3GPP TSG RAN WG1 #70 /

Transcription

1 3GPP TSG RAN WG1 #70 /

2 3GPP TSG RAN1 5 RAN1 Rel-11 RAN1 CA EnhancementCoMP Enhanced Downlink Control Channel Small Cell Enhancement LTE-A Rel-11 Rel RAN1 contributions Simultaneous HARQ-ACK and periodic CSI on PUCCH Format 3: Joint CodingSprarate Coding 2. Interband CA UL/DL TDD configurationitri 1 contribution RAN1 (inter-band carrier aggregation)(soft buffer partitioning) discussion [70-06] Rel-11 Rel-11 LTE TDD for DL-UL Interference Management and Traffic 2

3 Adaptation Rel-11 LTE-A Carrier Aggregation EnhancementsCoMP Further Enhanced Non-CA-based ICIC for LTEe-PDCCH CoMP Rel-11 RAN#56 Rel-12 contribution way forward contribution treat Rel-11 Rel-12 and Beyond Workshop Rel-12 small cell RAN Plenary RAN1 small cell enhancement Rel-12 small cell enhancement Rel-12 Samsung Alcatel lucent [7.2.3 Support of different TDD UL-DL configurations on different bands] RAN#51 (TDD)- LTE Rel-11 (working item) (TDD)- (full-duplex)(half-duplex) (inter-band carrier aggregation) (self-scheduling) (cross-carrier scheduling)(ran1 #70) (cross-carrier scheduling) (HARQ timing) [7.6 Enhanced downlink control channel(s)] RAN1#66bis LTE Rel-11 (new carrier type) (CoMP) (DL 3

4 MIMO) RAN1 (enhanced downlink control channel) (capacity) (inter cell interference coordination; ICIC) (spatial reuse) (beamforming) (diversity) MBSFN subframes (RAN1 #70) (reference signal)(search space) 4

5 GPP TSG RAN1#70 Meeting LTE-A Rel-11 RAN1 - (cross-carrier scheduling)(self-scheduling) - PDSCH (HARQ timing) - PUSCH (HARQ timing) - (resource element) 5

6 - PRB pairs - (search space)(aggregation level) - (reference signal)dm-rs port epdcch transmission - transport channel bits Enhanced Downlink Control Channel Small Cell Enhancement 13 Aug.- 17 Aug InterContinental Qingdao (Qingdao, China) RAN1 (13 Aug.- 17 Aug. 2012): 6

7 Tentative Schedule RAN1#70 8 am 9 am Coffee 10:30-11:00 Lunch Coffee 16 :00-16: 30 Monday Tuesday Wednesday Thursday Friday Enh DL CCHs (7.6.3) 1, 2, 3, 4 LSs 5 LTE Maintenance 7.1 Quasi Co- Located APs 7.7 Net Pos 7.4 DL CoMP CSI DL CoMP Ctrl Sig PUCCH TxD CA Enh TDD CA Enh HSPA Maint. 6.1,6.2 HSUPA MIMO 64QAM (6.6.4) DL CoMP RS FeICIC Enh DL CCHs UL CoMP Enh. Cell_FACH HSDPA Multi- Point HSPA 6.8 HSDPA 4x4 Enh DL CCHs ! CA Enh TDD (contd) CoMP FB Modes (6.3.5) CoMP Adhoc 7.5.1, On PUCCH CSI-RSRPRA for epdcch 8.30 HSPA 6.7 TDD H S P A C R re v ie w s 6.3.5, 6.4.1, 6.5.3, Net Pos 7.4 CA Enh Mult. TA QCL APs ( contd) 7.7 Revisiting AIs as neces sary Ocean Room Stage 7&8 Stage 10&11 7

8 1. Hauwei 250 ordinary RAN1#70 RAN2#79RAN3#77RAN4#64 RAN5#56 (Working GroupWG)CHTTL 1 RAN1#70 / CHTTLITRIHTCIII MediaTekASUSAcer 7 UTRA (WCDMA/HSPA) open ((Work ItemWI)) 5 Agenda Item 6.3: Four Branch MIMO Transmission for HSDPA Agenda Item 6.4: Further Enhancements for Cell_FACH Agenda Item 6.5: HSDPA Multiflow Data Transmission Agenda Item 6.6: MIMO with 64QAM for HSUPA Agenda Item 6.7:Study on HSPA feedback and signalling efficiency enhancements for LCR TDD E-UTRA (LTE/LTE-A) open () 6 Agenda Item 7.2: LTE Carrier Aggregation Enhancements Agenda Item 7.3: Further Enhanced Non-CA-based ICIC for LTE Agenda Item 7.4: Network-Based Positioning Support for LTE Agenda Item 7.5: Coordinated MultiPoint operation Agenda Item 7.6: Enhanced Downlink Control Channel(s) Agenda Item 7.7: Quasi-co-located Antenna Ports 2. Support of different TDD UL-DL configurations on different bands RAN1 (scheduling)(self-scheduling) (cross-carrier scheduling)(pcell) (UL-DL configuration) 8

