground mount RRU's on the left. Three CDMA RRU's present here, one for each sector. No LTE at this site initially. Samsung GMO site photo. New Samsung NV base cabinets at the left and ground mount RRU's directly in front. Three RRU's present here, one for each sector. Samsung GMO sites can run CDMA and LTE if set up that way. Alcatel Lucent GMO site photo. New AlcaLu NV base cabinets on the right and ground mount RRU's on the center. Six CDMA RRU's present here, two for each sector (one behind each also). AlcaLu GMO LTE sites will require two RRU's per sector.
  15. 24 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Monday, March 10, 2014 - 8:47 AM MDT After official unveiling at Mobile World Congress in Barcelona a few weeks ago, the Samsung Galaxy S5 made public its authorizations in the FCC OET (Office of Engineering and Technology) database at the start of this weekend. All of the domestic variants are there, including the A3LSMG900P, which in its tri band LTE configuration and "P" designation is the obvious Sprint variant. As expected of a Sprint high end handset, the Galaxy S5 ticks off all of the checkboxes: tri band LTE, UE category 4, global roaming capability, 802.11ac, NFC, wireless charging, etc. It also appears to improve upon the RF output of last year's single band Samsung Galaxy S4 and Galaxy Note 3. From a common uplink EIRP standpoint, the Galaxy S5 can max out up to 3-4 dB greater on band 25 LTE 1900, hitting 26.85 dBm in the middle of the PCS band, falling off 1-2 dB at the extremes of the band. Additionally, band 41 LTE max output looks relatively healthy at 25.03 dBm. In a pleasing move, the FCC authorization docs for the Galaxy S5 do include an antenna diagram -- something that is unfortunately becoming less common, per my mention in the recent HTC M8 FCC article. But in this case, we are able to show a visual of the dual WLAN antennas required for Wi-Fi 2x2 MIMO support, allowing MCS index raw data rates up to 300 Mbps over 802.11n and 866.7 Mbps over 802.11ac. Previously, two spatial stream Wi-Fi has been limited to some laptops and a select few tablets. Thus, the Galaxy S5 is pushing the handset envelope in that regard. See the antenna diagram below: Of course, with no separate CDMA2000 and LTE antennas, as depicted in the diagram above, the Galaxy S5 does not support SVDO nor SVLTE. No surprises there, since Sprint tri band LTE handsets have all been single radio path with e/CSFB. But continuing on the Wi-Fi front, the Galaxy S5 does include a unique simultaneous transmission mode: Wi-Fi and LTE. Now, this is not simultaneous Wi-Fi and LTE in the typical sense that Wi-Fi is used to tether an LTE connection. This is a dual IP stack connection over both Wi-Fi and LTE that Samsung dubs Download Booster, allowing packets to be split up and delivered by both connections, thereby increasing data speeds. Editorially, S4GRU has some concerns about inclusion of the bonded connection Download Booster, since it may engender "unlimited" data users to remain connected to LTE, too, while on secure Wi-Fi at home, work, school, etc. In most cases, Wi-Fi alone is sufficiently fast for all smartphone activities. And that is why S4GRU has long advocated offloading to Wi-Fi -- when/where possible and secure -- so as to help maintain valuable LTE capacity for truly mobile users. That said, we are curious to see the real world implementation of Download Booster before passing judgment. Finally, many hoped that the Galaxy S5 might be the first Sprint handset to support LTE Advanced carrier aggregation because Sprint plans to use its acquired Clearwire spectrum to aggregate multiple band 41 20 MHz TDD carriers. That capability, though, will have to wait for the presumed Samsung Galaxy Note 4, Galaxy S6, or some other device. The carrier aggregation omission is worth mentioning only because the A3LSMG900A variant headed to AT&T does support inter band downlink carrier aggregation. This allows the Galaxy S5 to bond up to 10 MHz FDD of AT&T's low frequency band 17 LTE 700 with up to 10 MHz FDD of its mid frequency band 2 LTE 1900 or band 4 LTE 2100+1700. Indeed, AT&T carrier aggregation is already in use in Chicago, as Gigaom's Kevin Fitchard reported last week. Well, that is the FCC skinny on the upcoming Sprint variant Galaxy S5. Nothing revolutionary on the cellular side of things, but with MIMO and Download Booster, it does offer up some interesting Wi-Fi enhancements. Sources: FCC, Samsung
  16. 23 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Monday, March 3, 2014 - 5:37 PM MST No one is publicly sure what the codenamed HTC M8 will finally be called. HTC One 2, HTC One More, or maybe pull an Apple move and just call it yet again the HTC One. Regardless, all of the big four domestic variants were added to the FCC OET (Office of Engineering and Technology) database today. The last to have its authorizations appear online this afternoon was none other than NM80P6B700 -- the tri band LTE variant undeniably headed to Sprint. As has been our trend over the past six months, we will still call this a teaser article -- albeit make it more extensive than usual. And we may not do a full RF breakdown in the future. Now that tri band LTE and 802.11ac, for example, are de facto standards among top of the line handsets, while SVDO and SVLTE have been laid to rest, there is less news to report on the RF side. But we do want to run a brief RF ERP/EIRP numbers comparison among the high end HTC handsets that have graced the Sprint lineup over the past two years because, well, HTC has developed a bit of a reputation among S4GRU members for losing its lead in the RF performance department. Despite its moniker, the HTC EVO LTE was downright poor on LTE, and the follow up Sprint variant HTC One and HTC One max were average at best. Numbers wise, the HTC M8 looks like a step in the right direction. Per the customary caveats, the available test bench measurements represent only maximum uplink ERP/EIRP, so they do not necessarily reflect the full two way RF performance equation. However, they can provide a decent advance peek inside at the RF proficiency of a handset. In that regard, the HTC M8 offers some improvements over its predecessors. See the table snapshot below (or link to it on Google Docs): https://docs.google.com/spreadsheet/ccc?key=0ArY31Mr219-ydHh0c2xsUWFmbE1udW5vSnlSMjA3TFE&usp=sharing More and more, OEMs are hiding behind the shroud of confidentiality and not allowing public inspection of the antenna diagrams in their FCC OET filings. HTC now appears