Proceedings of the 2nd International Conference on ICT for Digital, Smart, and Sustainable Development, ICIDSSD 2020, 27-28 February 2020, Jamia Hamdard, New Delhi, India

Research Article

A Theoretical Framework of Smart System Modelling of Crop Based on GAN and IoT Platform

  • @INPROCEEDINGS{10.4108/eai.27-2-2020.2303294,
        author={Wasim Ahmad Ansari and Yang  Yuwang},
        title={A Theoretical Framework of Smart System Modelling of Crop Based on GAN and IoT Platform},
        proceedings={Proceedings of the 2nd International Conference on ICT for Digital, Smart, and Sustainable Development, ICIDSSD 2020, 27-28 February 2020, Jamia Hamdard, New Delhi, India},
        publisher={EAI},
        proceedings_a={ICIDSSD},
        year={2021},
        month={3},
        keywords={precision agriculture iot edge computing cloud computing automation gan},
        doi={10.4108/eai.27-2-2020.2303294}
    }
    
  • Wasim Ahmad Ansari
    Yang Yuwang
    Year: 2021
    A Theoretical Framework of Smart System Modelling of Crop Based on GAN and IoT Platform
    ICIDSSD
    EAI
    DOI: 10.4108/eai.27-2-2020.2303294
Wasim Ahmad Ansari1,*, Yang Yuwang1
  • 1: School of Computer Science and Engineering, Nanjing University of Science & Technology, China
*Contact email: dr.wasim.scholar@njust.edu.cn

Abstract

The advent of the communication and cutting edge technologies is playing a key role in reforming the various sectors and so as in the field of the agronomy specifically in precision agriculture (PAg). Greenhouses facing the problem of barren land basically need such systems potentially relevant to control re-circulation in the optimal way. In this paper the proposed system is segregated into three parts cyber physical system (CPS), edge computing and cloud computing-gan layers with use of IOT protocols like message queuing telemetry transport (MQTT) protocol or constrained application protocol (CoAP) for communication purpose [3]. This paper describes the theoretically concept of using GAN based on the two multilayer perceptron models generator and discriminator. GAN uses the analyzed data from the cloud part and then generator generates the data distribution (probability distribution). The discriminator discriminates the data as per the certain system requirement to enable more automation needed for crops growth. GAN part works on the min max concept in which discriminator is trained to maximize the level assignment probability whereas generator is getting trained to minimize the chances of mistakes done by discriminator to support more accurate result. The concept of the GAN will enable the system to enhance its intelligence in order to make more sharpen and accurate decision.