This concept is in a research stage right now. People are working on making
it technically feasible and economically viable by trying different things
with the belief that the technology has long-term potential. Also, as you
pointed out it would likely not apply to all areas (like rural areas where
rolling out fiber is too expensive today) but to places where either fiber
is already available (like most of China and many APAC countries) or in
dense urban areas where consolidation on BS processing makes short runs of
fiber pay for itself. There is a whole series of white papers and articles
on this technology. Sharing a few links
*http://labs.chinamobile.com/cran/* <http://labs.chinamobile.com/cran/> which
has the link to download their white paper (
http://www.alcatel-lucent.com/features/light_radio/index.html (this has
links to ALUs white papers on the topics)
On Sun, Apr 15, 2012 at 3:34 PM, Akibsayyed <akibsayyed@...> wrote:
> Hello partul
> Concept you are talking is completely new for list people. could you
> please foucus more on this.
> Laying fibre network is not feasible in rural area but in urban area.
> Still i doubt on part processing baseband stack on cloud part.
> Please focus more on this.
> Akib Sayyed
> Matrix Shell
> Mob +91-966-514-2243
> Sent from Android
> Parul Gupta <parul8ue@...> wrote:
> Thanks for your response David.
> Let me give you some background on what I am trying to do. We are trying
> to simulated a "Cloud RAN scenario" which has lately been seeing a lot of
> interest in operators and equipment providers (China mobile, ALU, NSN, IBM
> are all talking about this new architecture with different names). The
> basic idea is to run all the Baseband processing of a wireless stack (PHY
> MAC and upwards) in a centralized "cloud"-like platform and connect this to
> remote radio heads via fiber. Besides OpenBTS GSM stack, we have some other
> technology stacks (wimax, LTE) and I want to simulate a scenario how many
> such virtual BS can fit on to a server from a compute and memory
> requirements standpoint. The next step would be to test with I/O.
> This is why I need the SWLOOPBACK mode - to test only the compute part of
> multiple OpenBTS instances running with multiple instances of other
> technology BS. Now if the SWLOOPBACK feature is outdated, would there be
> quick alternate to do this? If not, and if I were to start coding this
> feature, would it be straight-forward or involve massive rewriting of the
> code? Any pointers would be very helpful.
> Thanks much in advance!
> On Fri, Apr 13, 2012 at 10:04 PM, David Burgess <david@...:
>> Parul -
>> I'm surprised you got loopback to work in 2.6. It was originally put in
>> there to support unit testing of some of the channel components. We have
>> not supported it for a while and were planning to remove it.
>> -- David
>> On Apr 13, 2012, at 3:29 AM, Parul Gupta wrote:
>> > Dear all,
>> > I am new to the list and also somewhat new to OpenBTS 2.8, but excited
>> > Does anyone know if the SWLOOPBACK functionality is tested and
>> > supported in 2.8? It is supposed to be a feature for loopback
>> > testing/debugging without any hardware and I have used it in OpenBTS
>> > 2.6 without any problems. But for v2.8, when I compiled with this flag
>> > defined in Transceiver52M/USRPDevice.h and
>> > TransceiverRAD1/RAD1Device.h, I got some compilation errors (in
>> > RAD1Device.cpp, RAD1ping.cpp). I was able to fix these easily, but it
>> > seems to question whether SWLOOPBACK is supported in this release.
>> > Also, after fixing these errors, when I run OpenBTS, it seems to hang
>> > after printing "Starting the system..."
>> > I also couldn't find anything on it in the User Manual
>> > (SoftwareP2.8Manual.pdf) that came with the code.
>> > Thanks very much in advance
>> > Parul
>> David A. Burgess, CEO
>> Range Networks, Inc.
>> San Francisco
>> cell +1 707 208 2622
>> Cellular networks made simple and affordable.
> Knowledge sharing and collaboration for social welfare -
Knowledge sharing and collaboration for social welfare - http://ngopost.org/