to have jumped on that bandwagon. Fortunately, the Sprint variant HTC M8 docs do reveal some antenna gain figures, and those numbers are not always divulged, diagrams or not. Of note are unity 0 dBi or positive 1 dBi antenna gains for >1 GHz bands. Compare these to the -3.5 dBi antenna gain for PCS 1900 MHz in the HTC EVO LTE. Additionally, though this is not apparent in the table because it lists only maximum figures, the ranges of max and min ERP/EIRP within the various frequencies in each CDMA2000 band class and within the various carrier bandwidths in each LTE band are more tightly clustered, more consistent than usual. This, likewise, could indicate enhanced antenna engineering. And, finally, the single radio path handsets that have arrived in conjunction with Sprint tri band LTE so far have generally been better RF performers. Will the HTC M8 -- or whatever it gets called -- follow suit? Early returns indicate so, but once S4GRU membership gets its hands on a few samples, field testing in the coming weeks will tell the full story. Source: FCC Thread: http://s4gru.com/index.php?/topic/5008-htc-m8new-flagship/
  17. 22 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Monday, February 3, 2014 - 8:47 AM MST Yes, it has been a while, but welcome to S4GRU's third installment in an ongoing series about the many signal metrics available on those engineering screens hidden inside most mobile devices. Both part one and part two date back to last spring, so check those out if you have not already or if you need a refresher. Part three has been a long time coming mostly for lack of a really relevant topic. But a question was just recently posed in The Forums here at S4GRU about EARFCNs and center frequencies for band 41 TD-LTE 2600. Previously, we covered that 3GPP relationship for band 25 LTE 1900 and touched upon it for band 26 LTE 800, but when we did so, band 41 had not yet made its domestic debut. So, now that band 41 -- christened Sprint Spark -- is being overlaid on Clearwire WiMAX sites in the top 100 markets and tri band LTE handsets are finding their way into more and more Sprint users' hands, it is due time for an educational look at those 20 MHz TDD carriers being deployed across the massive BRS/EBS 2600 MHz band. First, let us take a look at the BRS/EBS band plan itself. Both it and band 41 encompass 2496-2690 MHz for a total of 194 MHz. The BRS spectrum is licensed -- mostly but not entirely in every market to Sprint subsidiaries. The EBS spectrum is also licensed but to educational institutions, which may then choose to lease the spectrum to commercial entities. So, even though band 41 is maximally 194 MHz wide, Sprint does not necessarily control all of that spectrum. And some of that spectrum -- such as the EBS J block and BRS/EBS K block -- is not intended for broadband uses. In other words, contiguity is periodically interrupted. Plus, WiMAX carriers still occupy much of that BRS/EBS spectrum. All told, band 41 in the US is not quite the huge blank slate that some make it out to be for Sprint to deploy 20 MHz TDD carriers. For reference, see the BRS/EBS band plan: Next, we will examine a couple of band 41 engineering screenshots drawn from The Forums: Just as we did for band 25 in part one of this series, we can extract the channel numbers (i.e. EARFCNs) and enter them into an equation to calculate the band 41 center frequencies: uplink/downlink center frequency (MHz) = 2496 + [0.1 × (EARFCN - 39650)] Because this is TDD, not FDD, we need to use only the "DL" channel number. In TDD, there are no separate frequencies for uplink and downlink. The LG screenshot on the left properly indicates the same EARFCN for both uplink and downlink. But good old Samsung "enginerring" on the right registers a different channel for the uplink, EARFCN 58978, a number which is an invalid value. So, when working with TDD, disregard any spurious "UL" channel number. To finish up our calculations, the range for band 41 EARFCNs is 39650-41589, so EARFCN 39991 is toward the low end of the the band, equating to a center frequency of 2530.1 MHz. And EARFCN 40978 comes out to a center frequency of 2628.8 MHz. Separated by nearly 100 MHz, the former is in the lower EBS segment, while the latter is in the contiguous BRS segment, as depicted in the aforementioned band plan graphic. Now, that 20 MHz TDD carrier at EARFCN 40978 is the one that we have documented most commonly across Sprint Spark markets. This was not surprising, since it is deployed in the up to 55.5 MHz of contiguous BRS spectrum that Sprint is licensed, not EBS spectrum that Sprint just leases. That said, we are seeing more and more reports of other EARFCNs, such as EARFCN 39991 detailed above. In other words, the band 41 EARFCN -- unlike the one and only PCS G block band 25 EARFCN -- can vary from market to market because of differences in spectrum licensing/leasing and remaining WiMAX carriers. Sprint's ultimate plan is to deploy multiple 20 MHz TDD carriers per market, putting it in an enviable position for satisfying the public's rapidly growing appetite for mobile data. However, do not misinterpret the multitude of current EARFCNs. We have no evidence to this point that the various EARFCNs indicate multiple 20 MHz TDD carriers in the same market. That is coming but probably will not be widespread prior to the WiMAX sunset slated for no earlier than 2015. In conclusion, S4GRU has created a tracking thread for the various band 41 EARFCNs as they pop up from market to market. Additionally, in our DL Center, we have made available a comprehensive WiMAX/TD-LTE carrier bandwidth and center frequency spreadsheet (screenshot below) that is continually updated as new EARFCNs get reported. If you are interested, we hope that many of you will continue to help us "crowdsource" this band 41 data so that we can get a clearer picture on Sprint Spark and BRS/EBS spectrum utilization. Sources: 3GPP, FCC
