Refined Metrics, Sensing Limits, and Resource Allocation in OTFS-RSMA LEO ISAC
Bruno Felipe Costa and Taufik Abrão · 2025
Read Refined Metrics, Sensing Limits, and Resource Allocation in OTFS-RSMA LEO ISAC by Bruno Felipe Costa and Taufik Abrão (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-02624.txt.
Introduction
Bruno Felipe Costa and Taufik Abrao design a combined communication and sensing system for satellites in low Earth orbit in a 2025 study. The framework pairs a modulation built for fast motion with a sharing scheme that splits signals among users. Satellite links suffer imperfect channel knowledge and leftover interference after cancellation, which the authors build into their measures.
They derive sharper signal quality formulas and fundamental sensing limits through an estimation bound that includes echo gain. Resource allocation is then posed as a fairness problem, maximizing the worst user's share under those limits. The results guide waveform and power choices for orbit networks.
Catalog record
| CREATOR | Bruno Felipe Costa and Taufik Abrão |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 25 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-02624
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Refined Metrics, Sensing Limits, and Resource Allocation in OTFS-RSMA LEO ISAC Bruno Felipe Costa and Taufik Abrão Abstract This paper develops an integrated Orthogonal Time Frequency Space (OTFS) - Rate-Splitting Multiple Access (RSMA) framework employing advanced Signal Processing (SP) techniques tailored for this demanding environment. We derive refined communication performance metrics, specifically Signal-to-Interference-plus-Noise Ratio (SINR) expressions capturing the practical effects of Imperfect Channel State Information (ICSI) and Imperfect Successive Interference Cancellation (ISIC) . Moreover, fundamental sensing limits are established via Cramér-Rao Bound (CRB) derivation incorporating parameter-dependent echo gain, linking waveform SP properties to estimation accuracy. The resource allocation is formulated as a non-convex optimization problem aiming for Max-Min Fairness under constraints derived from these SP metrics. Illustrative results, obtained via Genetic Algorithm (GA) optimization, crucially demonstrate that the proposed RSMA scheme uniquely enables the simultaneous satisfaction of stringent communication and sensing constraints metrics, a capability not achieved by conventional Space-Division Multiple Access (SDMA) . Such results highlight the efficacy of integrated OTFS-RSMA precoding and optimization approach for designing robust and feasible Low Earth Orbit (LEO) - Integrated Sensing and Communication (ISAC) systems. I Introduction Next-generation wireless systems, including 6G, envision Space-Air-Ground Integrated Network (SAGIN) architectures providing ubiquitous connectivity and environmental awareness. Integrated Sensing and Communication (ISAC) is a key enabler, enhancing spectral efficiency by using dual-function waveforms [ 1 ] . Low Earth Orbit (LEO) satellite systems are attractive platforms for global ISAC deployment [ 2 ] , but introduce significant Signal Processing (SP) challenges. The high satellite velocity causes large Doppler shifts, degrading conventional modulations like Orthogonal Frequency Division Multiplexing (OFDM) [ 3 ] . Furthermore, managing multiuser interference and ensuring reliable sensing under power constraints and dynamic channel conditions requires sophisticated SP solutions. Orthogonal Time Frequency Space (OTFS) modulation emerges as a powerful SP technique to combat high Doppler [ 4 ] . By operating in the Delay-Doppler (DD) domain, OTFS transforms the doubly-selective channel into a quasi-static representation, offering inherent resilience [ 5 ] . This DD domain operation also aligns naturally with radar parameters (range/delay, velocity/Doppler), making OTFS suitable for ISAC [ 1 ] . Key SP challenges in OTFS include efficient channel estimation [ 6 ] , robust data detection (e.g., using Message Passing (MP) / Belief Propagation (BP) algorithms exploiting channel sparsity [ 7 ] ), and managing inherent DD interference [ 5 ] . Figure 1 : LEO motion impact on DD channel: Time-varying geometry causes Doppler shifts, delay variations, and multipath spread. To manage multiuser interference, Rate-Splitting Multiple Access (RSMA) provides a flexible SP framework [ 8 ] . By splitting messages and employing advanced precoding, RSMA offers robust interference control compared to Space-Division Multiple Access (SDMA) or Non-Orthogonal Multiple Access (NOMA) , especially under practical impairments like Imperfect Channel State Information (ICSI) [ 9 ] . Integrating the Doppler resilience of OTFS with the interference management flexibility of RSMA is promising for LEO ISAC [ 10 , 11 ] . However, the core SP task lies in designing resource allocation strategies that optimally balance communication Quality of Service (QoS) (considering fairness [ 12 ] ), sensing precision, and power efficiency under realistic channel conditions and receiver imperfections. This paper delves into the advanced SP aspects