EAI - Reinventing the governance and economy of research and innovation
The leading professional community for research career enhancement in the digital age


Dr SeungHwan Won

‎Malaysia Campus, University of Southampton

“Intelligent Precision Agricultural Platform Employing Internet of Thing”


With the aid of Internet of Thing (IoT) and big data based processing, agriculture becomes a very data-centric area. The use of private information gained from various sensors’ based devices as well as public information such as weather data will provide high potential for increasing agricultural productivity. Proactive adoption and in-depth analysis of Machine Learning (ML) algorithms in disruptive manners may lead to gigantic impact on the advent of Intelligent Precision Agricultural Platform (IPAP). Therefore, both precise analysis of collected data and prediction capability at the IPAP will play a pivotal role in enhancing agricultural productivity. Four core components for the IPAP consists of (1) Data sensing, (2) Reliable & wide-coverage communication, (3) Data storage, and (4) Post-data processing, described in Figure 1.

Figure 1: Example of intelligent IoT system facilitating agricultural productivity and performing environmental monitoring

Two unlicensed frequency spectrums, namely 433 and 915 MHz are allocated for LoRa technology. For our project, only 915 MHz spectrum band will be considered owing to its maximum allowable transmit power, 500 mW, which can guarantee wider coverage. Under the situation, its maximum allowable cell radius in rural area may be in-between 10 and 15 km. A remote computer can control the actions of the end-devices and their gateway. Furthermore, it post-processes data collected from them and finds hidden features. Based on finally generated information, farmers and agronomists use both web-based monitoring software operating at the remote computer and mobile application software operating at smartphone to find a better way of resolving critical issues at proper moment, to predict potential future risks, and to minimise detrimental impact on entire farm management, which will be key factors of our proposed IPAP to be deployed in realistic situations.

Even though the post-data processing also encompasses the conventional statistical analyse based on data measured, we would delve deeper into how ML approaches will be adopted in our proposed system. With the development of IoT infrastructure, collecting massive amounts of specific information becomes realistic for the commercial purpose. How to process big data will make a critical difference in many applications. Big data analytics being line with ML techniques leads to the highest interests in many research topics. Among diverse ML approaches, Deep Learning (DL) technique has manifested its unique and powerful impact on the advancement of many engineering issues. A key capability of the DL aided approach is the analysis and learning of massive unsupervised data sets, making it an invaluable tool for big data analytics processing mainly uncategorised and unlabelled data. The exploitation of the trendy DL in our IPAP will provide a better solution to characterise hidden behaviours of data collected from real agricultural fields. It enables to make automated and even intelligent agricultural management real and to minimise any potential risks. Accordingly, the proposed plan will become the keystone for establishing a sustainable and reliable infrastructure for intelligent precision agricultural sector, which will contribute a significant impact in enhancing productivity of diverse agricultural areas.

In conclusion, during the keynote speech, we will introduce the above-mentioned to audience step-by-step and also exemplify the targeted system being developed both in Philippines and in Malaysia.


SeungHwan Won is an Associate Professor at USMC, and teaches both programming and advanced programming as well as mathematics, random process, and wireless communications.

He graduated from the School of Engineering, Korea University in 1999 (BS) and 2001 (MSc).

He completed his PhD in the area of Wireless Communications at the University of Southampton in 2008, under the supervision of Prof. Lajos Hanzo.

The title of his thesis was “Initial Synchronisation in the Multiple-Input Multiple-Output Aided Single- and Multi-Carrier DS-CDMA as well as DS-UWB Downlink”.

SeungHwan has broad industrial engineering experience, with employment both at LG Electronics R&D (2001-2004) and Samsung, S. Korea (2008-2013) where he held the position of Senior Engineer in the Modem Team, Digital Media & Communications Business.

He has published 14 journal papers and is associated with 21 US patents.

EAI Institutional Members