HSPA DL throughput

[mardaphoca]

Hi, I am a Electrical Engineering student and I have a trouble understanding some issues about the HSDPA technology...

According to my understanding, the DL throughput depends on the CQI value that the UE measures. But the medium is shared, so if the node B transmits with a certain Bit Rate, that is shared among the users.

So, I'm using a Sierra Wireless HSPA modem, and through the AT comands I can obtain the both the CQI and the EC/IO.

Is there a way to estimate the throughput with these two metrics?


I thank you

best regards

[mon3em]

Hi

there is many parameters & functionality defining throughput.

Radio condition(CQI), UE hardware catogary , # HSDPA Codes per cell & # HSDPA users per cells.

THROUGHPUT for use is defined by TBS (transport block size) which is transmitted to user according to above parameters & mainly according to CQI.

CQI it self is a direct image for Ec/No .... please to check 3Gpp documents for details.

mon3em

[mardaphoca]

Hi,

Thanks for your quick answer.

I was trying to figure out if it was possible to predict how many users are in cell through the Ec/Io. Or maybe other parameters.

With AT commands I can define the UE category and get the CQI.

The CQI would reflect the Nr of codes and transport block size.

Maybe this thinking is not correct?

Regards

[mon3em]

CQI is reflecting Ec/No >>>> signal quality

refer for 3gpp_doc for this introduction

mon3em

[mardaphoca]

Hi,

I've been through the 3gpp docs and it seems that CQI derivation is proprietary in the UE and is not standardised in the specs.
There are aproximations that i assume that can be used in link budget operations:

CQI = 0 if SINR<-3.96 dB
CQI = SINR/1.02 +4.81 if -3.96<SINR<26.04 dB
CQI = 30 if SINR >26.04 dB

So, is SINR measurable? or is it an aproximation in link budget calculations?
Is SINR related to Ec/Io or Ec/No ?

Thank you

[qaqa]

Hey Guys

don't forget the scheduler in the NodeB for determining the HSDPA Throughput.

[mardaphoca]

So do you think that calculation for cqi is accurate?

thanks

[brianm]

That CQI calculation is incorrect as it depends on a lot of factors listed above including type of UEs booked on the cell/site

[Ummi]

above is correct practically used. Remember if we'r on broadband environment must adopt RSVP then make E2E QoS, so you will have more troughput if you pay more money. Using this method depend on profile of user in HLR with negotiation to SAPC/SASN