  18. 22 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Thursday, November 8, 2012 - 1:10 PM MST Update: Six weeks later, Sprint and U.S. Cellular have finally filed their PCS 1900 MHz license assignment applications in the FCC ULS database. From the filing, we have learned that USCC will not relinquish all of its PCS spectrum in Springfield and Champaign, so the primary market spectrum table below has been updated to reflect that clarification. In a nutshell, Sprint will acquire a consistent PCS B block 20 MHz partition and disaggregation in all affected counties in the Chicago MTA and a consistent PCS A block 10 MHz partition and disaggregation in all affected counties in the St. Louis MTA. For a complete list of the counties included in the spectrum transaction, see this spreadsheet from the FCC filing. Yesterday, Sprint and U.S. Cellular announced an agreement to transfer PCS 1900 MHz spectrum and subscribers in several midwestern markets -- notably, Chicago, St. Louis, Ft. Wayne, South Bend, Springfield (IL), and Champaign -- from USCC to Sprint. While this transaction does entail that USCC will exit its largest and home market, Chicago, it is not a merger. Overall, USCC will give up 585,000 subs but will retain over 5 million current subs, and the deal involves no transfer of wireless infrastructure. Rather, the existing USCC CDMA2000 infrastructure in the affected markets will be retired within approximately two years, as subs are transitioned to the Sprint network. The exact boundaries of the PCS licenses and subs to be transferred from USCC to Sprint have not yet been revealed on a county by county basis. So, this article will be updated once the FCC assignment applications are filed or any other further info arises. In the meantime, know that this is a spectrum transaction, bar none. Chicago is Sprint's largest market in which it holds only 20 MHz of PCS A-F block spectrum. In nearly all other top 10 markets, Sprint holds 30 MHz of PCS A-F block spectrum. And Ft. Wayne is a proverbial red headed stepchild market -- Sprint's only top 100 market with only 10 MHz of PCS A-F block spectrum. So, most importantly, this transaction provides a 20 MHz PCS injection into Sprint's spectrum holdings in both Chicago and Ft. Wayne. For a look at the five largest markets included in the deal, see the spectrum table below: Moreover, Sprint's existing PCS D block 10 MHz and PCS E block 10 MHz licenses in Chicago are non adjacent. As such, Sprint has to run an extra set of guard bands -- one set of guard bands for each license. Those extra guard bands take up valuable spectrum, limiting Sprint to only six instead of seven CDMA2000 carriers in its 20 MHz of spectrum and leaving more sites in Chicago spectrum constrained than in any other big market. Synergistically, though, the PCS B block 20 MHz license that Sprint will acquire in Chicago is directly adjacent to its existing PCS D block 10 MHz license, giving Sprint a fully 30 MHz contiguous swath of PCS spectrum, which will allow Sprint to deploy additional CDMA2000 carriers and larger LTE bandwidth (10-15 MHz FDD) when the time comes to add LTE capacity. See the license contiguity in the band plan diagram below: Speaking of LTE, that is one of the key reasons why USCC is willing to part with its Chicago market. In most of its markets, USCC holds some combination of Cellular 850 MHz, PCS 1900 MHz, AWS 2100+1700 MHz, and Lower 700 MHz spectrum. But in Chicago, USCC controls only the aforementioned 20 MHz block of PCS spectrum. USCC entered the Chicago market just 10 years ago when it acquired PrimeCo, which had been divested as part of the merger that created Verizon Wireless. Since then, USCC has been unable to acquire additional spectrum in Chicago, leaving it effectively incapable of deploying LTE in its largest market while continuing its CDMA2000 operations. So, the deal with Sprint provides an exit strategy for Chicago in what was otherwise a dead end market for USCC. In the other five of the six markets detailed above, USCC likely could roll out LTE, as it holds additional AWS 2100+1700 MHz and/or Lower 700 MHz licenses in those markets. It should be noted, however, that those non PCS licenses are not being transferred to Sprint in this deal. But as it exits those markets, USCC will almost surely look to sell the other licenses, too, with VZW and T-Mobile being likely buyers for the AWS spectrum, AT&T a strong possibility for some of the 700 MHz spectrum. Sources: FCC, USCC, Sprint
  19. 22 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Friday, May 10, 2013 - 12:35 PM MDT Welcome back to S4GRU's continuing series focused on understanding many of the signal metrics displayed on your devices' engineering screens. If you missed part one a few weeks ago, that is a good place to start for background info. Last time out, we covered 3GPP2 band class 1 PCS 1900 MHz, in which Sprint has long operated its CDMA2000 network, and 3GPP band 25 PCS 1900 MHz, in which Sprint is currently deploying its LTE network. Today, let us begin with the last of Sprint's current native spectrum usage -- 3GPP2 band class 10 SMR 800 MHz. This is what 3GPP2 also calls the "Secondary 800 MHz band," and we will understand why when we finish up with band class 0 Cellular 850 MHz a bit later today. First, take a look at the following CDMA1X engineering screenshot: This handset is camped on Sprint's brand new band class 10 CDMA1X 800 overlay, which is replacing iDEN 800 and is currently available in select markets around the country. Now, as we did last time, we can take into account the band class and carrier channel number, then use the appropriate formulas to calculate both uplink and downlink center frequencies: uplink center frequency (MHz) = 806 + (0.025 × carrier channel) downlink center frequency (MHz) = 851 + (0.025 × carrier channel) In other words, the spacing in between potential carrier channel assignments in band class 10 is 0.025 MHz (or 25 kHz). This is due to the SMR 800 MHz band's legacy of dispatch and iDEN, both of which conform to 25 kHz channelization. And the band class 10 range of channel numbers extends from 0-719. So, using our formulas, band class 10 carrier channel 476 in the included screenshot has an uplink center frequency of 817.9 MHz, a downlink center frequency of 862.9 MHz. This is the one and only band class 10 carrier channel that Sprint will employ across most of the country. In parts of the Southeast where SouthernLINC also operates in rebanded SMR 800 MHz spectrum, Sprint users will instead see band class 10 channel 526, which has uplink and downlink center frequencies of 819.15 MHz and 864.15 MHz, respectively, just as S4GRU detailed in an article a year ago. As for band 26 LTE 800, well, that should be coming online in the next several months, but no devices are yet available. So, for both of those reasons, we cannot post any engineering screenshots. What we can anticipate, however, based on SMR 800 MHz spectrum constraints, is that Sprint's 5 MHz FDD LTE 800 carrier likely will be centered somewhere in the 821.1-821.5 MHz x 866.1-866.5 MHz ranges, translating to uplink and downlink EARFCN ranges of 