- This topic has 71 replies, 1 voice, and was last updated 5 years, 9 months ago by Bijoy.
13th May 2012 at 19:31 #50888BijoyGuest
As far as I Know,SAIC mobile stations were designed to demodulate a GMSK carrier and suppress GMSK- modulated interference. These mobile stations are typically designed to suppress one dominant interferer. They can be used to suppress the second sub-channel in an AQPSK modulated signal and therefore can be used as receivers for the VAMOS technique.
However, the SAIC methodology does not exploit the fact that when AQPSK modulation is used, the second sub-channel has a constant 90 degree phase shift with respect to the first sub-channel. Moreover, a SAIC receiver will experience the second sub-channel as the dominant interference and it will lose effectiveness suppressing additional external interferers,hence low SCPIR compared to VAMOS level 1 MS.
VAMOS level 1 MS exploit knowledge of the two TSC’s. When an ordinary QPSK demodulator or a proper modification of a QPSK modulator to deal with AQPSK is used, then joint detection is performed. This type of receiver offers optimal performance whenever the signal is corrupted by Gaussian white noise only. But such a receiver is unable to effectively suppress any external interference.Hence higher SCPIR compared to SAIC level 1 MS and so comes the need of SAIClevel 2 MS.
Bijoy13th May 2012 at 19:36 #50889BijoyGuest
Last line correction
Hence higher SCPIR compared to SAIC MS but not so good (specially for -ve SCPIR)and so comes the need of SAIC level 2 MS.14th May 2012 at 05:50 #50890pixGuest
You are describing the difference between VAMOS-I/SAIC and the VAMOS-II. The latter is able to effectively suppress the AQPSK interferer.
VAMOS level 1 MS is just a SAIC MS which is able to use the TSC Set 2. There is no special interference algorithms based on AQPSK usage. Therefore, their “SCPIR thresholds” are the same.
It can be seen in the 3GPP “MUROS” tech spec. The link level simulations show no difference between SAIC and VAMOS-I, as far as I remember. They both are described as “DARP”, with no further difference.
VAMOS-II on the other hand is much more efficient and can work with SCPIR as low as -10dB !
For SAIC/VAMOS- I, the SCPIR must be greater than -4dB.
pix14th May 2012 at 07:22 #50891BijoyGuest
Don’t you think support of additional TSC’s should improve the VAMOS performance of vamos level 1 MS?
Bijoy14th May 2012 at 08:17 #50892pixGuest
Yes you are absolutely right, that’s a good & valid point.
But I wonder by how many dB’s the SCPIR is improved ? max 2dB I would say ?
Also, this software enhancement might actually benefit the paired MS too :
If a VAMOS-I MS is paired with a SAIC MS, then both are equally enjoying the orthogonality of their TSC. In this case, the SAIC MS is as efficient as the VAMOS-I.
But when a SAIC is paired with a SAIC, that’s where there is a slight degradation.
It’s great to discuss with you because you are very educated & have a very sharp mind 😉
Regards15th May 2012 at 05:39 #50893paraHOGuest
Pix / Bijoy
I went away to check what updates for VAMOS before answer contest question and you both have already answered the question.
I know VAMOS multiplexes a pair of users on the two VAMOS subchannels of a single physical radio resource and this allows up to four users on each time slot. This assumes each user is separate. What happens SCPIR in cases where single MS has activated two radio channels?15th May 2012 at 10:31 #50894pixGuest
you snooze, you loose !
“What happens SCPIR in cases where single MS has activated two radio channels?”
Why would a MS have 2 radio channels ?
VAMOS works only for TCH, and therefore each MS needs only one TCH. As if it were in a regular TCH timeslot.
4 MS (a, b, c, d) on one VAMOS TCH timeslot means :
2 MS (a, b) in Half-Rate, paired on the HR-subchannel 0. a is on vamos-subchannel 0, b is on vamos-subchannel 1.
2 MS (c, d) in Half-Rate, paired on the HR-subchannel 1. a is on vamos-subchannel 0, b is on vamos-subchannel 1.
In the physical TCH timeslot, there is two HR subchannels, which are physically independant (as an image : HR0 = even bursts and HR1 = odd bursts).
On HR0 (even bursts), the a and b are tx-ing and rx-ing at the same time.24th May 2012 at 06:53 #50895BijoyGuest
Next Q pls.
Bijoy24th May 2012 at 18:11 #50896pixGuest
your turn to shoot, bijoy. I’m so exhausted from work, I need to have fun !25th May 2012 at 05:44 #50897BijoyGuest
As u wish.Here are a few?
1.Why SF512 is not used in UL?
2.If a UE is assigned consecutive TTI’s(HSPDCH)in HSDPA,it uses the same HS-SCCH.What if this HS-SCCH is erroneous?
Bijoy4th March 2013 at 16:44 #50898RajGuest
Need some information regarding SACCH position for VAMOS :
Shifted SACCH is only for DTX cases?
If above is true then what will be SACCH mapping for User1 if both the
calls are full rate (VAMOS)18th August 2015 at 11:21 #50899BijoyGuest
After 2 long years,I’m back and found my questions still unanswered.
Anyways,here are the answers.
1.SF-512 is not used in UL because each Ue has its own code tree and we have no channelization code shortage,there is no point in having so many SF’s.
2. If an error is detected in HS-SCCH, the Ue sends back over the uplink channel HS-DPCCH a NACK indicating an error. If no error is detected, the Ue checks whether the user data on the allocated HS-PDSCH(s) are erroneous. If they are not, an ACK is transmitted over the HS-DPCCH, else a NACK is transmitted.