Advanced Channel Decomposition Techniques in OTFS: A GSVD Approach for Multi-User Downlink

TL;DR

This work tackles reliable two-user downlink communication in OTFS by applying a generalized singular value decomposition to the dual-user channels and designing corresponding precoding and detection matrices. The authors derive analytical expressions for three channel-matching scenarios and demonstrate BER improvements over conventional multi-user OTFS with MMSE receivers or precoders, under both perfect and imperfect CSI, with a zero-forcing special case in one regime. The GSVD-based architecture diagonalizes the effective channel, enabling simple per-stream equalization and, in some regimes, sharing of data streams across users, while preserving user privacy in others. The results indicate significant practical gains for high-mobility scenarios and establish a foundation for extending the approach to more users and uplink configurations, highlighting robust performance under channel estimation errors. Overall, the method offers a principled way to manage inter-user interference in OTFS downlink through channel decomposition and diagonalization, with potential impact on future 6G designs.

Abstract

In this paper, we propose a multi-user downlink system for two users based on the orthogonal time frequency space (OTFS) modulation scheme. The design leverages the generalized singular value decomposition (GSVD) of the channels between the base station and the two users, applying precoding and detection matrices based on the right and left singular vectors, respectively. We derive the analytical expressions for three scenarios and present the corresponding simulation results. These results demonstrate that, in terms of bit error rate (BER), the proposed system outperforms the conventional multi-user OTFS system in two scenarios when using minimum mean square error (MMSE) equalizers or precoder, both for perfect channel state information and for a scenario with channel estimation errors. In the third scenario, the design is equivalent to zero-forcing (ZF) precoding at the transmitter.

Figures (5)

• Figure 1: Schematic of different scenarios in GSVD-based channel decomposition for precoding and detection matrix design.
• Figure 2: vs for Scenario I, comparing -based precoding and precoding.
• Figure 3: vs for Scenario II, comparing -based precoding and equalization.
• Figure 4: vs for Scenario III, comparing -based precoding and precoding.
• Figure 5: vs for Scenario II, comparing -based precoding and equalization in the presence of estimation error.