Rate-Splitting Multiple Access for Overloaded Multi-group Multicast: A First Experimental Study

TL;DR

The paper reports the first experimental validation of RSMA-based MGM and a three-way comparison with SDMA- and NOMA-based MGM in an overloaded, multi-group setting. Using a 2-antenna TX and two groups of two users each, nine LOS cases reveal that RSMA achieves higher max-min fairness throughput than the alternatives, thanks to flexible common/private stream allocation designed via a WMMSE-optimized precoder ${\mathbf P}$ under imperfect CSIT. The study implements a practical RSMA PHY on an NI USRP-based prototype, with Stage 1 CSIT acquisition and Stage 2 RSMA signal design, and evaluates MCS-limited throughputs that reflect finite alphabet and decoding reliability. The results show RSMA consistently delivers superior fairness across cases, while also highlighting counter-intuitive scenarios where BLER and MCS constraints shape throughput ordering, underscoring RSMA’s practical robustness to interference and channel variability.

Abstract

Multi-group multicast (MGM) is an increasingly important form of multi-user wireless communications with several potential applications, such as video streaming, federated learning, safety-critical vehicular communications, etc. Rate-Splitting Multiple Access (RSMA) is a powerful interference management technique that can, in principle, achieve higher data rates and greater fairness for all types of multi-user wireless communications, including MGM. This paper presents the first-ever experimental evaluation of RSMA-based MGM, as well as the first-ever three-way comparison of RSMA-based, Space Divison Multiple Access (SDMA)-based and Non-Orthogonal Multiple Access (NOMA)-based MGM. Using a measurement setup involving a two-antenna transmitter and two groups of two single-antenna users per group, we consider the problem of realizing throughput (max-min) fairness across groups for each of three multiple access schemes, over nine experimental cases in a line-of-sight environment capturing varying levels of pathloss difference and channel correlation across the groups. Over these cases, we observe that RSMA-based MGM achieves fairness at a higher throughput for each group than SDMA- and NOMA-based MGM. These findings validate RSMA-based MGM's promised gains from the theoretical literature.

Figures (5)

• Figure 1: An illustration of RSMA-based MGM.
• Figure 2: Signal structure within the two-stage transmission protocol used to implement RSMA-based MGM. In conventional (SDMA-based) 802.11g, every DATA (OFDM) symbol contains four precoded pilot subcarriers used to correct phase errors in the estimate of a user's precoded channel. This provides protection against decoding errors caused by rotated constellations post equalization, and is known as fine phase shift (FPS) FinephaseShifting. However, in RSMA-based MGM, a user is associated with two precoders in general. Hence, we use ${\mathbf{p}}_c$ to precode the FPS pilots in DATA symbols 1, 4, 7 etc., ${\mathbf{p}}_1$ to precode the FPS pilots in DATA symbols 2, 5, 8 etc., and ${\mathbf{p}}_2$ to precode the FPS pilots in DATA symbols 3, 6, 9 etc. This is because the existence of the common/private stream precoders is channel-dependent, and the above scheme ensures that a user can correct phase errors at least once every 3 DATA symbols.
• Figure 3: Measurement setup
• Figure 4: Fairness comparison between SDMA-based, NOMA-based and RSMA-based MGM. The number beside each data point indicates the measurement case. The black line ($y=2x$) corresponds to max-min fairness, and points that are closer (in terms of Euclidean distance) to this line represent fairer outcomes.
• Figure 5: The throughput performance for both multicast groups of SDMA-, NOMA- and RSMA-based MGM.

Theorems & Definitions (8)

• Remark 1
• Remark 2
• Remark 3: Achievability
• Remark 4
• Remark 5
• Remark 6
• Remark 7
• Remark 8: Message-splitting