China
Africa
Explore chapters and articles related to this topic
Agents, Objects, and Frames
Published in Adrian A. Hopgood, Intelligent Systems for Engineers and Scientists, 2021
There is no reason why an agent should remain permanently on a single computer. If a computer is connected to a network, a mobile agent can travel to remote computers to carry out its designated task before returning back home with the task completed. A typical task for a mobile agent might be to determine a person’s travel plan. This will require information about train and airline timetables, together with hotel availability. Instead of transferring large quantities of data across the network from the train companies, airlines, and hotels, it is more efficient for the agent to transfer itself to these remote sites, find the information it needs, and return. Clearly, there is potential for malicious use of mobile agents, so security is a prime consideration for their viability.
Securing IoT with Blockchain
Published in Vijayalakshmi Saravanan, Alagan Anpalagan, T. Poongodi, Firoz Khan, Securing IoT and Big Data, 2020
An IoT battery-powered device may be integrated directly with a blockchain client. This allows blockchain features to be embedded in IoT devices themselves for direct interaction between them. A multilevel blockchain system (MBS) is proposed by Mbarek et al. [36] to secure an IoT that uses mobile agents to enforce the flexibility and speed of transactions in the blockchain. Mobile agents roam throughout the network of IoT devices to aggregate useful data and generate hashed blocks of data, reducing time delays and solving other issues like scalability and synchronization. MBS consists of three hierarchical levels through which IoT devices can send their data securely: micro-level consisting of IoT devices, meso-level consisting of cluster heads, and macro-level consisting of the blockchain platform. The MBS platform is made up of four entities: the IoT device (collects and transmits data), ordering service (accepts transactions and creates blocks), endorsing peers (checks validity of smart contracts), and committing peers (runs validation). It includes meso, macro, and micro agents with different roles and locations in the architecture. Simulation is done using Hyperledger Fabric with 1,000 nodes and the end results are satisfactory in terms of energy consumption and response time.
Distributed Artificial Intelligence and Agents
Published in Weiming Shen, Douglas H. Norrie, Jean-Paul A. Barthès, Multi-Agent Systems for Concurrent Intelligent Design and Manufacturing, 2019
Weiming Shen, Douglas H. Norrie, Jean-Paul A. Barthes
Mobile agents are software agents capable of moving from one machine to another automatically. One advantage compared to a static agent residing on a particular machine is that this can decrease the network communication load. Indeed, if we consider for example an agent involved in E-commerce trying to buy videos on specific topics, then the agent will first have to select which sites to visit, then for each interesting site it will request samples of videos in order to select those of interest, and after having repeated the process for a number of sites, it will be able to place an order. The whole process will generate considerable traffic on the network. An alternative approach consists in sending the agent from the index site (a yellow pages service) to each provider in turn and doing the selection process locally. If the agent is small enough this will generate much less traffic on the net. The feasibility of this approach requires that (1) the agent be small enough to be shipped efficiently; and (2) the agent be able to execute on various platforms. Several approaches have been developed allowing code to migrate from machine to machine, to accomplish this.
Fast data acquisition algorithm for remote monitoring system of smart home
Published in International Journal of Computers and Applications, 2022
The application of mobile agent technology in the management of the Internet of things has solved the above problems. Mobile agent is a program that can migrate autonomously from one host to another in a heterogeneous network and interact with other agent or resources. Mobile agent technology is a product of the combination of distributed technology and agent technology. It has the characteristics of reducing network load, supporting platform independence, self-executing, and expanding easily. So when the first domain using mobile agent to realize intra-domain data acquisition, it does not have to send a large number of variables back to the first domain through the SNMP for computing, but instead of transferring the acquisition work to the mobile agent at the device node, and the first domain just needs to receive the returned calculation result, execute the related operation. Therefore, the technology can improve the performance of data acquisition to a great extent.
Study of order lifecycle tracking systems in building material equipment manufacturing enterprises
Published in Journal of Industrial and Production Engineering, 2018
Zhao Peng, Shunsheng Guo, Lei Wang, Jun Guo, Xixing Li
Mobile agent technology can be used between the internal and external information systems of the enterprise. Based on the basic principle of a mobile agent, the mobile agent will migrate to different terminals to carry out tasks. These terminals include types of local computers, and the mobile agent will handle different node information on the local computer. Finally, the information will be sent to the host. The problem of data acquisition in the order tracking system can be solved using mobile agent technology.
An efficient cloud prognostic approach for aircraft engines fleet trending
Published in International Journal of Computers and Applications, 2020
Zohra Bouzidi, Labib Sadek Terrissa, Noureddine Zerhouni, Soheyb Ayad
To evaluate the efficiency of the presented mobile agent paradigm, six induction motors with different failure modes in a motor-tested system are used to imitate distributed manufacturing processes. Mobile agents distribute signal-processing algorithms (e.g. feature extraction) to cloud nodes instead of transmitting raw detection measurements to the central server, and can significantly reduce the traffic load on the network.