Complementary to the fine-grained channel state information (CSI) and coarse-grained received signal strength indicator (RSSI) measurements, the mid-grained spatial beam attributes [i.e., beam SNR (bSNR)] during the millimeter-wave (mmWave) beam training phase were recently repurposed for Wi-Fi sensing applications, such as human activity recognition and indoor localization. This article proposes a multiband Wi-Fi sensing framework to fuse features from both CSI from 5-GHz bands and the mid-grained bSNR at 60 GHz with feature granularity matching (GM) that pairs feature maps from the CSI and bSNR at different granularity levels with learnable weights. To address the issue of limited labeled training data, we propose to pretrain an autoencoder-based multiband Wi-Fi fusion network in an unsupervised fashion. For specific sensing tasks, separate sensing heads can be attached to the pretrained fusion network with fine-tuning. The proposed framework is thoroughly validated for three sensing applications using in-house experimental data sets: 1) pose recognition; 2) occupancy sensing; and 3) indoor localization. Comparison to a list of baseline methods demonstrates the effectiveness of GM. An ablation study is performed as a function of the amount of labeled data, the latent space dimension, and learning rates.

中文翻译：

---

具有匹配特征粒度的多频段 Wi-Fi 传感


作为细粒度信道状态信息 (CSI) 和粗粒度接收信号强度指示符 (RSSI) 测量的补充，毫米波 (mmWave) 波束期间的中粒度空间波束属性 [即波束 SNR (bSNR)]训练阶段最近被重新用于 Wi-Fi 传感应用，例如人类活动识别和室内定位。本文提出了一种多频段 Wi-Fi 传感框架，可融合 5 GHz 频段的 CSI 和 60 GHz 的中粒度 bSNR 的特征，并具有特征粒度匹配 (GM)，该特征粒度匹配 (GM) 将不同粒度级别的 CSI 和 bSNR 的特征映射进行配对具有可学习的权重。为了解决有限标记训练数据的问题，我们建议以无监督的方式预训练基于自动编码器的多频段 Wi-Fi 融合网络。对于特定的传感任务，可以将单独的传感头连接到经过微调的预训练融合网络。所提出的框架使用内部实验数据集针对三种传感应用进行了彻底验证：1）姿势识别； 2) 占用感应； 3）室内定位。与一系列基线方法的比较证明了 GM 的有效性。消融研究是根据标记数据量、潜在空间维度和学习率来执行的。