Jump to content


Photo

LTE Fragmentation(any "hopefully" a solution)


  • Please log in to reply
No replies to this topic

#1 xcharles718

xcharles718

    Member Level: 3G EVDO

  • S4GRU Premier Sponsor
  • PipPip
  • 408 posts
  • Phones/Devices:Nexus 5(T-Mobile)
  • LocationBrooklyn, NY

Posted 08 April 2012 - 09:53 AM

There are currently a total of 43 different LTE band classes in the world. Very likely, more bands classes will come into fruition in the future. However, many of these band classes are somewhat redundant. Here's one such example.

Band Class 2 covers the U.S. PCS Spectrum bands(at least most of them). Band 2 covers frequencies 1930-1990MHz on downlink and 1850-1910MHz on uplink. This is the FCC’s PCS A-F blocks.
Around the time Sprint announced the EVO 4G(April 2010), Sprint requested a band extension from the 3GPP for Band 2. They asked for this so that it can also cover the G block(has you might know, is what Sprint’s LTE network will operate on at launch). The extension was eventually approved, but it ended up creating a new Band #(in this case, Band 25). This is not a true extension of Band 2, but pretty much covered what Sprint requested.
Band Class 25 covers the frequencies 1930-1995MHz for downlink and 1850-1915MHz for uplink.
Posted Image

As you can probably forumate from above, Band 25= Band 2 + G Block.

I can understand that creating a new band class does help prevent confusion on if a device supports a certain frequency, or that manufacturers can just create new devices using the newer bands instead of the older ones that covers less possible networks.
The thing is that LTE is young, new, and still adapting to the world's different frequency blocks across the globe. Band classes will continue to grow for at least the next few years. Some carriers may opt to buy new spectrum and deploy LTE there. Other(like Sprint & T-Mobile USA) would refarm and use spectrum they already own.
This makes building a multi-carrier device a bit of a pain at this time since most devices are incompatible with other networks, even one within a close MHz range. One example is the 700MHz spectrum in the U.S.(which will be discussed in another article)

One solution I see for this is the true extension of band classes to include new bands. For this, I would opt for all band classes to be given “draft” status, along with a revision label be given to every time a new request for an extension of a band class is approved instead of a new band class number. For example, Band 2 & 25.
With the addition of a spectrum block to band 2, instead of creating Band 25(or 26 or etc.), use a revision marker for the same band class.
So if in 2010, PCS bands(without G block) for LTE was called Band 2, PCS w/G Block would become the new Band 2. A revision number would be associated with each small change to the band along with “draft” status to the class number, but all future devices w/ this band number will include the updated frequency range. After a few years, the band class would be finalized and no additional changes can or will be made. This will prevent redundant band classes, intentional use of older bands(anticompetitive practices), and excessive phone revisions of the same phone(no more “one size fits one” and more “one size fits many”).

I’ll end this write-up with the growing possibility of LTE roaming in the near future. South Korea carrier’s SK Telecom & LG Telecom(LG U+) currently operate their LTE networks at 800MHz(Band 5). Sprint has just received approval for Band Class 26(an extended Band 5 to support ESMR).
Posted Image

If both the Korean carriers decided to sell new devices with Band 26 support along with Sprint, could this be the beginning of internationality compatible LTE devices?
Posted ImagePosted Image

Sources: Niviuk(http://niviuk.free.fr/lte_band.php), 3GPP(Band 25[http://www.3gpp.org/.../RP-110804.zip ]
Band 26[http://www.3gpp.org/.../RP-120305.zip] ), Engadget, LG U+ (Korean[http://www.uplus.co.kr/])
This line you're reading right now is intentionally left blank...but it's not really blank, is it?




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users