9 (PCell)(UL-DL configuration) RAN1 #68 (PCell) (PCell) TDD (UL-DL configuration) HARQ-ACK timing of PCell PDSCH, the scheduling timing of PCell PUSCH, the HARQ timing of PCell PUSCH should follow the PCell timing. -- RAN1 #68 PCell timing is the same as Rel-8/9/10. (RAN1 #70) (cross-carrier scheduling)(harq timing) (SCell)(PDSCH) (HARQ timing) (PDSCH)(HARQ timing) (PCell) (UL-DL configuration)(cross-carrier scheduling) (PCell)(UL-DL configuration) (self-scheduling) Table: The reference PDSCH HARQ/scheduling timing on SCell PDSCH HARQ timing Pcell SIB-1 UL-DL Configuration on SCell follows TDD UL-DL configuration # A A A A A A 1 B A C A A B Scell SIB-1 2 B B C C A B UL-DL 3 B C C A A B Configuration 4 B B C B A B 5 B B B B B B 6 B Notes: The number in the grid is the reference UL-DL Case A Case B Case C 9

10 configuration which SCell PDSCH HARQ timing follows. In case of cross-carrier scheduling: PDSCH HARQ timing on SCell follows P-Cell timing for PDSCH, regardless of the number of aggregated CCs In case of self-scheduling (as shown in the following table): For both full-duplex and half-duplex case: Case A: The set of SCell(s) downlink subframe is a subset of PCell PDSCH HARQ timing on SCell follows PCells TDD UL-DL configuration Case B: The set of SCell(s) downlink subframe is a superset of PCell PDSCH HARQ timing on SCell follows SCells TDD UL-DL configuration Case C: The set of SCell(s) downlink subframe is neither a subset nor a superset of PCell PDSCH HARQ timing on SCell follows a reference TDD UL-DL configuration Table: The reference PDSCH HARQ/scheduling timing on SCell PDSCH HARQ timing Pcell SIB-1 UL-DL Configuration on SCell follows TDD UL-DL configuration # Scell SIB UL-DL Configuration Notes: The number in the grid is the reference UL-DL configuration which SCell PDSCH HARQ timing follows. Case A Case B Case C (PUSCH)(HARQ timing) 10

11 (scheduling timing) (PCell)(UL-DL configuration) (cross-carrier scheduling)(pcell)(ul-dl configuration)(self-scheduling) Table: The reference PUSCH HARQ/scheduling timing on SCell HARQ/scheduling Scheduling cell SIB-1 UL-DL Configuration timing of PUSCH on Scheduled Cell follows TDD UL-DL Configuration # B B B B B D 1 D B C B B D Scheduled cell 2 D A C C B D SIB-1 UL-DL 3 D C C B B D Configuration 4 D A C A B D 5 D A A A A D 6 D B B B B B Notes: The number in the grid is the reference UL-DL configuration which SCell PDSCH HARQ timing follows. Case A Case B Case C Case D In case of cross-carrier scheduling (as shown in the following table): Case A: The set of SCell(s) uplink subframe is a subset of PCell and if the PUSCH RTT of the scheduling cell SIB1 configuration is 10ms PUSCH HARQ/scheduling timing on SCell follows PCell SIB1 configuration -- RAN1 #68 Case B: The set of SCell(s) downlink subframe is a superset of PCell and if the PUSCH RTT of the scheduling cell SIB1 configuration is 10ms PUSCH HARQ/scheduling timing on SCell: FFS Case C: The set of SCell(s) downlink subframe is neither a subset nor a superset of PCell PUSCH HARQ/scheduling timing on SCell: FFS Case D: if the PUSCH RTT of the scheduling cell SIB1 configuration is not 10ms 11

