User-Specific Channel Estimation Overhead Optimization and Resource Allocation for Multi-User OTFS Systems

Accurate channel estimation is one of the major challenges in deploying orthogonal time frequency space (OTFS) systems, because the inter-grid interference (IGI) between the pilot and data caused by multi-path channels significantly reduces estimation accuracy. Existing solutions embed the unified guard zero-symbols to prevent IGI in multi-user OTFS systems, but they ignore that users have varying abilities to mitigate IGI based on their specific channel conditions. Consequently, using the same guard for different users leads to redundant guard symbols, which reduces spectrum efficiency. In this letter, we leverage user-specific statistic channel characteristics to design a tailored channel estimation overhead optimization and resource allocation scheme to enhance the spectrum efficiency for multi-user OTFS systems. Specifically, we first derive the mathematical expression of the transmission capacity. Then we formulate a total capacity maximization problem by jointly optimizing the channel estimation overhead and bandwidth, subject to individual rate requirements. To solve this non-convex problem, we introduce an alternative optimization algorithm and derive closed-form expressions for updating the solutions in each iteration.