R1-2509385
discussion
FLS#1 on evaluation assumptions for 6GR air interface
From Huawei
Summary
This Huawei-moderated Tdoc (R1-2509385) for the RAN1#123 meeting summarizes and proposes updated evaluation assumptions for 6GR air interface studies across 8 main sections, containing approximately 50 proposals/observations covering antenna modeling, SLS parameters, new traffic models for AI/ML and immersive communication, link budget templates, and NTN assumptions.
Position
Huawei proposes updating BS antenna modelling parameters for 700MHz, 2GHz, and 7GHz with specific (M,N,P,Mg,Ng; Mp,Np) tuples. They model UE antenna using five predefined combinations of antenna elements and TXRUs with two alternatives (UPA array or TR 38.901 handheld locations). For AI/ML traffic, they propose a model parameterized by Token as minimum application-layer data unit, with packet size as integer multiple of token size, burst arrivals of N packets with M Type-1 and N-M Type-2 packets of different importance, and Token success rate computed as successfully transmitted tokens divided by total transmitted tokens—rejecting packet success rate as a replacement. They support extending the XR traffic model for immersive communication with higher data rates and frame generation rates for gaming, plus a haptics model with co-generated packets. They favor FTP Model 3 extensions over FTP Model 1 for incorporating PDB and multiple packet sizes, proposing two options where Option1 assigns multiple independent traffic flows per UE. For link budget, they support Candidate 1 template updated with required SNR linked to target data rate at specified BLER and MaxCL calculation from TR 38.913. For 6GR NTN, they propose Ku-band at 14GHz UL/11GHz DL, satellite array modelling with full-connection TXRU virtualization, and consideration of VLEO, LEO, and GEO orbit types.
Key proposals
- Proposal 2.1.2-1 (Sec 2.1): Updates BS antenna modelling parameters (M,N,P,Mg,Ng; Mp,Np) for 700MHz to (8,4,2,1,1;1,4), for 2GHz outdoor Comb1 to (8,2,2,1,1;1,2), and for 7GHz outdoor Comb1 to (24,16,2,1,1;4,16) and Comb3 to (48,16,2,1,1;8,16).
- Proposal 2.2.2-rv2 (Sec 2.2): Defines UE antenna modelling configurations as 5 combinations (1T2R, 2T4R/4T4R, 4T8R/8T8R, 8T16R/16T16R, 16T32R/32T32R) with Alt1 (UPA array parameters) and Alt2 (TR 38.901 handheld/candidate antenna locations).
- Proposal 3.1.2-rv2 (Sec 3.1): Assumes system-level simulation layout with single-layer hex grid for all scenarios, two-layer macro+micro for Dense Urban and Urban Macro, and defines specific ISD values per scenario.
- Proposal 3.2.2-1rv1 (Sec 3.2): Defines total transmit power per BS per deployment scenario and frequency, e.g., around 7GHz Dense Urban Macro BS Option1: 44dBm/20MHz, Option2: 43dBm/20MHz, Micro BS: 33dBm/20MHz, with note that BS Tx power scales with bandwidth under max limits.
- Proposal 4.1.2-rv1 (Sec 4.1.2): Defines generic traffic model for AI/ML services parameterized by packet size (integer multiple of token size), packet arrival (N packets per burst with M Type-1 and N-M Type-2 of different importance), PDB, and Token success rate requirement with packet-level failure counting.
- Proposal 4.2.2-rv1 (Sec 4.2): Defines two amended XR models: Model-1 (eXR without Haptics) adding immersive gaming values (data rate 100/300/500 Mbps, frame rate 90/120 fps, STD 25% of mean) and UL-heavy video uploading values; Model-2 (eXR with Haptics) with co-generated haptics packets and PDB of 12ms or 30ms.
- Proposal 4.3.2-2rv1 (Sec 4.3): Presents two options for extending FTP Model 3 with multiple packet sizes: Option1 (2-3 independent traffic flows per UE) and Option2 (2-3 traffic flows per cell with UE classes consuming traffic fractions αc).
- Proposal 4.5.2-rv1 (Sec 4.5): Defines IoT traffic types (Triggered/polled reporting, Autonomous reporting, Remote actuation, Firmware/software upgrade) with specific traffic models in Tables 1-4, e.g., DL trigger 150 bytes, UL payload 1000 bytes, 1 report/hour/UE.
- Proposal 5.1.2 (Sec 5.1): Updates link budget template for Candidate 1: Row(19) updated to Required SNR = SNR for Target data rate with Y% BLER, Row(30) keeps but removes FFS, adds Row(22bis-a) for MaxCL calculation from TR 38.913.
- Proposal 6.1.2 (Sec 6.1): Sets Ku-band carrier frequency for NTN at 14GHz for UL and 11GHz for DL.
- Proposal 6.2.2 (Sec 6.2): Models satellite antenna as array with TXRU virtualization Option-2B (full-connection), Mg=Ng=1, W matrix with amplitude and phase coefficients, Bessel function from 38.811 for simplified evaluations.
- Proposal 3.3.2-rv2 (Sec 3.3): Defines UE distribution and speed: 10 users/TRxP, Indoor Hotspot 100% indoor 3km/h, Dense Urban 80% indoor 3km/h + 20% outdoor 30km/h, Rural 50% indoor 3km/h + 50% outdoor 120km/h.
- Observation 1 (Sec 4.1): States Token success rate directly reflects AI/ML service quality and cannot be replaced by packet success rate used in XR traffic model.
- Observation 3 (Sec 4.1): Notes Tokens are typically encrypted and encapsulated within standards protocol layers for transmission over air.
- Proposal 3.4.1-rv1 (Sec 3.4): Defines other SLS parameters including ISD per deployment, BS/UE antenna heights, noise figures (5dB below 7GHz, 7dB above 15GHz), channel models from TR 38.901, and MMSE-IRC receiver as baseline.