12 PUSCH HARQ/scheduling timing on SCell: FFS In case of self-scheduling: For the full-duplex case: PUSCH HARQ/scheduling timing on SCell follows SCell SIB1 configuration -- RAN1 #68 For the half-duplex case: PUSCH HARQ/scheduling timing on SCell follows SCell SIB1 configuration -- RAN1 #68bis 3. Enhanced downlink control channel RAN1#66bis LTE Rel-11 (new carrier type)(comp)(dl MIMO) RAN1 (enhanced downlink control channel) (capacity) (inter cell interference coordination; ICIC) (spatial reuse) (beamforming) (diversity) MBSFN subframes RAN1 (reference signal) (search space) (resource element) (epdcch) (resource element, RE) epdcch is not transmitted in PRB pairs that contain PBCH or PSS/SSS (RAN1 #70 Agreement) QPSK is the only modulation scheme for epdcch (RAN1 #70 Agreement) The group of REs defined in spatial diversity transmission is 1 RE (RAN1 #70 Agreement) When distributed transmission is used, spatial diversity is used and each RE in a given 12

13 PRB pair belonging to a given DCI is associated by specification with one of two APs alternately following the ereg mapping (FFS which two APs) (RAN1 #70 Agreement) ereg definition: For distributed transmission: A PRB pair is divided into 16 eregs in both normal and special subframes and for normal and extended CP regardless of the presence of other signals. 16eREGs has #0 to #15 indices For localized transmission: FFS The specification supports the case that an ecce is formed by N eregs in distributed and localized: N= 4 in following cases. (This corresponds to 4 ecces per PRB pair in localized transmission.) In normal subframe (normal CP) or special subframe configs 3,4,8 (normal CP) N= 8 in following cases. (This corresponds to 2 ecces per PRB pair in localized transmission.) Special subframe configs 1,2,6,7,9 (normal CP) Normal subframe (extended CP) and special subframe configs 1,2,3,5,6 (extended CP) Aggregation levels supported for epdcch are: In normal subframes (normal CP) or special subframe configs 3,4,8 (normal CP), and the available REs in a PRB pair is less than Xthresh, For localised: 2, 4, 8, working assumption 16 subject to feasible search space design For distributed: 2, 4, 8, 16, working assumption 32 subject to feasible search space design Working assumption that Xthresh = 104 In all other cases: For localised: 1, 2, 4, working assumption 8 subject to feasible search space design For distributed: 1, 2, 4, 8, working assumption 16 subject to feasible search space design (search space)(aggregation level) (common search space, CSS)(UE-specific search space, USS)(CSS) 13

14 (USS) (downlink assignment) (uplink grant) (CSS)(USS) Search spaces, PDCCH monitoring, and the number of blind decodes for epdcch Total number of epdcch USS blind decodes per CC is 32 or 48 depending on configuration of UL MIMO The UE is not expected to receive epdcch in a special subframe with special subframe configuration 0 or 5 in normal CP, or special subframe configuration 0, 4, or 7 in extended CP. The ereg to RE mapping is fixed in specifications given the frame structure type, subframe configuration and CP length Special subframes with the same DMRS positions have the same ereg to RE mapping The ereg to RE mapping does not depend on the PRB pair#, subframe#, legacy control region size, DwPTS length or presence of other signals such as CRS,CSI-RS,PRS, ereg indices are sequentially mapped to the REs without REs for DMRS (24 for normal CP and 12 for extended CP) in a frequency first and then time manner, within each PRB pair It is FFS whether to support cyclic shift of the assigned ereg indices in each OFDM symbol or further rearrangement in the OFDM symbols carrying DMRS. Note that Xthresh = 104 was derived to keep the worst case coding rate close to 0.8. Note that epdcch is not mapped to GP or UpPTS. When a UE detects its DL assignment defining a PDSCH allocation which overlaps with the PRB pair(s) containing its DL assignment, the UE shall assume that the PDSCH scheduled by its DL assignment is rate-matched around the PRB pair(s) containing its DL assignment. In addition, the UE shall assume that the PDSCH scheduled by its DL assignment is not mapped to that PRB pair(s) containing its DL assignment on any layer. In subframes not configured for monitoring epdcch, UE monitors CSS and USS on PDCCH according to Rel-10 behaviour In subframes where UE monitors epdcch USS on a given carrier: it does not monitor PDCCH USS on the same carrier it can at least be configured to monitor either localised, or distributed epdcch candidates in a given subframe it also monitors CSS on PDCCH 14

