No Need of Data Pre-processing

Device-free context awareness is important to many applications. There are two broadly used approaches for device-free awareness, i.e., video-based and radio-based. Video-based can deliver good performance, but privacy a serious concern. Radio-based applications have drawn researchers' attention instead, because it does not violate radio signal penetrate obstacles. The existing works design explicit methods each radio-based application. Furthermore, they use one additional step extract features before conducting classification exploit deep learning as tool. Although this feature extraction helps explore patterns of raw signals, generates unnecessary noise information loss. CSI without initial data processing was, however, considered no usable patterns. In article, we the first propose an innovative learning–based general framework both classification. key novelty article that be generalised all with CSI. We also eliminate extra work from signals. conduct extensive evaluations show superior performance our proposed method its generalisation.