The latest researches on access control model are dramatically different from conventional ones. Nowadays, most attention is paid to accessing across organizational boundaries. So, how to identify the applicant and determine authorization with limited information; how to express and exchange control rules expediently; how to protect confidential information and enhance collaboration simultaneously, are the most concerned problems. However, for large commercial organizations, a fine management of internal functions is of the same importance as external service management. It is very troublesome to control authorizations merely with attributes and composition of policies introduced from attribute-based access control (ABAC). So, we introduce a united access control model for systems in collaborative commerce, combining the advantages of conventional role-based access control (RBAC), task-based authentication control (TBAC) and that of recent ABAC and automated trust negotiation (ATN). Innovational ideas in the model are analyzed, and the implement architecture is discussed. The paper concludes with a summary of the united model’s benefits and future work.

As the capability of an individual Web service is limited, it's necessary to create new functionalities with existing Web services. Web services composition is the ability to create a new value-added service by incorporating some existing web services together. A model based colored Petri net (CPN) to provide semantic support for web service composition is proposed. The basic composite constructs in the model are sequence, concurrent, choice and loop. A closed composing algebra is defined to obtain a framework which enables declarative composition of web services. Finally modeling composite processes of Web services based on CPN is applied to a case of naval vessel command and control system.

Abstract-After analyzing the characteristics of Immunity Intrusion Detection System, by utilizing prominent characteristics of genetic algorithm and vaccine mechanism, a new hybird immunity intrusion detection model based on genetic algorithm and vaccine mechanism was established. The modeling process is described in detail, such as feature extraction of vaccine, genetic operates to memory detectors and the improvement for detection method. Via application vaccine mechanism into intrusion detection system, the new model has the function of misuse detection and anomaly detection simultaneously. In order to improve the detection matching efficiency, we also present a novel matching algorithm RBNDM. Finally, we evaluated our model using the KDD Cup 1999 Data set. The experiments show that this model can increase the true positive rate of the IDS.

No abstract

After analysis the characteristics of AIS-based Intrusion Detection System, a new AIS-based intrusion detection model based improved Genetic Algorithm is established. By utilizing prominent characteristics of genetic algorithm, such as automatic optimizing, global researching, and adaptability, the new model uses genetic operator to improve the candidate detectors generating algorithm and reduce detectors redundancy. The detectors generated by new model have good fitness and better detection ability. Experiments show that this model can effectively increase the true positive rate of the IDS.