26761-26765 and 8761-8765, respectively. I will be out in the field with my spectrum analyzer in the coming months, ready to capture and publish a first peek at the LTE 800 carrier. And expect a follow up article on LTE 800 engineering later this year. Now, let us conclude with a look at Sprint roaming service in 3GPP2 band class 0 Cellular 850 MHz. Or this is what 3GPP2 has traditionally referred to as the "800 MHz band." And that, as I piqued earlier, is why the "Secondary 800 MHz band" name comes into play for band class 10 SMR 800 MHz. In more recent years, the "800 MHz" nomenclature has become problematic, as it makes distinguishing between band class 0 and band class 10 difficult for less informed users. For a good example of this, see the iPhone 4S tech specs, which mislead many into thinking that it supports Sprint's band class 10 CDMA1X 800 overlay. For this reason, I have long advocated using "Cellular 850 MHz" as distinct terminology. That background aside, let us examine a CDMA1X engineering screen of a Sprint device roaming on VZW: This handset is idling on channel assignment 425. Again, we can use the appropriate formulas to calculate both uplink and downlink center frequencies: uplink center frequency (MHz) = 825 + (0.03 × carrier channel) downlink center frequency (MHz) = 870 + (0.03 × carrier channel) So, that VZW channel 425 is centered at 837.75 MHz x 882.75 MHz, which is toward the bottom of the Cellular B block license, as we will see in just a moment. First, in the Cellular 850 MHz band, channelization is 0.03 MHz (or 30 kHz), as that dates back to the original analog AMPS standard, which used 30 kHz FM channels and got us all started on this cellularized wireless network journey. Second, we encounter a complication with band class 0. The above formula works only for a subset of channel assignments, 1-799. For channel assignments 991-1023, we have to use slightly modified formulas: uplink center frequency (MHz) = 825 + [0.03 × (carrier channel − 1023)] downlink center frequency (MHz) = 870 + [0.03 × (carrier channel − 1023)] The reason for this complication is complicated itself. When the FCC originally created the Cellular 850 MHz band plan in the 1980s, it was 825-845 MHz x 870-890 MHz, divided into two equal 10 MHz FDD (10 MHz x 10 MHz) licenses: Cellular A block (825-835 MHz x 870-880 MHz) and Cellular B block (835-845 MHz x 880-890 MHz). Each block consists of 333 AMPS channels, A block covering 1-333, B block running 334-666. Not long after, the FCC expanded the Cellular 850 MHz band, but it could not do so by simply adding spectrum exclusively at the bottom or the top of the band plan. Because of spectrum constraints and equal license bandwidth, the FCC had to add a sliver at the bottom of the band plan and two at the top of the band plan. The additions became known as "A low," "A high," and "B high." See my band plan graphic below: Since "A high" (1.5 MHz FDD) and "B high" (2.5 MHz FDD) continue as upper end extensions of the band plan, they follow the original center frequency formula, adding channels 667-799. "A low" (1 MHz FDD) tacked on at the bottom of the original plan is the anomaly. It requires its own center frequency formula and adds channels 991-1023. Also, note the missing channels 800-990. Those are a mystery, unbeknownst even to me. Additionally, because it is only 1 MHz FDD, "A low" is not frequently used for CDMA2000 carrier channels, which are always 1.25 MHz FDD in bandwidth. So, many of the carrier channel assignments in "A low" are invalid, since they would cause the CDMA1X or EV-DO carrier to extend off the lower edge of the band. If "A low" is utilized, the only permissible channel assignments are 1013-1023, all of which cause the CDMA2000 carrier to extend into the original A block. So, if you ever encounter a band class 0 channel assignment in the 1013-1023 range, you have found something of a rare bird. Well, that covers the relationships among bands, band classes, carrier channel assignments, EARFCNs, and center frequencies. Next time, we will turn our attention to another signal metric. I am thinking maybe SIDs and NIDs or PN offsets but have not decided yet. See you then... Sources: 3GPP, 3GPP2, FCC
  20. 21 points
    by Robert Herron Sprint 4G Rollout Updates Wednesday, July 18, 2012 - 12:59 PM MDT As many of you already know, the ability to connect and keep connected to LTE signals in Sprint's launch markets has been problematic for Sprint customers. What's going on? Here at our forums at S4GRU, we have been busy talking with our members and trying to figure out the issue since LTE markets started going live last Thursday. Some members were able to connect early and often. Some had problems and were able to eventually connect. Some have been unable to stay connected. Some have never connected, even standing right next to a live LTE site. We have now complied a lot of data from our members and have drawn some observational conclusions. There are two main issues, connecting to LTE can be a challenge on some devices (especially the EVO LTE), and the signal thresholds are not optimum to keep connected to LTE before getting pushed back to the 3G EVDO network. I am in a confirmed LTE area with a strong signal but cannot connect Some LTE devices just do not want to connect to LTE. With the EVO LTE, some of our members have complained that they even stood next to a confirmed operating LTE site and the 4G icon would not appear. For most of these folks though, cycling from CDMA/LTE mode, back to CDMA only mode and then back to CDMA/LTE mode forces the phone to look for a LTE signal. This has worked for most people I have talked with who have an EVO LTE and know they are in a strong LTE signal area. This also seems to help some other LTE devices sometimes. There most likely is an issue where these devices are not scanning for LTE service like they should be. When you go out of CDMA/LTE mode and then come back in, the first thing the device does is scan for LTE service. Hopefully, Sprint will get an OTA out that fixes this issue soon. I can connect to LTE but it goes back to 3G EVDO Other members have discussed how they can connect to 4G LTE, whether automatically or by forcing it (as described above), but it goes back to 3G EVDO. They cannot keep a Sprint LTE connection. This may happen right away, or this may happen as they drive down the road, or after they pick up the device. By the best we can tell, this is being caused by the LTE signal thresholds programmed in the device(s). When your device is in CDMA/LTE