Currently, Radio Frequency Identification (RFID) is receiving considerable attention nowadays as an enabling technology that can improve supply chain operations, asset tracking and inventory controlling. Along with the instant application of the object identification with RFID technique, people are applying themselves to establish "the internet of things" whose core and foundation is RFID Code Resolution Networks (CRN). As the code information provider, RFID CRN servers are located at different geographical position and network environment. The increasing size of network, complexity of the tree-like topological structure and the severe situation of network security make the secure running and daily management of the network much more difficult. Based on such situation, it becomes necessary to monitor the RFID CRN.Firstly, this thesis analyzes Simple Network Management Protocol (SNMP) in detail. Through the research of typical RFID CRN, this thesis designs the monitoring system for RFID CRN based on SNMP. According to the discussion of some crucial problems such as the confirmation of monitoring parameters, the selection and creation of Management Information Base (MIB) nodes and the event mechanism, this thesis designs the models of and extends the function of the whole system which make it suitable for RFID CRN.Secondly, Based on NET-SNMP, SNMP4J and JGRAPH this thesis realizes the key components of the monitoring system including performance monitoring, topology generation and the subscription and detection of Trap information. This system has clear structure as well as popularity, extensibility and portability, and can be easily integrated with other SNMP monitoring system. The Monitoring tool can generate real-time time-series chart, network topology diagram and data comparison chart, which are useful for network manager.At last, through the test of the resource cost and response time of monitoring agent as well as monitoring tool's function test, this thesis proves that the monitoring system can work reliably and monitor the network effectively. Under the reasonable monitoring load, the monitoring system will not influence the performance of the networks. |