Low-Power Wake-Up Signal Design in 3GPP 5G-Advanced Release 19

The Low-Power Wake-Up Signal (LP-WUS) and Low-Power Synchronization Signal (LP-SS), introduced in 3GPP 5G-Advanced Release 19, mark an important advancement toward power-efficient IoT communications. This paper provides a comprehensive overview of the LP-WUS procedures in the RRC_IDLE and RRC_INACTIVE states and summarizes the key physical-layer design aspects. The LP-WUS is intended to be detected by a low-power energy detector (ED), allowing the main radio (MR) to remain switched off, thereby enabling substantial power savings compared to conventional 5G paging mechanisms. As such, LP-WUS is considered the baseline for 6G WUS design. Furthermore, different receiver architectures are evaluated, highlighting the inherent trade-offs between power-saving gains and coverage performance.

💡 Research Summary

The paper provides a comprehensive examination of the Low‑Power Wake‑Up Signal (LP‑WUS) and Low‑Power Synchronization Signal (LP‑SS) introduced in 3GPP 5G‑Advanced Release 19, focusing on their operation in the RRC_IDLE and RRC_INACTIVE states. The authors begin by highlighting the power‑consumption problem inherent to conventional 5G DRX cycles, where a device must keep its main radio (MR) active for periodic measurements and paging checks, leading to tens of milliwatts of idle power draw. LP‑WUS addresses this by allowing a dedicated low‑power radio (LR) to monitor a specially designed wake‑up signal while the MR remains in an ultra‑deep sleep, thereby dramatically reducing the baseline power consumption.

The paper details the configuration and timing framework that enables LP‑WUS. UE configuration is signaled via the lowPower‑Config‑r19 element in SIB1. The network groups UEs into sub‑groups (SGs) using either core‑network‑assigned indices or UE‑ID‑derived indices, supporting up to 31 SGs per paging occasion (PO). A mapping equation determines which paging occasion (iPO) a UE should monitor based on its UE‑ID, the number of paging frames per DRX cycle, and the number of POs per paging frame. Each paging occasion can be associated with one, two, or four wake‑up occasions (LOs), and each LO can contain multiple monitoring occasions (MOs) per beam. The flexibility to configure N_LO_PO = 1, 2, 4 and N_MO_LO = 1–4 allows the network to balance paging granularity, latency, and overhead.

LP‑SS is introduced as a periodic low‑power synchronization signal that enables the LR to perform the same measurements (e.g., SS‑RSRP, SS‑RSRQ) that the MR would normally acquire from SSBs, and to obtain coarse time‑frequency synchronization for LP‑WUS decoding. Its periodicity (160 ms or 320 ms) and duration (4, 6, 8 OFDM symbols) are configurable, providing a trade‑off between resource overhead and synchronization accuracy.

From a physical‑layer perspective, LP‑WUS employs on‑off keying (OOK) modulation, a binary amplitude‑shift keying scheme that can be detected by a simple energy detector. The OOK symbols are generated in the time domain, transformed by a DFT of size 132 subcarriers, and mapped onto the existing 5G OFDM grid. The signal bandwidth is fixed at 132 subcarriers, and the number of OOK symbols per OFDM symbol (M) can be 1, 2, or 4, giving flexibility in time‑frequency resource allocation. Information bits (B ≤ 5) are encoded using a three‑step process: (i) Reed‑Muller based channel coding for false‑alarm and missed‑detection targets, (ii) rate‑matching to fit the available OOK symbols, and (iii) Manchester coding, which doubles the bit rate but yields a robust energy‑based decision metric at the receiver.

The ON‑sequence for each OOK symbol is a cyclically extended Zadoff‑Chu (ZC) sequence. The network can configure up to N_root = 2 distinct ZC roots, and each root can be cyclically shifted to generate multiple sequences (N_seq). The cyclic shift spacing is maximized to improve correlation properties, and the number of sequences is limited by the chosen M (e.g., N_max_seq = 16 for M = 1). This design enables coherent detection when phase information is available, while still supporting non‑coherent energy detection.

Receiver architectures are evaluated in three categories: (1) pure energy detection, which offers the lowest power consumption but limited sensitivity; (2) correlation‑based sequence matching, which provides higher detection probability and extended coverage at the cost of additional DSP operations; and (3) hybrid schemes that first perform a coarse energy check and then apply correlation only on candidate windows, achieving a middle ground. Simulation results show that LP‑WUS can reduce overall device power consumption by 30 %–70 % compared with conventional paging, while maintaining acceptable miss‑detection rates. However, the coverage loss depends on the number of SGs, the chosen N_root, and the receiver’s processing budget.

In conclusion, the authors demonstrate that LP‑WUS, together with LP‑SS, enables IoT devices to keep their main radio off for the majority of the DRX cycle, thereby achieving substantial battery‑life extensions without violating latency requirements for paging. The flexible grouping, timing, and physical‑layer design make LP‑WUS a strong candidate as the baseline wake‑up mechanism for future 6G networks. The paper also outlines open research directions, including multi‑antenna sequence design, dynamic subgroup management, and real‑world power‑performance validation.