15 working assumption that the UE can be configured to monitor both localised and distributed epdcch candidates in a given subframe If both are configured, the total number of USS blind decodes on the carrier is not increased The subframes where UE monitors epdcch USS are defined by at least rules in the specs not special subframe configurations 0 and 5 for normal CP, 0 and 4 for extended CP working assumption that configuration by higher layer signalling can also be provided (details of the higher layer signalling are FFS) epdcch and PRB pairs An epdcch set is defined as a group of N PRB pairs Working assumption: N = {1 for localised (FFS), 2, 4, 8, 16 for distributed (FFS), } A distributed epdcch is transmitted using the N PRB pairs in an epdcch set A localized epdcch shall be transmitted within an epdcch set FFS whether a localised epdcch can be transmitted across more than one PRB pair K 1 epdcch sets are configured in a UE specific manner Maximum number for K is selected later among 2, 3, 4, and 6 The K sets do not have to all have the same value of N The total number of blind decoding attempts is independent from K The total blind decoding attempts for a UE should be split into configured K epdcch sets Each epdcch set is configured for either localized epdcch or distributed epdcch The K sets consist of KL sets for localized epdcch and KD sets for distributed epdcch (where KL or KD can be equal to 0), and not all combinations of KL and KD are necessarily supported for each possible value of K Details FFS PRB pairs of epdcch sets with different logical epdcch set indices can be fully overlapped, partially overlapped, or non-overlapping. Note that excessive configurations should be avoided. Note that the details of the second subbullet are dependent on the conclusions on ereg definition. Note that it may be possible to forbid certain combinations of N and K Note that the used values of N and K may depend on the system bandwidth. 15

16 Remaining Details of RE Mapping for epdcch epdcch PRB pair ereg ereg mapping distributed transmission ecce ereg localized transmission PRB pair ecce epdcch PRB pair Huawei HiSilcon PRB pair 8 ereg ereg mapping block mapping normal subframe MBSFN subframe PRB pair overhead 2 4 ecceecce mapping checkerboard structure Intel Corporation ereg 9 RE ereg mapping cell common NTT DOCOMO PRB pair 16 ereg set PRB pair 20MHz CATT ecce PRB pair PRB pair overhead 2 4 ecce Qualcomm Inc. normal subframe PRB pair 2 4 ecce ecce PRB pair interleaver inter-cell interference epdcch resource distributed epdcch localized epdcch PRB pair share ecce set Samsung distributed transmission ereg 4 RE ecce 36 RE ereg mapping REG mapping Samsung Rel-11 support distributed epdcch localized epdcch multiplexing Renesas Mobile Europe Ltd PRB pair 16 ereg normal subframe special subframe ereg interleaved mapping RE Nokia Nokia Siemens Network ecce ereg 16

17 local transmission ereg multiplexing epdcch operation modes (localized epdcch distributed epdcch)nokia Nokia Siemens Network ecce ereg mapping cyclic mapping maximum distance interleaving Alcatel-Lucent Alcatel-Lucent Shanghai Bell normal subframe PRB pair 4 ecce special subframe PRB pair 2 4 ecce ecce mapping distributed checkerboard pattern Research In Motion UK Limited normal CP PRB pair 4 ecce extended CP PRB pair 2 ecce Sharp PRB pair 16 ereg ecce ereg localized transmission distributed transmission Panasonic PRB pair 16 ereg normal subframe special subframe ecce normal subframe 4 ereg NEC PRB pair normal subframe 3 4 ecce PRB pair special subframe 2 3 ecce; ecce N ereg N subframe 9 ZTE ereg mapping PRB mapping LG Electronics ecce 2 4 ereg interference randomization cellereg index PRB pair Ericsson ST Ericsson normal subframe special subframe PRB pair 16 ereg ecce 4 ereg Agreements: - QPSK is the only modulation scheme for epdcch - EPDCCH is not transmitted in PRB pairs that contain PBCH or PSS/SSS 17

