Wireless body area networks (WBANs) is a new research hotspot with great development prospects. The non-contact sensing based on radio frequency signal can solve the issues of personal comfort and privacy. Detection of cervical motion range and cervical strain in time are important in diagnosis and prevention of cervical spondylosis. In this paper, channel state information is used to achieve smart perception and monitoring, timely and efficient detection of different postures and abnormal bending of the neck. It provides an efficient way for protecting cervical health, and also some help for doctors to understand the causes of cervical vertebral disease in a timely manner. The classification accuracy of the four activities reached 99.4%, 99.7%, 99.5% and 99.3%, respectively.
Keywords: cervical spine health protection; cervical spondylosis; neck activity; non-contact sensing.