mode, it seeks LTE first. If it finds LTE, it should connect, but if it doesn't have a strong enough signal, it shunts the user off to 3G, sometimes in seconds. The devices have a minimum LTE signal programmed in them (it appears to be somewhere in the midrange and could be slightly different between devices). If the LTE minimum signal that is programmed is maintained, the device stays connected to LTE, no problems in most instances. So if you have a strong LTE signal after you connect, you will most likely keep it. However, if you have a midrange LTE signal, you will likely get bumped back down to 3G EVDO if you should move farther away from the signal, or maybe even pick up the device. Most devices will drop a few dBm of signal when being held, and it could be enough to move you back down to 3G. If you have a weak LTE signal, then you are almost doomed. You will not likely be able to stay connected to LTE for very long and most likely be shunted back to 3G pretty quickly. An exception to this we noted is if you are also in a weak 3G area. But if you have a strong 3G signal and a weak LTE signal, you are most likely not going to be able stay connected with the current thresholds programmed. What can Sprint do? I know this is very frustrating for most of you. We all want it to work flawlessly. But this is part of the process. Sprint needs our constructive feedback so they can make changes. There are drawbacks to being early adopters. That is the point of this article. To wrap up some meaningful and thoughtful feedback from our members into a single comprehensive piece and give to Sprint for them to work on. As well as educate our members and readers into some of the information we have been able to determine. Sprint, please work with your OEM's right away on OTA's to adjust the thresholds for keeping LTE signals, as well as addressing the cycling through CDMA only mode to get devices to make initial LTE connection. One key point that I hear over and over again is that LTE customers would rather have a weak LTE signal than a strong 3G signal. Weak LTE still performs better than even good 3G in most instances. If anyone from Sprint would like to reach out to me and provide any updates of what they are doing to address the problem, I would love to receive a PM, email or Direct Message on Twitter. I will then be happy to provide an update for our members and readers. What can we do now? With WiMax devices, we could actually change the WiMax signal threshold ourselves. However, we have not been able to locate anywhere in the LTE devices where that can be done. So we are in a waiting position to see if Sprint will help us. If you absolutely cannot connect to LTE or stay connected to LTE, you can force your device into LTE only mode. If you do this, you will lose access to 3G EVDO and 1x services while in this mode. But it is reversible at any time. It requires your device's MSL code, though. You can get your MSL code by using some apps like MSL Reader, or by using a Terminal Emulator with some models. You can search the web for ways to get your MSL code of your particular device. Some people can even get it from Sprint CSR's. Once you have your MSL code, go into your phone dialer and enter ##DATA# (*#*#DATA#*#* on the Galaxy Nexus). A menu will open and ask you to select Edit or View. Select Edit. Enter your MSL number. Now in each device, there may be some variability in the next steps. Select the Others button, then choose HDR/1X selection. Select LTE Only mode. Now you will only be able to connect to LTE, no 3G or 1x. However, you will not be able to make/receive phone calls or text in this mode. Data only. Once in LTE Only mode, you will only be able to connect to LTE signals. Even weak LTE signals. And you will not have to worry about getting shoved off into 3G. This will allow you to test your LTE, and make sure your LTE is indeed working in your device. You could stay parked in this mode if you wanted to. You just wouldn't be able to use the phone or text. To restore to normal, just go back in the same way and select LTE/CDMA or LTE/CDMA/EVDO, depending on your device. If going into LTE only mode you are still unable to connect to LTE, then you are either not in LTE coverage like you thought you were, or your device has a problem. This is all we know at this time. We will update with more information as we learn it. Thank you to all who helped gather this information from the field. You guys are what make S4GRU an awesome place!
  21. 21 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Friday, August 17, 2012 - 1:14 PM MDT CDMA1X and EV-DO carrier channels are shared resources. In CDMA1X, many subscribers share the same carrier channel, their individual traffic kept theoretically orthogonal by code division. Likewise, in EV-DO, individual traffic is separated by time division. But what happens when Sprint (or any other CDMA2000 network provider) has deployed greater than one CDMA1X and/or EV-DO carrier channel on a given cell site? How does your handset determine which carrier channel to utilize? You might like to think that your handset would automatically choose the least loaded CDMA1X and/or EV-DO carrier channel. But that is not really the case. Instead, when multiple carrier channels are available, each cell site broadcasts a channel list message of the available carrier channels on that site. Upon receiving this list of multiple carrier channels, each handset then invokes a hashing algorithm to select which carrier channel to use. Think of it like a multi lane highway, but each car must choose a particular lane based on the car's license plate number. For CDMA1X, the hashing algorithm -- which is a kind of pseudo random number generator -- is seeded with the handset's ESN or the subscriber's MDN/MSID (i.e. phone number). Unless the subscriber changes devices or phone numbers, these values remain static, hence the carrier channel hash is quite predictable. And Sprint, for reference, seems to use MDN/MSID based hashing. Nearly a decade ago, I built a spreadsheet that emulates the CDMA1X hashing algorithm, downloadable as an XLS file. However, for EV-DO, the carrier channel hash is not quite so outwardly predictable. To seed the hashing algorithm, EV-DO uses a session number, which obviously varies from data session to data session. Each time that a handset powers up, crosses a SID/NID boundary, or even toggles airplane mode, for example, generates a new EV-DO data session, hence a new session number. And it is this session number that determines the