18 - The group of REs defined in spatial diversity transmission is 1 RE - ereg definition: A PRB pair is divided into 16 eregs in both normal and special subframes and for normal and extended CP regardless of the presence of other signals 16eREGs has #0 to #15 indices - The specification supports the case that an ecce is formed by N eregs in distributed and localized N= 4 in following cases. (This corresponds to 4 ecces per PRB pair in localized transmission.) In normal subframe (normal CP) or special subframe configs 3,4,8 (normal CP) N=8 in following cases. (This corresponds to 2 ecces per PRB pair in localized transmission) Special subframe configs 1,2,6,7,9 (normal CP) Normal subframe (extended CP) and special subframe configs 1,2,3,5,6 (extended CP) Aggregation levels supported for EPDCCH are: In normal subframes (normal CP) or special subframe configs 3,4,8 (normal CP), and the available REs in a PRB pair is less than X thresh, For localised: 2, 4, 8, working assumption 16 subject to feasible search space design For distributed: 2, 4, 8, 16, working assumption 32 subject to feasible search space design In all other cases: For localised: 1, 2, 4, working assumption 8 subject to feasible search space design For distributed: 1, 2, 4, 8, working assumption 16 subject to feasible search space design Working assumption that Xthresh = 104 Total number of epdcch USS blind decodes per CC is 32 or 48 depending on configuration of UL MIMO - The UE is not expected to receive EPDCCH in a special subframe with special subframe configuration 0 or 5 in normal CP, or special subframe configuration 0, 4, or 7 in extended CP. - The ereg to RE mapping is fixed in specifications given the Frame structure type, subframe configuration and CP length Special subframes with the same DMRS positions have the same ereg to RE mapping 18

19 The ereg to RE mapping does not depend on the PRB pair#, subframe#, legacy control region size, DwPTS length or presence of other signals such as CRS,CSI-RS,PRS, - ereg indices are sequentially mapped to the REs without REs for DMRS (24 for normal CP and 12 for extended CP) in a frequency first and then time manner, within each PRB pair It is FFS whether to support cyclic shift of the assigned ereg indices in each OFDM symbol or further rearrangement in the OFDM symbols carrying DMRS. Note that X thresh = 104 was derived to keep the worst case coding rate close to 0.8. Note that epdcch is not mapped to GP or UpPTS. Indication to the UE of epdcch PRB Pairs epdcch dynamically PRB PDSCH resource allocation impact epdcch blocking localized distributed transmission Panasonic direct enumeration of PRB# PRB signaling bit ( full bitmap ) NEC group 4 bits 8 bits dynamically epdcch PRB pair Samsung distributed epdcch 1 bit 2 bits ecfi UE PRB pairs localized epdcch 2 bits CSI UE PRB pairs Fujitsu epcfich dynamically distributed epdcch resource size CATT PDSCH epdcch dynamically epdcch resource Alcatel-Lucent Alcatel-Lucent Shanghai Bell RRC signaling PRB pair PRB pair PDSCH LG Electronics epdcch PRB 19

20 Motorola Mobility epdcch RB peak rate Qualcomm Inc. epdcch resource allocation RRC semi-statically NTT DOCOMO blocking resource epdcch PRB pairs multiple epdcch sets per UE method epcfich method Renesas Mobile Europe Ltd dynamically epdcch PRB Ericsson ST Ericsson epdcch resource epdcch cluster cluster n setsepcfich supported Rel-11 AgreementProposal Conclusion: [I] Agreement: - When a UE detects its DL assignment defining a PDSCH allocation which overlaps with the PRB pair(s) containing its DL assignment, the UE shall assume that the PDSCH scheduled by its DL assignment is rate-matched around the PRB pair(s) containing its DL assignment. In addition, the UE shall assume that the PDSCH scheduled by its DL assignment is not mapped to that PRB pair(s) containing its DL assignment on any layer. [II] Proposal: - A set of reserved PRB-pairs for epdcch transmission may be signaled o all UE-specifically by RRC to the UEs. - UE shall assume these PRB-pairs are not used for PDSCH. [III] Conclusion: No additional solution in Rel-11. Remaining Aspects of PUCCH Resource Allocation PUCCH epdcch 2 way forward contribution: SharpNokia Siemens NetworksNokiaSamsung Docomo 20

