Overview
This cross-section entry covers the 5G NR signal structure aspects of the PRS-based ISAC paper. For the full summary including radar processing, CRLB derivation, and frame structure proposals, see the primary page: 5G PRS-Based Sensing (primary).
5G NR Reference Signal Comparison for Radar Sensing
The paper systematically compares four 5G NR downlink reference signals as candidates for radar sensing:
| Signal | Sequence | Time-domain | Frequency-domain | Sensing suitability |
|---|---|---|---|---|
| SS (PSS+SSS) | 127-chip M/Gold | Fixed in SSB (symbols 0,2) | 127 subcarriers | Poor — very limited resources |
| DMRS | Gold | Front-loaded (1–2 symbols) | Up to 240 consecutive subcarriers, Comb 4 | Moderate — restricted bandwidth |
| CSI-RS | Gold | 1/2/4 symbols | Up to 52 PRBs, multiplexed across ports | Poor — very sparse RE mapping |
| PRS | Gold, length 4096 | Multiple consecutive slots | Up to 272 PRBs, Comb 2/4/6/12 | Best — richest resources, longest sequence |
PRS wins because it has the longest Gold sequence and the most flexible time-frequency resource allocation of any 5G NR downlink pilot.
PRS Configuration Details (TS 38.211 / TS 38.214)
- Sequence ID: n_ID,seq^PRS ∈ {0, …, 4095} — uniquely identifies a PRS resource
- Comb patterns: Comb-2, Comb-4, Comb-6, Comb-12 (frequency-domain subcarrier spacing)
- Symbol configurations: 2, 4, 6, or 12 OFDM symbols per slot
- Bandwidth: 24–272 PRBs (granularity of 4 PRBs)
- Frequency ranges: FR1 (450 MHz–5.9 GHz) with µ=0,1,2; FR2 (24.2–52.6 GHz) with µ=2,3,4
- Purpose: Primarily designed for UE positioning via time-difference-of-arrival (TDOA) from multiple base stations
Subcarrier Spacing and Sensing Tradeoff
| µ | Δf (kHz) | T_s (µs) | Best for |
|---|---|---|---|
| 0 | 15 | 71.35 | FR1, large range window |
| 1 | 30 | 35.68 | FR1 general |
| 2 | 60 | 17.84 | FR1/FR2 overlap |
| 3 | 120 | 8.92 | FR2, better velocity resolution |
| 4 | 240 | 4.46 | FR2 mmWave, finest velocity |
Larger µ → smaller T_s → better velocity CRLB but worse range CRLB (fundamental tradeoff).