output of the hashing algorithm. To demonstrate this process, I positioned myself in one location about a quarter of a mile distant from the north sector of a local cell site. Over the course of several minutes, I grabbed three screen caps of the EV-DO engineering screen on one of my handsets. In between each screen cap, I cycled airplane mode at least once, each cycle generating a new data session. In the span of four minutes, I was able to get my handset to hash to each of the three EV-DO carrier channels deployed on this site. When I arrived at the site, my handset hashed to PCS 0175, which is the third EV-DO carrier channel (F3) in the channel list message. The second and third hashes after toggling airplane mode several times were to PCS 0150 (F2) and to PCS 0100 (F1). See the Channel Number field depicted in the screen caps: In addition, here is a raw RF look with a spectrum analyzer at the seven CDMA2000 carrier channels deployed on this cell site sector: The four CDMA1X carrier channels are PCS 0050, PCS 0075, PCS 0125, and PCS 0200. As is oft the case, the three aforementioned EV-DO carrier channels -- PCS 0100, PCS 0150, PCS 0175 -- are distinguishable by their slightly higher RF power output. Furthermore, for those curious, PCS 0025 (at the far left of the graph) and PCS 0225, PCS 0250, and PCS 0275 (at the right of the graph) are fallow spectrum on this site. If deployed, PCS 0025 would be the next EV-DO carrier channel (F4), PCS 0275 the final EV-DO carrier channel (F5), while PCS 0225 and PCS 0250 would be additional CDMA1X carrier channels. Back to the hashing algorithm, while it attempts to distribute users more or less evenly among available EV-DO carrier channels, it does not take into account several other factors, such as loading and backhaul. For example, if you are stuck on a carrier channel and sector with a few data hogs who have stronger signal than you do, your data speeds will likely suffer as the "fair and proportional" scheduler integral to the EV-DO airlink attempts to maximize total throughput by allocating greater time slots to the users with better signal quality. Additionally, backhaul may not be distributed evenly among deployed carrier channels, so it is possible that some carrier channels may have inherently greater data capacity than others do. Another benefit of rehashing to a different carrier channel is that you may be able to connect to a closer cell site. Because not all cell sites have the same number of deployed EV-DO carrier channels, carrier channel hashing is an imperfect process. To illustrate, the cell site (call it cell site "A") that I detailed above for this trial has three EV-DO carrier channels (F1, F2, F3), as duly noted. But the adjacent cell site to the north (call it cell site "B") has only two EV-DO carrier channels (F1, F2). A handset that hashes to F3 on cell site "A" will cling to carrier channel PCS 0175 even as it moves north well into the coverage area of cell site "B." Interference will not be a problem, as cell site "B" does not transmit PCS 0175, but signal strength (and data speeds) will diminish until cell site "A" drops below a network defined threshold, at which point the handset will handoff to cell site "B" and hash to PCS 0150. This can require substantial movement and/or time. So, if you always want the most crisp EV-DO handoffs, you can try to ensure that your handset always hashes to F1, the EV-DO carrier deployed on essentially every site in the market. To conclude, by no means is airplane mode a panacea for slow 3G data ills. EV-DO carrier channel deployment and backhaul can vary from site to site, while loading can also vary from site to site, even from minute to minute. And EV-DO networks in some cities are just generally overloaded. But if you are at work, in a restaurant, at a park, etc., and find yourself with unbearably slow 3G data or lower than usual signal strength for that location, try toggling airplane mode. A 30 second on/off cycle of airplane mode will start a new data session and could get your handset to rehash to another EV-DO carrier channel that is on a closer site, has better backhaul, and/or is currently less loaded. Sources: Qualcomm, author's field data
  22. 20 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Tuesday, September 3, 2013 - 11:35 AM MDT Welcome back from summer vacation. The S4GRU writing staff, too, has returned and is ready to catch up on some of the developments from over the last two or three months. In late July, Sprint released two tri band LTE mobile hotspots and one tri band LTE USB dongle. Meanwhile, Clearwire had lit up band 41 TD-LTE 2600 on numerous sites in several markets around the country. Around the same time, the LG G2 became the first Sprint tri band LTE handset to pass through FCC OET (Office of Engineering and Technology) authorization. As is our tradition by now, we add to S4GRU's stalwart series of articles on the FCC OET authorizations for the HTC EVO 4G LTE, Samsung Galaxy S3, Motorola Photon Q 4G, LG Optimus G, Samsung Galaxy Note 2, HTC One, and Samsung Galaxy S4 our look at the RF faculties of the LG G2: CDMA1X + EV-DO band classes 0, 1, 10 (i.e. CDMA1X + EV-DO 850/1900/800) LTE bands 25, 26, 41 (i.e. LTE 1900/800, TD-LTE 2600) band 25 LTE 3/5/10 MHz FDD carrier bandwidth band 26 LTE 1.4/3/5/10 MHz FDD carrier bandwidth band 41 TD-LTE 10/15/20 MHz TDD carrier bandwidth LTE UE category 4 W-CDMA bands 2, 5 (i.e. W-CDMA 1900/850) GSM 850/1900 802.11a/b/g/n/ac Wi-Fi 802.11n MCS index 7 (single spatial stream, 40 MHz carrier bandwidth, 400 ns guard interval) 802.11ac MCS index 9 (single spatial stream, 80 MHz carrier bandwidth, 400 ns guard interval) SVDO and SVLTE support absent RF ERP/EIRP maximum: 19.80 dBm (CDMA1X/EV-DO 850), 21.64 dBm (CDMA1X/EV-DO 1900), 23.09-27.08 dBm (LTE 1900), 17.77-21.29 dBm (TD-LTE 2600) CDMA1X/EV-DO Rx antenna diversity NFC antenna integrated into battery cover Antenna locations: (see FCC OET diagram below) Simultaneous transmission paths: (see FCC OET diagram below) The LG G2 is not only the first revealed Sprint tri band LTE handset but also the first Sprint category 4 UE of any kind because it is utilizing the Qualcomm Snapdragon 800 (MSM8974). The MSM8974 is a 28 nm process SoC that contains processor, cellular baseband, and WLAN/GNSS baseband all on one chipset -- à la the MSM8960 that dominated the first half of last year. The difference, of course, is that the MSM8974 has a quad core processor and a UE category 4 cellular baseband. The latter supports up to 150 Mbps on the downlink, 