21 epdcch semi-static starting PUCCH index ecce PUCCH dynamic offset dynamic offset PDCCH epdcch epdcch set epdcch based HARQ-ACKPUCCH format 1a/1b : necce ecce indexn(1)pucch PUCCH MPUCCH-EPDCCH epdcch semi-static starting PUCCH indexmepdcch-dynamic_offset dynamic offset ZTEEricssonST-EricssonHuaweiHiSiliconChina Telecom RIMNew Postcom Mediatek UE-specific PUCCH (NPUCCH) epdcch localized epdcch PUCCH format 1a/1b index : Floor ( (Index of the 1st ecce of the corresponding epdcch) / N) * N N PRB pair ecce distributed epdcch PUCCH format 1a/1b index Index of the 1st ecce of the corresponding epdcch Agreement: - Lowest ecce index of the corresponding EPDCCH is a component of PUCCH resource determination - A UE is configured with a semi-static PUCCH resource starting offset for each EPDCCH set; ecce is indexed per EPDCCH set - FFS until RAN1#70bis between - Option A) not to use dynamically signaled PUCCH resource offset by EPDCCH - Option B) dynamically signaled PUCCH resource offset by EPDCCH - Whichever of options A and B is chosen, RRC signalling will not be introduced. - FFS until RAN1#70bis for localized EPDCCH among 21

22 - Option X) not to use antenna port index - Option Y) to use antenna port index of EPDCCH - Option Z) to use antenna port index of PDSCH TDD aspects are FFS if solutions are needed, aim for solutions without RRC impact. 4. Small Cell Enhancement 3GPP 6 3GPP Workshop Small Cell Enhancement RAN Chairman Takehiro Nakamura (from NTT Docomo) Small Cell Enhancement scenario RAN1 Technical Report (TR)Small Cell scenarios 1. With or Without Marco Coverage With or Without Marco Coverage Under macro coverage: Small cell nodes are deployed under the coverage of an overlaid macro-cell layer. Outside macro coverage: Small cell nodes are deployed without an overlaid macro-cell layer. In principle, both standalone and non-standalone operation modes should be targeted for 22

23 i. Backward compatibility can be discussed independently. ii. F1 and F2 can be the same i.e. co-channel deployment. Spectrum aspects are different topics. Further question: what could be the difference compared with Rel-11 when F1 and F2 are the same? iii. The definition of small cell, considering coverage, transmission power and other aspects? Can be clarified in the discussion. iv. Is simultaneous connection possible for standalone small cells under marco coverage? v. The definition of standalone operation needs to be clarified. vi. Should we distinguish Marco with Pico? TR scenario 2. Sparse or Dense For throughput performance (RAN1 studies), Dense deployments should be prioritized compared to Sparse deployments. For dense deployments, enhanced technique, such as interference mitigation, would need to be studied. For mobility/ connectivity performance (RAN1/ 2/ 4 studies), both Sparse and Dense deployments should be prioritized. For sparse deployments, enhanced inter-frequency mobility between macro and small cell would need to be studied. For dense deployments, enhanced intra-frequency mobility between small cells would need to be studied. i. Traffic aspects can be addressed in the traffic section. ii. For mobility/connectivity performance. the co-channel case for sparse should be addressed in the discussion. iii. Solutions for dense deployments can also apply to sparse case. Common solutions are desirable. iv. Intra-small cell cluster case? v. Sparse or dense deployment of small cells are independent to with/without marco overlay. vi. Is density a function of the frequency of small cells? vii. The possibility of multiple carriers within the same frequency band to be 23

24 viii. ix. clarified. Even in the sparse case, capacity needs to be studied. Multi-RAT small cells to be clarified TR Throughput performance dense scenario mobility connectivity performance scenario 3. Outdoor or Indoor (UE Speed) One of the key points to differentiate outdoor/ indoor is the requirements for UE speed. For indoor UE, only low UE speed is targeted for. For outdoor, not only low UE speed, but also medium UE speed is required. For throughput performance (RAN1 studies), low UE speed should be mainly considered. For mobility/ connectivity performance (RAN1/ 2/ 4 studies), not only low UE speed, but also middle UE speed should be taken into account. i. The necessarily of this outdoor/indoor dimension mainly for the mobility purpose. ii. The scenarios for indoor sites to support outdoor UEs and vise versa. iii. Channel models can be different for outdoor and indoor. iv. Medium UE speed whether and how this speed presents in the small cells? TR indoor UE low speed 0 km/h ~ 15km/h outdoor UE medium-low speed 15km/h ~ 60km/h 4. Ideal or Non-Ideal Backhaul Both ideal/ non-ideal backhaul options should be studied regardless of other factors of scenarios, i.e. with/without macro coverage, sparse/dense and indoor/outdoor. Relaying/ self-backhauling should be taken into account as one technique for the backhauling of the small cell nodes. 24

