SPS IN LTE in UL & DL , along with DS

Question

How does SPS & DS Scheduling can be done at a time ?? Does UE always decode pdcch with sps-crnti once SPS is configured ?? Or UE can listen to pdcch with both CRNT & SPS-CRNTI ??

Kindly clarify my doubts..

Answer 1

As we know, an LTE UE can run both the applications voice as well as data simultaneously. Use of SPS is beneficial when data comes with a fixed rate. Usually voice application data comes with the interval of 20ms. SPS and Dynamic scheduling can work together for an UE.
In downlink data flow, SPS C-RNTI and normal C-RNTI will be used for SPS and dynamic scheduling respectively.
As SPS can be configured for UL or DL or both.

From MAC specification, its seems either UE will be able to use SPS C-RNTI or UE's C-RNTI per TTI. But for this I am not sure.
I am waiting for other professionals to respond on this topic.

Comments

Thank you Vimal for responding.
yeah , i do have the same query , if UE listen to PDCCH only using SPS-CRNTI , then how DS & SPS can occur at a time .

Answer 2

Excerpt from this link http://blog.3g4g.co.uk/2011/02/volte-semi-persistent-scheduling-sps.html
says, "dynamic scheduling of best effort data can occur on top of SPS, but the SPS allocations would take precedence over any scheduling conflicts"

The UE finds its PDCCH by monitoring a set of PDCCH candidates in every subframe. This is known as blind decoding. The UE demasks each control candidate's CRC using its RNTI. If no CRC error is detected, the UE considers it as a successful decoding attempt and reads the control information within the successful candidate.
With the possibilities of different RNTIs, PDCCH candidates, DCI and PDCCH formats, a significant number of attempts may be required to successfully decode the PDCCH. To overcome this complexity the UE first tries to blindly decode the first CCE in the control channel candidate set of a subframe. If the blind decoding fails, the UE tries to blindly decode the first 2, 4 then 8 CCEs sequentially, where the starting location is fixed for common search case. (Courtesy http://www.mathworks.in/help/lte/examples/pdcch-blind-search-and-dci-decoding.html)

In summary, UE needs to search for :
[1] (Temporary) C-RNTI in UE specific search space
[2] SPS C-RNTI in UE specific search space if SPS is configured

[3] SI-RNTI/P-RNTI in Common Search space (not talking about RA-RNTI as it only happens during Random access procedure)
[4] TPC-PUSCH-RNTI in Common search space if configured
[5] TPC-PUCCH-RNTI in Common search space if configured

UE specific search space has 6(aggregation1) + 6(aggregation2) + 2(aggregation4) + 2(aggregation8) = 16 candidates.
Two possible RNTI in UE search space shall make searching (worst case)= 16*2=32

Common Search space has 4(aggregation4) + 2(aggregation8) = 6 candidates.
Three possible RNTI in common search space shall make searching (worst case)= 6*3=18

SPS and C-RNTI can occur at the same time, and with the fact that SPS-RNTI based decoding is only used for activation/reactivation of SPS bearers and it uses C-RNTI other times, 2 RNTI search spaces are not that bad.