50 Mbps on the uplink -- though those speeds will not likely be seen on Sprint during the lifespan of this handset due to spectrum bandwidth constraints. Also, we cannot confirm at this point that the G2 is actually using the internal WLAN/GNSS baseband capabilities of the MSM8974 and not a separate chipset solution from Broadcom, as has been the trend with the HTC One and Samsung Galaxy S4. But having at least both processor and cellular baseband on the same chipset should be a step in the right direction. Regarding maximum ERP/EIRP, we are slightly modifying the way that we report those figures. Even within a given frequency band, max ERP/EIRP can vary according to frequency and modulation. So, if the FCC OET docs show greater than 1 dB of variance in a certain band and airlink, we now report that range instead of a single max figure. As we have stated in the past, FCC OET testing includes only transmitters, not receivers. Thus, we have to extrapolate overall RF prowess based on mobile uplink transmission capabilities, and that is an inexact science. In a nutshell, though, the CDMA2000 band class 0 and 1 power outputs appear to be a bit on the weak side, while the band 25 LTE power output looks good. The band 41 TD-LTE EIRP is lower than we would like to see -- especially considering BRS/EBS 2600 MHz propagation characteristics -- but TD-LTE 2600 will be used largely as an offload band for handsets, probably less so for hotspots. What is missing are band class 10 CDMA2000 and band 26 LTE ERP figures. The original FCC OET filing in late July included only conducted power figures for that band class and band. Conducted power is what is delivered to the antenna, not what is actually radiated from the antenna. And even though the G2 at the end of August has already had one Class II Permissive Change filing, we expect at least one more Class II filing with additional ERP figures before this Sprint variant hits the streets. Last year, the LG Optimus G had its initial, incomplete FCC OET filing in the summer, followed by a Class II change in the fall and a release in November. In our Optimus G article last summer, we wrote not to expect the street date right away, and we will offer the same caveat here. The rumored November release for the Sprint variant G2 should probably be the expectation. Source: FCC
  23. 18 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Tuesday, September 10, 2013 - 1:10 PM MDT To cut right to the chase, Apple announced at its live event today two new iPhone models: high end iPhone 5S and mid range iPhone 5C. Both are coming to Sprint and both support Sprint LTE -- but only dual band Sprint LTE. The new crop of iPhone models for the next year will not be tri band LTE handsets on Sprint. The LTE bands supported by iPhone 5S (A1453) and iPhone 5C (A1456) are substantial and as follows: band 1, 2, 3, 4, 5, 8, 13, 17, 18, 19, 20, 25, 26. For Sprint users, that means dual band LTE 1900/800. While 13 total bands seems impressive, a few of those bands -- such as band 2/25 and band 5/26 -- are subset/superset bands. The big takeaway for Sprint users, though, is that band 41 is absent this year. So, TD-LTE 2600 will be coming soon to several tri band Android handsets but not to the dual band two new iPhone models. Band 38 TD-LTE 2600 is limited to the Asia/Oceania variants. Also worthy of note, Sprint and SoftBank share the same iPhone 5S and iPhone 5C variants this year. Whether that is merely coincidence is impossible to determine. But Sprint and SoftBank have talked about combined economy of scale as a benefit of their tie up. To conclude, the new iPhone FCC OET docs have not yet trickled out, but as they do, we will have more info to come. Stay tuned... Source: Apple Thread: http://s4gru.com/index.php?/topic/4442-the-iphone-5s-iphone-5c-not-tri-band-lte-was-next-iphone-to-be-announced-on-september-10/
  24. 18 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Friday, June 1, 2012 - 11:58 PM MDT Update: Sprint has scheduled an exclusive Samsung event for the evening of June 12 in Boston. It looks like the Samsung Galaxy S3 coming out party has been set. Just as the HTC EVO 4G LTE is setting up for its delayed national street date tomorrow June 2, it may sooner than expected be getting another high profile cousin in Sprint's burgeoning line up of Network Vision ready, LTE capable devices. First, word leaked this afternoon that Sprint is prepping landing and pre-order web pages for its version of the Samsung Galaxy S3. Then, this evening, S4GRU uncovered the Samsung SPH-L710 (aka Galaxy S3) exhibits that had hit the FCC OET (Office of Engineering and Technology) database earlier today. So, the ball seems to be rolling toward an imminent launch for the Galaxy S3. And, as we did with the EVO 4G LTE a few weeks back, here is an RF focused technical rundown of the upcoming Samsung flagship Sprint handset: CDMA1X + EV-DO band classes 0, 1, 10 (i.e. CDMA1X + EV-DO 850/1900/800) LTE band 25 (i.e. LTE 1900; PCS A-G blocks) LTE 5 MHz carrier bandwidth LTE UE category 3 SVDO and SVLTE support, including SVDO or SVLTE and simultaneous 802.11a/b/g/n Wi-Fi tether Maximum RF ERP: 17.78 dBm (CDMA1X 850), 20.77 dBm (EV-DO 850), 24.05 dBm (CDMA1X 1900), 23.85 dBm (EV-DO 1900), 17.21 dBm (CDMA1X 800), 17.56 dBm (EV-DO 800), 22.01 dBm (LTE 1900) NFC antenna integrated into battery LTE antenna configuration: 1 Tx, 2 Rx (i.e. 2x2 downlink MIMO) All in all, the Galaxy S3 does not present any really big technical surprises. As RF capabilities go, it follows very closely in the footsteps of the EVO 4G LTE. One of the few notable differences is that the Galaxy S3, like its Galaxy Nexus sibling, supports only 5 MHz x 5 MHz LTE carriers, while the EVO 4G LTE can do both 5 MHz x 5 MHz and 10 MHz x 10 MHz LTE bandwidths. However, Sprint has no definite plans to deploy 10 MHz x 10 MHz LTE during the typical lifespan of either of these handsets. And, otherwise, the Galaxy S3 does appear to have the general edge in RF transmit power. Sources: FCC, Inside Sprint Now
  25. 17 points
