| Recently, the train industry has been developed rapidly, and many high speed train have been used. The state and fault information of the train should be detected and dealt with quickly for the safety of the train. But traditional train control system can not meet the need. The train safeguard alarm system, developed by our team, detects the run parameters of the train, analyzes the data, judges the state of the train, and gives an alarm for the hidden troubles.Along with the coming of the information epoch, the rapid development of network technology is changing people's work and life. Network has been used in everywhere of the society, including railway. Train communication network is used in many trains to control on-board equipments, and it has been a perfect solution in train control system. So it is necessary to develop the train safeguard alarm system based on network.The LonWorks technology is developed in recent years and available as an open standard to all manufacturers. It integrates computer, network and control into one technology and displaces proprietary centralized systems with open, highly distributed, interoperable systems. Neuron chip, whose network protocol has all 7 layers of the ISO/OSI reference model, is the core controller of LonWorks network, and it has the abilities of both communication and application control. LonWorks now has been widely used in the automatic areas ranged from industry, building, transportation, to power system, medical treatment.In this paper, according to the application demands of train safeguard alarm system, several widely used communication network and field bus are studied and compared. LonWorks is used as the network through the train, and CAN bus is used in the vehicle. The solution and structure of the whole network are discussed, and the details of the development are also presented, including the development of the node, the hardware and software of the gateway, how to construct the network, and debug it.After finishing the development of the network, system reliability is also discussed and many measures are taken to improve it. Finally, the static debugging and the dynamic debugging on-board are carried out, and the performance of the system is improved through these experiments. |