A mobile agent based communication protocol to optimize message delivery cost

• This paper proposes a communication protocol for mobile agents.
• Proposed protocol selects optimal transmission cost path for message delivery.
• The protocol is generalized for all applications regardless of MRR.
• We used NS3 to simulate our communication protocol.

Agents are applicable to many computing areas like distributed systems because of their autonomy and mobility factors. Reliable and efficient communication between agents in a mobile agent system is a challenging task. To deliver the messages efficiently and reliably, mobile agent communication schemes generally have two phases. The first phase is agent tracking phase which deals with locating the agent current location. Second phase deals with the delivery of message to the agent at its current location. The existing approaches suffer from high communication overhead or problems such as message chase problem, triangle problem, or delayed delivery of messages. Approaches such as I-update deliver the message very efficiently without any overhead in message delivery phase but increase the memory and network overhead in agent tracking phase. In this paper, we present a novel approach that minimizes message delivery delay while maintaining a balance between the costs of agent tracking phase and message delivery phase. Our approach enhances the home agency scheme by maintaining track of agent migrations and costs of message delivery through different channels at the home agency. This approach allows the home agency to send the message to an agent via optimal transmission cost path, which minimizes message delivery delay and reduces the probability of occurrence of message chase problem.