25 What non-ideal means would need to be clarified/ studied further in the SI phase. i. For further discussion: whether non-ideal or limited backhaul is prioritized. High-quality backhaul is not precluded. ii. Is out-band relay or in-band relay be targeted for? To be discussed. TR ideal non-ideal network latency scenario Core Network Cost and Energy EfficiencySecurity Small Cell Femto Cell Pico Cell Small Cell 5. Coordinated Multi-Point (CoMP) Operations Measurement for DL CoMP DL CoMP interference measurement CoMP Scheme interference hypothesis interference hypothesis CQI : 1. Alt1: configured IMR IMR interference hypothesis : CQI link adaptation : CSI-RS overhead CSI feedback overhead 2. Alt2: configured 1 IMR UE interference hypothesis ( TPs outside from CoMP measurement set) UE ( TP NZP CSI-RS) : CQI & link adaptation Alt 1 CSI-RS overhead : CSI feedback overhead 3. Alt3: configured 1 IMR TPs outside from CoMP measurement set 25

26 TP CQI CQI CQI : CSI overhead :CSI link adaptation R WayFoward Ericsson, InterDigital, MediaTek, Motorola Mobility, New Postcom, Panasonic, Renesas, Samsung, ST-Ericsson, Texas Instruments, ZTE Alt2 NZP-CSI-RS UE emulation R WayFoward RAN1 spec UE measure interference implementation RES interference measurement CoMP UE DOCOMO R WayFoward hopping hopping IMR granularity IMR granularityaccurate IMR CATT TDD subframe IMR 2REs per PRB 2 REs/PRB 4 REs/PRB 4 REs/PRB LS RAN4 2 REs/PRB 26

27 maximum IMR maximum IMR 3 CSI process CSI Feedback Modes for CoMP CoMP (remote radio headsrrh) (cell-edge)( CoMP) LTE-A Rel. 11 CoMP RIPMIRI ( channel state information CSI)(precoder) CoMP Rel. 11 CSI-RS CSI (periodic)(aperiodic) periodic feedback PUCCH aperiodic feedback PUSCH Periodic Feedback CSI Carrier Aggregation (CA) CoMP (per CSI-RS resource feedback) CA (component carrier CC) (time-division multiplexing or TDM) CSI 27

28 Process (period) offset subframe CSI Process CSI-RS IMR CSI: Conclusion: Working Assumption is independent configuration of multiple CSIs for periodic report o At least in the case of PUCCH: o In case 2 or more CSIs are configured in the same reporting instance(s), FFS the details of Collision handling Compression/multiplexing Observation: By configuring 2 or more CSIs with the same set of reporting instances, it is possible to compress/multiplex multiple CSIs into the same set of reporting instances discussion - All the Rel 10 CSI reporting modes are supported for CoMP in Rel All the Rel 10 CSI reporting types are supported for CoMP in Rel The Rel 10 rules for collisions between different CSI reports in the non-ca case also apply for non-ca CoMP for the case of collision between CSI reports within one CSI process. Rel 10 CSI reporting modes CSI reporting types CSI reports Rel 10 CA CC CSI report CoMP CSI / CSI-RS resource Collision Handling Way Forward R Way Forward on CoMP Feedback Modes Samsung, DoCoMo, Intel, InterDigital, LGE, NEC, Ericsson. Also supported by ST-Ericsson - Working assumption is that Rel-11 supports the feedback configuration and reporting for simultaneous CA and CoMP. 28

29 Strive for reduction of UE complexity in CSI report design, e.g. limiting number of CSI processes, etc Huawei CATT Rel 11 CoMP CA Rel 10 index CSI Process CSI Process - Indexing scheme for CSI processes: Alt1: Indexing is defined within a given CC Alt 2: Indexing is defined across all configured CCs Working Assumption Alt 1 dropping rule - Dropping rule is supported based on reporting type and CSI process/cc index Alt 1: Reporting type CSI process index CC index Alt 2: Reporting type CC index CSI process index Alt 1 dropping UE CSI process index CSI process index CA CC index CSI Multiplexing CSI reports subframe CA enhancements CA enhancements PUSCH payload size CSI processes CSI multiplexing Alt1: FFS, taking into account the outcome of CA CSI multiplexing discussion Alt2: At least for single carrier case, support Multiplex of CSIs on PUSCH Alt3: At least for single carrier case, support Multiplex of CSIs on PUCCH format 3 Compression CSI subframe Ericsson Way Forward: R WF on rank and subband constraints between CSI Processes Ericsson, Alcatel Lucent, Alcatel-Lucent Shanghai Bell, CATT, ST-Ericsson, Huawei, Hi Silicon, LG Electronics, New Postcom, Samsung 29