of such an integrated OTFS - RSMA LEO ISAC system. The main contributions of this work are fivefold: i ) A concise system model incorporating advanced signal processing techniques, namely OTFS and RSMA , tailored for realistic LEO channel characteristics (including DD sparsity) and practical impairment models ( ICSI , and Imperfect Successive Interference Cancellation (ISIC) ). ii ) Derivation of refined communication performance metrics ( Signal-to-Interference-plus-Noise Ratio (SINR) ) using Linear Minimum Mean Square Error (LMMSE) analysis, explicitly accounting for ICSI and ISIC effects crucial for robust SP system design. iii ) Derivation of the fundamental sensing performance limits via ( Cramér-Rao Bound (CRB) ) based on the Fisher Information Matrix (FIM) , considering parameter-dependent echo gain variations and explicitly linking estimation accuracy to the second-order moments of the Delay-Doppler energy distribution of waveforms, determined by the precoders \bm{\mathcal{P}} . iv ) Formulation and analysis of a fairness-aware (Max-Min) resource allocation problem from an SP perspective, highlighting its non-convex nature and discussing potential SP -centric optimization approaches. v ) Numerical demonstration, by deploying evolutionary heuristic GA optimization, in which the integrated RSMA strategy uniquely enables the simultaneous satisfaction of stringent communication (common rate QoS ) and sensing ( CRB QoS ) constraints, showcasing a feasibility advantage over conventional SDMA within the proposed OTFS-RSMA LEO ISAC networks. II System Model: Signal Processing Perspective We consider a downlink Multi-User (MU) - Multiple-Input Single-Output (MISO) ISAC system where a LEO satellite ( N_{t} antennas) serves K single-antenna users using OTFS modulation and RSMA . We focus on the core SP equations and models. OTFS Modulation Principles . OTFS maps information symbols x[l,m] onto an M\times N DD grid ( l : delay, m : Doppler). These are transformed to the Time-Frequency (TF) domain via the Inverse Symplectic Finite Fourier Transform (ISFFT), yielding intermediate TF symbols denoted as X[n,i] (where n is the time index, 0\leq n<N , and i is the frequency index, 0\leq i<M ), and then mapped to the time domain via the Heisenberg transform [ 4 ] . The receiver reverses this using the Wigner transform and SFFT. Its key SP advantage for LEO satellite networks is transforming the time-varying channel into a quasi-static DD representation, mitigating Doppler effects [ 5 ] . RSMA Strategy . RSMA splits user k ’s message into common s_{c} ( \mathbb{E}[|s_{c}|^{2}=1 ) and private s_{p,k} ( \mathbb{E}[|s_{p,k}|^{2}=1 ) symbols. Linear precoding in the DD domain provides flexible interference management, a core SP task [ 13 ] . LEO Channel and DD Representation . The physical channel between satellite antenna n_{t} and user k is time-varying, characterized by a DD impulse response h_{k,n_{t}}(\tau,\nu;t) comprising P paths [ 4 ] : \begin{split}h_{k,n_{t}}(\tau,\nu;t)=\sum_{p=1}^{P}\alpha_{p,k,n_{t}}&(t)e^{j\psi_{p,k,n_{t}}(t)}\\ &\!\!\!\!\!\!\!\!\times\delta(\tau-\tau_{p,k}(t))\delta(\nu-\nu_{p,k}(t)).\end{split} (1) Parameters (\alpha,\psi,\tau,\nu) depend on orbital dynamics, geometry, and fading [ 14 , 15 ] . This physical channel results in an effective N_{dd}\times N_{dd} DD channel matrix \mathbf{H}_{k}(t) relating transmitted and received DD symbols. Crucially for SP , due to the typically small number of paths P\ll MN in satellite links, \mathbf{H}_{k}(t) is often sparse, which can be exploited by advanced receivers. III Advanced Performance Metrics: Derivations Accurate assessment and optimization require precise SP metrics that account for practical impairments. We detail the derivations for refined SINR and CRB . Refined SINR under ICSI and ISIC . We derive approximate SINR expressions for LMMSE receivers designed with \hat{\mathbf{H}}_{k} , explicitly incorporating ICSI (variance \sigma_{e}^{2} ) and ISIC (factor \Theta_{k} ), following established methodologies [ 16 ] . The filter vectors are \mathbf{w}_{c,k},\mathbf{w}_{p,k}\in\mathbb{C}^{1\times N_{dd}} . III- 1 Common Stream SINR Derivation The desired signal power is P_{S,c}=|\mathbf{w}_{c,k}\hat{\mathbf{H}}_{k}\bm{\mathcal{P}}_{c}|^{2} . Interference comes from private streams. The effective noise includes filtered AWGN and the impact of ICSI on all signal components. The total interference-plus-noise power D_{c} is approximated as: D_{c}\approx\sum_{j=1}^{K}|\mathbf{w}_{c,k}\hat{\mathbf{H}}_{k}\bm{\mathcal{P}}_{p,j}|^{2}+\|\mathbf{w}_{c,k}\|^{2}(\sigma_{n}^{2}+\sigma_{e}^{2}P_{\text{tot}}). (2) The term \sigma_{e}^{2}P_{\text{tot}} accounts for the variance introduced by the channel error \mathbf{E}_{k} acting on the total transmitted signal power P_{\text{tot}} , projected onto the filter space. The refined SINR for the common stream at user k is: \small\mathrm{SINR}_{c,k}^{\text{(ref)}}\approx\frac{|\mathbf{w}_{c,k}\hat{\mathbf{H}}_{k}\bm{\mathcal{P}}_{c}|^{2}}{\sum_{j=1}^{K}|\mathbf{w}_{c,k}\hat{\mathbf{H}}_{k}\bm{\mathcal{P}}_{p,j}|^{2}+\|\mathbf{w}_{c,k}\|^{2}(\sigma_{n}^{2}+\sigma_{e}^{2}P_{\text{tot}})} (3)