AI-Powered Internet of Things 

Courses included in the program:

Project: Smart Cities: Digital Transformation of Cities

Duration: 01.20122 - 02.20223

Funding agency:  IEEE

Project Coordinators (PIs): Dr. Madusudan Singh, WSU, Korea and Dr. Dhananjay Singh, HUFS, South Korea  

Description: The project has successfully explored and addressed the challenges and opportunities presented by Industry 4.0 through the integration of Internet of Things (IoT) and Machine Learning (ML) technologies in the modeling of a smart factory. By combining sensor-based data collection with intelligent data-driven decision-making systems, the project developed a comprehensive framework for enhancing industrial automation, predictive maintenance, real-time monitoring, and production optimization.

 Throughout the project, advanced digital tools, including microelectronics, AI algorithms, and embedded IoT devices, were applied to model key processes in a smart factory environment. Machine learning techniques were used to analyze large volumes of operational data, enabling improved forecasting, fault detection, and adaptive planning mechanisms.

The project outcomes demonstrated significant improvements in productivity, energy efficiency, and resource utilization. Additionally, new approaches for optimizing workflows, reducing downtime, and supporting data-informed decision-making were proposed and validated. By contributing novel methods and practical solutions, this project has laid the groundwork for future applications of Industry 4.0 technologies in industrial settings, helping pave the way toward more intelligent, adaptive, and sustainable manufacturing systems.


High-level abstraction of the processing flow when requesting a task in the Elastic-RAN model: (1) the user performs a request that is captured by an antenna; (2)this antenna passes this request to the Pool Orchestrator responsible for a particular geographic region; (3) later, this orchestrator dispatches the incoming task to one of its BBU Pools (4) for the appropriate processing; (5) in parallel, Elasticity Manager performs periodic monitoring of different metrics for resource reorganization. 

Project: Using Internet of Things and Machine Learning in Modeling a Smart Factory in the Context of Industry 4.0

Duration: 04.2019 - 04.2022

Funding agency:  CAPES , Brazil 

Project Coordinators (PIs): Dr. Rodrigo Righi, UNISINOS, Brazil and Dr. Dhananjay Singh, HUFS, South Korea    

Description: The project has successfully explored and addressed the challenges and opportunities presented by Industry 4.0 through the integration of Internet of Things (IoT) and Machine Learning (ML) technologies in the modeling of a smart factory. By combining sensor-based data collection with intelligent data-driven decision-making systems, the project developed a comprehensive framework for enhancing industrial automation, predictive maintenance, real-time monitoring, and production optimization.

 Throughout the project, advanced digital tools, including microelectronics, AI algorithms, and embedded IoT devices, were applied to model key processes in a smart factory environment. Machine learning techniques were used to analyze large volumes of operational data, enabling improved forecasting, fault detection, and adaptive planning mechanisms.

The project outcomes demonstrated significant improvements in productivity, energy efficiency, and resource utilization. Additionally, new approaches for optimizing workflows, reducing downtime, and supporting data-informed decision-making were proposed and validated. By contributing novel methods and practical solutions, this project has laid the groundwork for future applications of Industry 4.0 technologies in industrial settings, helping pave the way toward more intelligent, adaptive, and sustainable manufacturing systems.


Project: Semantic edge computing and IoT architecture

Duration: 03.2012 - 02.2020

Funding agency: EU Horizon 2020.

Project Coordinators (PIs): Dr. Dhananjay Singh, HUFS, South Korea  

Description: In this project, we have designed a Semantic Fusion Model (SFM) that involves the use of edge devices in Internet of Things (IoT) networks to perform local data processing and analysis, rather than relying on centralized servers or cloud systems. The architecture integrates SFM to process and integrate information from sensors in IoT networks. The smart embedded system in this architecture uses semantic logic and value-based information to enhance its intelligence. Additionally, this project discusses the various applications, services, and visual aspects of IoT using technologies such as Radio Frequency Identification (RFID), 6lowpan, and sensor networks, as well as the challenges that need to be addressed in the implementation of IoT.


Publications: 

https://doi.org/10.3390/s22020640 [SCIE: 3.57]

Project: IP-Based Wireless Sensor Networks: 6LoWPAN Tealtime Testbed 

Duration: 03.2007 - 02.2017

Funding Support: Brain Korea-21, DSU, NIMS, HUFS, MtoV Inc. 

Project Coordinators (PIs): Dr. Dhananjay Singh, HUFS, South Korea  

Description: In this project, we have developed a framework for IPv6-Low Power Wireless Personal Area Networks (6LoWPAN) that utilizes short adaptation identifiers (AIDs) in place of full IPv6 addresses to enable effective IPv6 header compression in communication between IEEE 802.15.4 nodes and the IPv6 domain. To facilitate this, we have implemented a mechanism for translating AIDs to IPv6 addresses and maintaining an AID-IPv6 translation table at the gateway and In-node. When a packet is transmitted, it carries an AID value in the adaptation header instead of the OUT-node's IPv6 address, which is then translated back to IPv6 at the gateway using the AID-IPv6 translation table. In addition, we have designed an effective frame format for the adaptation layer to support both global and local communication in this context. As a reminder, the primary goal of 6LoWPAN is to provide internet connectivity to low-power networks using the IEEE 802.15.4 standard. This allows IN-nodes within the network to communicate with OUT-nodes in the IPv6 domain. 


Publications: 

Project: Ubiquitous IT: Wireless Sensor Networks with Simulation and Testbed 

Duration: 03.2008 - 02.2018

Industry: Brain Korea-21, DSU, NIMS, HUFS, MtoV Inc. 

Project Coordinators (PI): Dr. Dhananjay Singh, HUFS, South Korea  

Description: In this project, we have developed a wireless ad-hoc sensor network that enables portable devices to establish communication without the need for a central infrastructure. However, the lack of a central infrastructure and the ability of devices to move randomly introduces challenges such as routing and security. To address these issues, we have examined several ad-hoc routing protocols, including AODV, DSR, DSDV, OLSR, and ZRP, which propose solutions for routing within a fixed ad-hoc sensor network. However, we are also interested in communication between a wireless device in the ad-hoc network and a fixed device in a fixed network, such as the Internet. Therefore, we have modified the ad-hoc routing protocol AODV to support this type of interconnection between the wireless ad-hoc sensor network and the fixed network. 


Publications: 

https://doi.org/10.3390/s19183835 [SCIE-3.57]