Content for  TR 38.835  Word version:  18.0.1

This clause captures the capacity performance evaluation results with and without intra- and inter frequency RRM measurement restrictions. Scheduling restrictions apply for the UEs during time-intervals where it is performing intra-frequency RRM measurements at FR2, or gap assisted inter-frequency RRM measurements. Such scheduling restrictions does not allow to transmit PUCCH/PUSCH/SRS or receive PDCCH/PDSCH/CSI-RS. When XR frame arrival collides with RRM measurements, the XR frame will be delayed for the duration of SMTC window or measurement gap (MG). The performance with SMTC/MG scheduling restrictions (scheme 7.2 and 7.4 in Table B.1.7-1 and Table B.1.7-2) has been compared against no SMTC/MG scheduling restrictions as well as with adaptation of MG period. Particularly, the following schemes have been evaluated:

• Scheme 7.1: Without SMTC scheduling restrictions. In this scheme, no scheduling restrictions for FR2 intra-frequency measurements were assumed. This can be realized via gNB-2-UE signalling to configure the UE with time-mask where it shall always prioritize PDCCH/PDSCH decoding and/or PUSCH transmission in line with XR traffic, even if colliding with SMTC windows. Alternatively, the UE-to-gNB signalling to make the gNB scheduler aware of when s-MeasureConfig induced scheduling restrictions apply could be applied.
• Scheme 7.2: With SMTC scheduling restrictions. In this scheme, scheduling restrictions every 20 ms time period for an SMTC window of 5 ms for FR2 intra-frequency measurements were assumed. In case the SMTC windows with scheduling restrictions collide with time periods where the gNB would need to schedule the XR transmission, XR frame transmission will be delayed.
• Scheme 7.3: No scheduling restrictions during measurement gap. In this scheme, no scheduling restrictions during measurement gap for inter-frequency measurements were assumed.
• Scheme 7.4: With scheduling restrictions during measurement gap. In this scheme, scheduling restrictions for the duration of measurement gap length of 6 ms are applied. Measurement gaps (MG) are configured to allow UE to do inter-frequency neighbour cell measurement and the corresponding RF tuning for RRM purposes (e.g. mobility, load balancing, CA set-up). In case the MG with scheduling restrictions collide with time periods where the gNB would need to schedule the XR transmission, XR frame transmission will be delayed.
• Scheme 7.5.1: Adaptation of MG period. The MG period is adapted to be 2 times when PCell RSRP is 2dB better than the best neighbour cell and 4 times when PCell RSRP is 4dB better than the best neighbour cell. The UE speed is 30km/h.
• Scheme 7.5.2: Adaptation of MG period. The MG period is adapted to be 4 times when PCell RSRP is 2dB better than the best neighbour cell and 8 times when PCell RSRP is 4dB better than the best neighbour cell. The UE speed is 30km/h.

Based on the evaluation results in Table B.1.7-1 and Table B.1.7-2 the following observations can be made:

• For FR2, InH, DL, with 100MHz bandwidth for VR/AR single-stream traffic model, 30Mbps, 10ms PDB, 60 FPS, with SU-MIMO and 2TxRU, it is observed from Source [Nokia] that the capacity is increased from 3.1 UEs per cell with SMTC scheduling restrictions for FR2 intra-frequency measurements (5 ms SMTC window every 20 ms) to 7 UEs per cell without SMTC scheduling restrictions for FR2 intra-frequency measurements (capacity gain is 126%). For 15ms PDB the capacity gain is 44%.
• For FR1, DU, DL, with 100MHz bandwidth for VR/AR single-stream traffic model, 30Mbps, 10ms PDB, 60 FPS, with SU-MIMO and 64TxRU, it is observed from Source [MediaTek] that the capacity is increased from 0 UEs per cell with MG scheduling restrictions for inter-frequency measurements for all UEs (6 ms MG length every 40 ms) to 10.6 UEs per cell without MG scheduling restrictions for inter-frequency measurements. Similar trend is observed for MG configuration 80 ms.
• For FR1, DU, DL, with 100MHz bandwidth for VR/AR single-stream traffic model, 30Mbps, 10ms PDB, 60 FPS, with SU-MIMO and 64TxRU, it is observed from Source [MediaTek] that the capacity is increased from 2 UEs per cell with MG scheduling restrictions for inter-frequency measurements for all 20% UEs (6 ms MG length every 40 ms) to 10.6 UEs per cell without MG scheduling restrictions for inter-frequency measurements (capacity gain is 400%). Similar trend is observed for MG configuration 80 ms.
• For FR1, DU, DL, with 100MHz bandwidth for VR/AR single-stream traffic model, 30Mbps, 10ms PDB, 60 FPS, with SU-MIMO and 64TxRU, it is observed from Source [MediaTek] that the capacity is increased from 0 UEs per cell with MG scheduling restrictions for inter-frequency measurements for all UEs (6 ms MG length every 40 ms) to 0.5 UEs per cell with adaptation of MG period for inter-frequency measurements, where MG period is adapted to be 2 times when PCell RSRP is 2dB better than the best neighbour cell and 4 times when PCell RSRP is 4dB better than the best neighbour cell. Similar trend is observed for MG configuration 80 ms.
• For FR1, DU, DL, with 100MHz bandwidth for VR/AR single-stream traffic model, 30Mbps, 10ms PDB, 60 FPS, with SU-MIMO and 64TxRU, it is observed from Source [MediaTek] that the capacity is increased from 2 UEs per cell with MG scheduling restrictions for inter-frequency measurements for all 20% UEs (6 ms MG length every 40 ms) to 7.2 UEs per cell with adaptation of MG period restrictions for inter-frequency measurements, where MG period is adapted to be 2 times when PCell RSRP is 2dB better than the best neighbour cell and 4 times when PCell RSRP is 4dB better than the best neighbour cell (capacity gain is 250%). Similar trend is observed for MG configuration 80 ms.
• For FR1, DU, DL, with 100MHz bandwidth for VR/AR single-stream traffic model, 30Mbps, 10ms PDB, 60 FPS, with SU-MIMO and 64TxRU, it is observed from Source [MediaTek] that the capacity is increased from 0 UEs per cell with MG scheduling restrictions for inter-frequency measurements for all UEs (6 ms MG length every 40 ms) to 0.8 UEs per cell with adaptation of MG period for inter-frequency measurements, where MG period is adapted to be 2 times when PCell RSRP is 2dB better than the best neighbour cell and 4 times when PCell RSRP is 4dB better than the best neighbour cell. Similar trend is observed for MG configuration 80 ms.
• For FR1, DU, DL, with 100MHz bandwidth for VR/AR single-stream traffic model, 30Mbps, 10ms PDB, 60 FPS, with SU-MIMO and 64TxRU, it is observed from Source [MediaTek] that the capacity is increased from 2 UEs per cell with MG scheduling restrictions for inter-frequency measurements for all 20% UEs (6 ms MG length every 40 ms) to 8.7 UEs per cell with adaptation of MG period restrictions for inter-frequency measurements, where MG period is adapted to be 2 times when PCell RSRP is 2dB better than the best neighbour cell and 4 times when PCell RSRP is 4dB better than the best neighbour cell (capacity gain is 335%). Similar trend is observed for MG configuration 80 ms.