On Fundamental Limits of Transmission Activity Detection in Fluid Antenna Systems

In this letter, we develop a unified Cramér-Rao bound (CRB) framework to characterize the fundamental performance limits of transmission activity detection in fluid antenna systems (FASs) and conventional multiple fixed-position antenna (FPA) systems. To facilitate CRB analysis applicable to activity indicators, we relax the binary activity states to continuous parameters, thereby aligning the bound-based evaluation with practical threshold-based detection decisions. Closed-form CRB expressions are derived for two representative detection formulations, namely covariance-oriented and coherent models. Moreover, for single-antenna FASs, we obtain a closed-form coherent CRB by leveraging random matrix theory. The results demonstrate that CRB-based analysis provides a tractable and informative benchmark for evaluating activity detection across architectures and detection schemes, and further reveal that FASs can deliver strong spatial-diversity gains with significantly reduced complexity.

💡 Research Summary

This paper develops a unified Cramér‑Rao bound (CRB) framework to quantify the fundamental performance limits of transmission activity detection in fluid antenna systems (FASs) and conventional multiple fixed‑position antenna (FPA) systems. The authors first relax the binary activity variables (0/1) to continuous real‑valued parameters, which makes the parameter space differentiable and allows the application of CRB theory. This relaxation aligns the bound‑based analysis with practical threshold‑based decisions used in real detectors.

Two representative detection formulations are considered. The first is a covariance‑oriented model, where the receiver exploits second‑order statistics of the received signal. For a multi‑antenna FPA system with M antennas, the received signal is Y = HBS + Z, where H is the channel matrix, B = diag(b₁,…,b_K) contains the (relaxed) activity coefficients, and Z is white Gaussian noise. By defining effective power parameters θ_k = σ_h² b_k², the empirical covariance matrix R ≈ Σ_k θ_k s_k s_kᴴ + σ_z² I is obtained. Using the Slepian‑Bangs formula, the Fisher information matrix (FIM) element becomes J_{ij}=M·|s_iᴴ R⁻¹ s_j|². When the pilot sequences are orthogonal (SᴴS = L p̄ I), a closed‑form CRB (equation 8) is derived via the Woodbury identity. In the non‑orthogonal case the K×K FIM must be constructed and inverted, revealing that pilot non‑orthogonality and multi‑user interference cause an information loss that scales with the ratio K/L.

The second formulation is a coherent model, appropriate for single‑snapshot detection. The signal model is y = Φb + z, where Φ = diag(g)·Sᵀ contains the effective spatial‑temporal signatures (g are the channel gains) and b is the vector of continuous activity coefficients. The FIM for real parameters b is J_{ij}=2σ_z² Re{φ_iᴴ φ_j}. By partitioning the FIM and applying the block‑matrix inversion lemma, the diagonal element of the inverse is