AI 中文总结
本文针对OFDM-ISAC系统，分析了MF与RF接收机的感知性能，提出几何星座整形框架以权衡通信可靠性与感知精度，在实际空中传输中取得显著增益。
AI 中文摘要
正交频分复用（OFDM）是集成感知与通信（ISAC）系统的关键波形，因其具备高频谱效率且与现代无线标准具有固有兼容性。然而，随机数据调制下其基于估计理论的基础感知性能仍未得到充分探究。本文针对基于OFDM的多目标距离估计ISAC系统，开展了统一且明确的性能分析，重点关注调制星座对感知性能的显著影响。我们构建了一套全面的估计理论框架，以表征匹配滤波（MF）和互易滤波（RF）两种感知接收机架构下的距离估计均方误差（MSE）。理论分析表明，在多目标且杂波丰富的环境中，MF接收机的感知性能受星座四阶矩（峰度）的根本限制，该参数决定了与数据相关的旁瓣干扰电平；而RF接收机则以噪声增强为代价消除此类干扰，其性能由星座二阶矩的倒数决定。基于上述闭式MSE推导，我们提出了一种针对感知接收机的几何星座整形（GCS）框架，通过基于最小欧氏距离（MED）和接收机相关感知指标联合优化星座几何，实现通信可靠性与感知精度的灵活权衡。结果表明，所提星座整形方案在实际空中传输实现中可提供显著性能增益，并能针对不同接收机架构定制感知与通信的权衡。
英文摘要
Orthogonal frequency division multiplexing (OFDM) is a key waveform for integrated sensing and communication (ISAC) systems due to its high spectral efficiency and inherent compatibility with modern wireless standards. However, its fundamental estimation-theoretic sensing performance under random data modulation remains largely unexplored. This paper presents a unified and explicit performance analysis of OFDM-based ISAC systems for multi-target range estimation, focusing on the distinct impacts of the modulation constellation on the sensing performance. We develop a comprehensive estimation-theoretic framework to characterize the range estimation mean-square error (MSE) for both matched filtering (MF) and reciprocal filtering (RF) sensing receiver architectures. Our theoretical analysis reveals that in multi-target and clutter-rich environments, the sensing performance of the MF receiver is fundamentally limited by the fourth-order moment (kurtosis) of the constellation, which determines the data-dependent sidelobe interference level. In contrast, the RF receiver eliminates such interference at the cost of noise enhancement, with its performance governed by the inverse second-order moment of the constellation. Building on these closed-form MSE derivations, we propose a sensing-receiver specific geometric constellation shaping (GCS) framework. By jointly optimizing the constellation geometry based on the minimum Euclidean distance (MED) and receiver-dependent sensing metrics, we enable a flexible trade-off between communication reliability and sensing precision. Our results demonstrate that the proposed constellation shaping provides significant performance gains and facilitates a tailored sensing and communication trade-off across different receiver architectures in practical over-the-air implementations.
Comments13 pages, 12 figures