30 working assumptions Take the following two bullets as working assumption. A RI-reference-process can be configured for a CSI process PMI/CQI of the process is calculated conditioned on the RI of its RI-reference-process,if configured, that is reported in the same or the most recent preceding subframe A subband-reference-process can be configured for a CSI process Subband CQI of the process reflecting transmission over the same subband as indicated for the subband-reference-process of the same bandwidth part, that is reported in the same or the most recent preceding subframe UE RI reference CSI process RI subband index CSI processes CSI processes PMI/CQI reference CSI process RI s0 s1 s2 s3 s4 s5 s6 s7 s8 s9 s0 s1 s2 s3 s4 s5 s6 s7 s8 s9 CSI process 0 report T1 T2 T3 T3 T1 T2 CSI process 1 report T4 T5 T5 T4 T1 T2 T3 T4 T5 Reporting type for RI report Reporting type for wideband PMI/wideband CQI report Reporting type for subband CQI and preferred subband label report Reporting type for wideband CQI report Reporting type for subband CQI report RI reference PMI reference RI reference for Subband CQI PMI reference for Subband CQI 30

31 s0 s1 s2 s3 s4 s5 s6 s7 s8 s9 s0 s1 s2 s3 s4 s5 s6 s7 s8 s9 CSI process 0 report T1 T2 T3 T3 T1 T2 CSI process 1 report T4 T5 T5 T4 T1 T2 T3 T4 T5 Reporting type for RI report Reporting type for wideband PMI/wideband CQI report Reporting type for subband CQI and preferred subband label report Reporting type for wideband CQI report Reporting type for subband CQI report Subband reference ( Nokia) discussion Rel 11 Aperiodic Feedback aperiodic feedback DCI 2-bits CSI request field UE CSI-RS resource RRC field bit CoMP CA future-proof Agreement In the case of a single CC configuration where multiple CSIs are configured for COMP, 2-bit CSI request field will be used in DCI format 0 (if in UE SS) and DCI format 4 for triggering of aperiodic CoMP feedback The candidate CSI reports are configured by RRC 1-bit CSI request remains in format 0 in the case of CSS FFS the payload content of the report Multiple CSI feedbacks could be multiplexed within one report instance FFS how to configure these multiple feedbacks into one report instance 31

32 FFS the semi-static and dynamic signaling details FFS if CSI request field is extended to other than 2 bits, by adding new bits or using existing codepoints FFS simultaneous usage of CSI request field for CoMP and Carrier Aggregation contributions DCI CSI Request UL CoMP UL CoMP SRS power control reference signal SRS power control Ericsson R A-SRS SRS power control RS A-SRS TPC overhead UE handle SRS PC UE A-SRS FFS offset discussion UL reference signal Rel-101 UL 4 SRS aperiodic SRS Rel-11 HetNet SRS UE CSI SRS resource cell CoMP UE CoMP UE CoMP UE SRS RRH cell ID RRH SRS base sequence SRS resource TP UE ProposalSRS sequence cell ID UE-specific 32

33 virtual cell ID Inetl contribution R virtual cell ID Rel-10 cell-specific sequence UL CoMP UE-specific SRS sequence virtual cell id virtaul cell id SRS resource configuration implementation KDDI5 R WayFoward A-SRS frequency hopping frequency hopping mechanism Rel-8 periodic SRS A-SRS Rel-10 proposal trigger WF frequency hopping wider-band sounding power-limited UE frequency hopping multi shot Rel-11 Rel-12 SI WI Rel-12 RAN Planery Enhanced Small Cell discussion small cell Rel-12 (support of different TDD UL-DL configurations on different bands)(soft buffer partitioning) 33

34 RAN1 #70 bis (cross-carrier scheduling)(self-scheduling) (epdcch)(ran1 #70 bis)(dci) (PDSCH) Small Cell Enhancement RAN chairman discussion Scenarios Requirements 3GPP RAN#57 3GPP RAN1 Study Item Small Cell Samsung Alcatel lucent 34