Design And Implementation Of Life Safety Monitoring System For Elderly Living Alone Based On Internet Of Things

At present,the population in China is confronted with severe aging problems.The size and proportion of the elderly population are continuously increasing.However,the burden caused by aging is massive because of China’s deficient pension system.To make matters worse,a large proportion of the elderly are living alone.Guarding them is difficult and costly,which hinders the process of adapting to aging.Therefore,for a better pension system,it’s urgent to improve the health care level of the solitary elderly.With the development of IoT technology,such as ZigBee,NB-IoT(Narrowband Internet of Things)and LoRaWAN(Long Range Wide Area Network),there are a variety of monitoring systems to protect the life safety of the elderly living alone.However,most of these systems use cameras or wearable devices,which violates privacy and causes inconvenience.In addition,some features of existing IoT communication protocols,such as scalability and roaming,can not meet the needs of small and fixed scenes,like elderly monitoring,but increase the cost of systems.Combining advantages of LoRa(Long Range)and NB-IoT,this thesis designs and implements a monitoring system to ensure the life safety of the elderly living alone.The system uses custom lightweight MAC protocol to construct a LoRa sensor network that collects the indoor environment and activity information.By analyzing the activity data,the system protects the life safety without perception and body contact.It also provides emergency help for the elderly by NB-IoT.The main work of this thesis is as follows:1.A custom lightweight LoRa MAC protocol is designed and implemented.It specifies the physical and the link layer,which respectively uses LoRa and regulates channel access,frame format,encryption,retransmission and acknowledgement.In this thesis,the protocol is implemented on the embedded side and server side separately.2.A LoRa sensor network based on the custom MAC protocol is designed and implemented.It consists of sensor nodes and gateway nodes,while the former includes temperature and humidity nodes,body detection nodes and single door status nodes.A sensor node is composed of a SX1276 radio module,a STM32L073 control module and a sensor module.Besides the same radio and control module,a gateway node also has a 4G transmission module.3.A one-click help system based on NB-IoT is designed and implemented.The system includes help devices with buttons and alarm function.4.A network server is designed and implemented.The network server is divided into three parts:the MQTT(Message Queuing Telemetry Transport)backend,a LoRa server and an API(Application Programming Interface)server.It also has a web interface to provide system management and data visualization for users.5.The life safety anomaly detection algorithm is implemeted and the whole system is tested.By estimating the life pattern of the elderly,the algorithm uses box plot rule to find anomalies over time intervals.Finally,this thesis tests the system function and analyze the results.