    by Andrew J. Shepherd Sprint 4G Rollout Updates Wednesday, September 18, 2013 - 2:40 PM MDT S4GRU has been on a bit of a roll of late. We had an unusually quiet summer here on The Wall, as our writing staff was frequently away on summer vacation. But we have come back with a vengeance. Earlier this month, we got the next Nexus handset scoop on the rest of the tech press with our discovery of the reappearance of the LG D820 authorization docs at the FCC OET (Office of Engineering and Technology) and supposition that the D820 circumstantially has a striking amount in common with the presumed upcoming Nexus 5. Additionally, last week, we brought you a quick dissection of the two new Sprint variant iPhone models with dual band LTE only minutes after Apple released the specs, not to mention, another FCC OET discovery -- a tri band LTE Samsung Galaxy Mega 6.3 clearly headed to Sprint All three of those articles were "Teasers." We have decided to publish shorter articles more often when longer articles may not be immediately feasible for our writing staff. Then, down the road, expect more in depth follow up pieces, such as on the Nexus 5, iPhone 5S/5C, and Galaxy Mega 6.3. Today, though, we bring you another full length FCC OET airlink article on the presumed Sprint variant Samsung Galaxy Note 3. Late on Wednesday last week, authorizations for multiple Samsung Galaxy Note 3 variants -- all bearing the SM-N900 base numbering scheme -- started appearing in the FCC OET database. Other tech outlets had previously gotten a list of the model numbers and tied the N900A to AT&T, N900T to T-Mobile, N900V to VZW, and N900S to Sprint. But while the "A," "T," and "V" variants all arrived in the FCC OET, the "S" variant did not make an appearance, and additional info suggested that the "S" variant is instead for SK Telecom in South Korea. Meanwhile, the N900R4, the "R4" variant, also had its authorizations posted to the FCC OET. And at least one tech site incorrectly pegged it as the one for Sprint and other CDMA2000 operators. But that clearly missed the absence of band class 10 CDMA1X 800, which has been a staple of Sprint devices for roughly two years now. In the end, the "R4" variant is coming to CDMA2000 operators, but USCC and C Spire are the likely destinations. So, that left the N900P, the "P" variant, for Sprint, and the circumstantial evidence of being the only version with band class 10 CDMA1X 800 supports that inference. Thus, adding to S4GRU's continuing series of articles on the FCC OET authorizations for the HTC EVO 4G LTE, Samsung Galaxy S3, Motorola Photon Q 4G, LG Optimus G, Samsung Galaxy Note 2, HTC One, Samsung Galaxy S4, and LG G2 is Sprint's expected Samsung Galaxy Note 3: CDMA1X/EV-DO band classes 0, 1, 10 (i.e. CDMA1X/EV-DO 850/1900/800) LTE band 25 (i.e. LTE 1900) LTE 5/10 MHz FDD carrier bandwidth LTE UE category 4 W-CDMA bands 2, 5 (i.e. W-CDMA 1900/850) GSM 850/1900 802.11a/b/g/n/ac Wi-Fi 802.11n MCS index 7 (single spatial stream, 40 MHz carrier bandwidth, 400 ns guard interval) 802.11ac MCS index 9 (single spatial stream, 80 MHz carrier bandwidth, 400 ns guard interval) SVLTE support, including SVLTE and mobile hotspot (both 2.4 GHz and 5 GHz) SVDO absent RF ERP/EIRP maximum: 19.82-20.93 dBm (CDMA1X/EV-DO 850), 18.91-21.30 dBm (CDMA1X/EV-DO 1900), 21.85-23.63 dBm (LTE 1900) RF conducted power maximum: 24.75 dBm (CDMA1X/EV-DO 800) CDMA1X/EV-DO Rx antenna diversity NFC antenna integrated into battery cover Antenna locations: (see FCC OET diagram below) Simultaneous transmission paths: (see FCC OET diagram below) By now, many of our readers have a solid understanding of how to analyze the bullet points listed above. As such, we will hit just some of the highlights. Samsung has finally started adding LTE 10 MHz FDD carrier bandwidth authorization to its Sprint devices. That may be inconsequential during the lifetime of these devices, but it is nice to see, nonetheless, as other OEMs have included 10 MHz FDD capability from the beginning. The Note 3 is reportedly another world phone for Sprint. While the specs for FCC testing include only those bands licensed in the US, the GSM 850/1900 capabilities are actually quad band: GSM 850/900/1800/1900. Likewise, W-CDMA 850/1900 expands to include bands 1 and 8 for W-CDMA 2100+1900/900 outside of the US. The Note 3 also follows the trend of devices released this summer that now support mobile hotspot via 5 GHz Wi-Fi. That can be handy in congested 2.4 GHz Wi-Fi environments, and congested 2.4 GHz Wi-Fi environments are now basically everywhere that people live and work. Devices with SMR 800 MHz capability seem to be coming through the FCC OET now with only conducted power testing, instead of radiated power testing. I suspected previously that a Class II Permissive Change would be required to include ERP testing, but that seems no longer to be the case. Looking back at the Sprint variant Samsung Galaxy S4 authorization docs, they submitted only conducted power specs for CDMA1X 800. Of course, the Galaxy S4 has already been out in the world all summer. I am not sure what to make of this, but Part 90 -- which governs SMR 800 MHz -- may not require ERP testing. So, going forward, we will report conducted power if that is the only data available. But the phrase that dare not be heard among the committed is "single band." One LTE band. That is the elephant in the room in this article. Yes, the "P" variant is band 25 LTE 1900 only. Sources have suggested to S4GRU that tri band LTE handsets headed to Sprint will not support SVLTE. As such, some have hypothesized that Sprint sacrificed tri band capability for SVLTE support in the Note 3 because it is, indeed, a "phablet." Using a "phablet" as a phone against one's ear looks clownish, so using the "phablet" in hand with an earpiece or Bluetooth headset may be expected, leading to more SVLTE usage. However, further examination of FCC OET docs on the multiple Note 3 variants and subsequently announced Sprint version Samsung Galaxy Mega 6.3 tri band "phablet" pokes some holes in that SVLTE theory. The major revelation, which will bring no solace to Sprint subs who had been expecting and are now pining for a Sprint tri band Note 3, though, is that the "R4" version headed to USCC manages to support both SVLTE and quad band LTE: band 4, 5, 12, 25 LTE 2100+1700/850/700/1900. See the antenna locations and simultaneous transmission paths diagrams: The big takeaway is that Samsung has managed to cram in SVLTE and multi band LTE for relatively minor regional CDMA2000 operators. Why not Sprint? Well, the Sprint variant Note 3 is a world phone, while the regional CDMA2000 operator variant is not. And the Sprint version has to support band class 10 CDMA1X 800 and band 25 LTE 1900 -- unlike any others. Does the engineering required for world phone and Sprint boutique band/class capabilities present an obstacle to multi band LTE? Is SVLTE worth the sacrifice? Let the discussion flow... Source: FCC, FCC Thread: http://s4gru.com/index.php?/topic/4318-samsung-galaxy-note-3/