Chunyan Han

Research and Application on Automatic Network Topology Discovery in ITSM System

Automatic network topology discovery is important for network management and network analysis in ITSM system. Issues like network environment configuration, performance testing and fault detecting all require accurate information of the network topology. According to the requirement of the ITSM system, in this paper we propose an efficient algorithm that is based on SNMP for discovering Layer 3 and Layer 2 topology of the network. The proposed algorithm only requires SNMP to be enabled on routers and managed switches. For Layer 3 topology discovery, this paper proposes a shared subnet based approach to analyze connections between routers. In addition, the algorithm uses a STP based method to discover Layer 2 topology, which does not require the completeness of switch’s AFT information. The algorithm has been implemented and tested in several enterprise-level networks, and the results demonstrate that it discovers the network topology accurately and efficiently.

Logic basis for anticipatory reasoning-reacting systems with three-dimensional moving objects

To design and develop anticipatory reasoning-reacting systems with three-dimensional moving objects, it is indispensable to decide the fundamental logic basis to underlie anticipatory reasoning about three-dimensional moving objects. This paper investigates the fundamental logic basis for anticipatory reasoning-reacting systems with three-dimensional moving objects. The paper presents some basic requirements for the fundamental logic basis and shows that three-dimensional spatio-temporal relevant logic is more suitable than classical mathematical logic and its various classical conservative extensions and various relevant logics to be the fundamental logic basis by two case studies.

Research on Multi-agent Dynamic Epistemic Relevant Logic for Air Traffic Control System

Traditional interactive air traffic control systems can’t meet the needs of the increasing complexity of the air traffic network. In this paper, we propose a new generation of air traffic control system and establish a logic system called Three-dimensional multi-agent dynamic epistemic relevant logic (3DMDERL for short) to fulfill the needs. We also describe the detail about the application of the logic system that we have established in the air traffic control field. The logic system combining with the new generation air traffic control system prevents many failures caused by information exchanging errors and data sharing and helps to guarantee the safety and efficiency of air traffic.

Logic-Based Prediction Model for Air Traffic Control System

Anticipatory Reasoning-Reacting Systems can solve the accidents caused by human factors effectively, and reduce the casualties of the crash. Also it can make a difference in airport scheduling, airport flow control and weather forecast. This paper constructs an Air Traffic Control System prediction model which is based on qualitative reasoning and quantitative calculation. Because 3D-STDRL has many advantages in three-dimensional moving objects reasoning field, this prediction model adopts 3D-STDRL as the basis of reasoning. Meanwhile this prediction model takes some appropriate algorithms as reasoning theoretical basis. This paper describes each functional module of the model and uses a specific instance to describe how the model works. Finally it proves that this model can make effective and accurate predictions and get proper solutions.

Aircraft Rerouting Decision-Making Model under Severe Weather

Severe weather is an important factor resulting flight delays. Based on the Aircraft Flight Regulations, this paper established an aircraft rerouting decision-making model under severe weather with the consideration of severe weather, FFAs, self-situation of aircrafts and aircraft flight coordination. Besides, relevant logic-based reasoning is used in this model to reduce the complexity of the mathematical calculations. At the end of the paper, we take the routes from Dalian to Harbin as a sample experiment data. The experiment successfully verified the validity of aircraft rerouting decision-making model.

Research and Application of Heterogeneous Network Topology Discovery Algorithm Based on Multiple Spanning Tree Protocol

Heterogeneous network topology discovery is significant for network management and network. According to the requirement of the real network environment ,in this paper we analyze the current mainstream data-link level topology discovery algorithms, including spanning tree topology discovery algorithms and forwarding-table topology discovery algorithm, discussing the advantages and disadvantages of each method and come up with a heterogeneous network topology discovery process based on multiple spanning Tree protocol including STP / RSTP, PVST +, MISTP / MSTP protocols, The algorithm has been implemented and tested in several enterprise-level networks, and the results demonstrate that it discovers the physical topology information, VLAN information , spanning tree logical topology information relatively accurate.

An Approach for IPSecVPN Network Topology Discovery Based on Layer 3 Topology

Network topology discovery is crucial to auto network management system. In research area, the reseach of Layer 3 topology discovery uses ipRouteTable [1,2], which efficiency is slow. This paper proposes a improved approach to Layer 3 discovery algorithm. Then, this paper gives the time complexity of algorithm analysis. And this paper proposes a approach to discover IPSecVPN protocol topology based on Layer 3 topology. In addition, the architecture of topology discovery module proposed is helpful to further topology discovery. Those alogrithms relied on SNMP MIB information have been implemented in Java and tested in several enterprises network. The discovering results demonstrate that it is accurate and efficient to discover Layer 3 topology and IPSecVPN protocol topology.

Research and Application of Multiple Spanning Tree Network Topology Discovery Algorithm

Network topology information was vital for network management. According to the requirement of the real network environment ,in this paper we studied the current mainstream level two and level three topology discovery algorithms, including simple spanning tree topology discovery algorithms and forwarding-table topology discovery algorithm, discussing the advantages and disadvantages of each method and come up with a network topology discovery method suited for multiple spanning Tree protocol topology, protocols including STP / RSTP, PVST +, MISTP / MSTP protocols, The algorithm has been tested in several enterprise-level environments, and the results showed that it discovered the physical connection information, VLAN information, spanning tree logical connection information considerably accurate.