Agents Based e-Commerce and Securing Exchanged Information

Abstract

Mobile agents have been implemented in e-Commerce to search and filterinformation of interest from electronic markets. When the information isvery sensitive and critical, it is important to develop a novel securityprotocol that can efficiently protect the information from malicioustampering as well as unauthorized disclosure or at least detect anymalicious act of intruders. In this chapter, we describe robust securitytechniques that ensure a sound security of information gatheredthroughout agent's itinerary against various security attacks, as wellas truncation attacks. A sound security protocol is described, whichimplements the various security techniques that would jointly prevent orat least detect any malicious act of intruders. We reason about thesoundness of the protocol using Symbolic Trace Analyzer (STA), a formalverification tool that is based on symbolic techniques. We analyze theprotocol in key configurations and show that it is free of flaws. Wealso show that the protocol fulfils the various security requirements ofexchanged information in MAS, including data-integrity,data-confidentiality, data-authenticity, origin confidentiality and datanon-repudiability.