Grouping Strategy for RFID-Based Activity Recognition in Smart Home

Radio frequency identification (RFID), as a kind of non-contact technology, is widely applied to objects tracking, missing products detection, indoor localization and activity recognition. In real life, it is impossible for RFID tags to completely avoid the problems on signal loss, channel conflict and tag broke which cause data loss, thereby reducing the recognition accuracy. In this paper, we propose a dictionary segmentation based sparse representation classification (DSSRC) algorithm to solve the problem. Extensive experiments are conducted to validate our DSSRC algorithm. Finally, our experimental results demonstrate that the proposed algorithm can reach high accuracy. In addition, our DSSRC algorithm can achieve higher recognition efficient compared with the sample dictionary based sparse representation classification (SDSRC) algorithm.