Hnin Yu Shwe

Cluster-parent based RPL for Low-Power and Lossy Networks in building environment

Low-Power and Lossy networks (LLNs) cover a wide scope of applications, including, but not limit to, industrial control, healthcare and building automation, which has proliferated into both industrial and academic areas in recent years. Routing plays a significant role to the overall architecture of LLNs. The IPv6 Routing Protocol for Low Power and Lossy Networks (RPL) is standardized by IETF to provide interoperability for LLNs. Traditional routing protocols deliver packets through single relay and cannot cope well with the unreliable wireless medium. In this paper, we propose a cluster-parent based RPL protocol (CRPL), which is an opportunistic coordination forwarding scheme. In CRPL, through a top-down approach, an optimization problem is formulated. And the optimal cluster-parent set is selected to minimize the end-to-end cost for individual node. Priority-based scheduling mechanism is applied to coordinate the cluster-parent set for packet forwarding. We compare CRPL with RPL, and another recent work, ORPL on NS3 platform. The performance metrics in terms of packet delivery ratio, number of retransmission and average path cost are measured under various scenarios. Simulation results show that CRPL outperforms RPL and ORPL with significant improvement on the network reliability and reduction of packet retransmission.

Building Efficient Multi-level Wireless Sensor Networks with Clustering

Wireless sensor networks will be responsible for the majority of the growth in smart building systems over the next decade. In resource constrained wireless sensor networks, it is very important to design the protocols with energy efficiency to prolong the lifetime of the sensor networks. Node clustering and data aggregation become popular since cluster-based sensor network can enhance the whole network throughput by aggregating the collected sensory information in each cluster. In such a network, the cluster head nodes play an important role in forwarding data originated from other common nodes to the sink. As a consequence, the cluster head nodes will have the problem of quick energy depletion upon multiple packets forwarding in high data load sensor networks. In this paper, we proposed a simple cluster-based linear network coding protocol in which random linear network coding is applied at cluster head nodes in order to minimize the number of forward packets to the sink. Simulation results are provided to show the efficacy of the proposed method in terms of the throughput and end-to-end delay.

Interference-aware resource allocation for device-to-device communications in cellular networks

Device-to-device (D2D) communications was proposed underlying cellular networks to increase system capacity to support high-data-rate multimedia services. However, D2D communications unavoidably bring along co-channel interference to conventional cellular users. In this paper, we propose interference-aware resource allocation (IARA) scheme for cellular networks with underlying D2D communications. The proposed IARA scheme is based on two-way channel gain estimations and interference level measurements. Related works are revisited and the proposed IARA scheme is discussed. Its performance is investigated through extensive computer simulations. IARA scheme outperforms random allocation (RA) scheme. We further propose to use adaptive transmission power for D2D users to protect cellular users from severe co-channel interference. Simulation results show that this allows for longer link distance for D2D users, leading to better resources reuse.

Cluster-Based WSN Routing Protocol for Smart Buildings

The design issues of resource constrained sensor network include the use of efficient routing protocol which have significant impact on the energy dissipation of the network. Since the energy saving plays the main challenge in the design and implementation of wireless sensor networks, we proposed the energy efficient routing protocol which is an integration of clustering and routing in sensor network. In our approach, we also performed network coding during the data routing in order to achieve the additional power saving in the cluster head nodes. Simulation results are provided to show the efficacy of the proposed method in terms of the throughput and end-to-end delay. In addition, the results show significant improvement of proposed method in the network lifetime over other technique.

Cubic-based 3-D localization for wireless sensor networks

Localization using sensor network has attracted much attention for its comparable low-cost and potential use with monitoring and targeting purposes in real and hostile application scenarios. Currently, there are many available approaches to locate persons/things based of GPS and RFID technologies. However, in some application scenario, e.g., disaster rescue application, such localization devices may be damaged and may not provide the location information of the survivors. The main goal of this paper is to design and develop a robust localization technique for human existence detection in case of disasters such as earthquake or fire. In this paper, we propose a 3-D localization technique based on the hop-count data collected from sensor anchors to estimate the location of the activated sensor mote in 3-D coordination. Our algorithm incorporates two salient features, grid-based output and event-triggering mechanism, to guarantee both improve accuracy and power efficiency. Simulation results indicate that the proposed algorithm can improve the localization precision of the human existence and work well in real environment.

Scalable Cloud-Based Data Storage Platform for Smart Grid

Real-time conditional monitoring and instinctive fault diagnosis plays an important role in the smart grid systems. Conventional data storage and management systems are facing the difficulties with the disseminated, heterogeneous and large volume of data. Conventional data management infrastructure utilizes centralized approach which comprises of large-scale server, disk array data storage hardware and relational database management system (RDBMS) for database services. This type of infrastructure may result in poor performance to support the requirement of smart grid applications for very large database with high data request rates at very low latency. Precise, fast, vulnerable and cooperative information system is the basis requirement for the smart grid applications. In this paper, we proposed a distributed data storage and management platform which guarantees high scalability and reliability data storage and management with the huge amount of information in the smart grid application. In order to evaluate and validate our scalable cloud-based data storage platform for smart grid, we deployed our proposed infrastructure in our cyber-physical test-bed which is a facility to test innovative technologies for building efficiency and urban sustainability.

Scalable Distributed Cloud Data Storage Service for Internet of Things

In the era of Internet of Things (IoT), a number of things are efficiently interconnected in order to combine, integrate the physical world to the information space. One of the application domains of urban IoT is smart grid where smart buildings, smart homes, vehicles, electrical grids, power plants, other components are interconnected to provide the services in an efficient way. In the smart grid systems, the data of conditional monitoring are increasing rapidly. Facing with these massive, distributed, heterogeneous, complex state data, conventional data storage, management will encounter great difficulties. Conventional infrastructure uses centralized approach, expensive large-scale server, disk array storage hardware, relational database management system, which leads to poor system scalability, higher costs,, essentially, is difficult to adapt to the requirement of higher reliability on real-time state data of the smart grid applications. Accurate, fast, open, shared information system is the basis for the smart grid applications. In this paper, we proposed a scalable distributed cloud data storage, management platform which provides reliable data storage, management of the vast amount of information in IoT applications. We have evaluated, validated our approach by applying our cloud data storage, management platform in our cyber-physical test-bed, a facility which is intended to test innovative technologies for urban sustainability.

An IoT-oriented data storage framework in smart city applications

Smart City applications are addressed to city authorities, infrastructure and utilities managers, and city stakeholders wishing to use smart city solutions in order to increase the competitiveness, cohesion, and sustainability of the city. Facing with these massive, distributed, heterogeneous and complex state data, conventional data storage and management will encounter great difficulties. Conventional infrastructure uses centralized approach, expensive large-scale server, disk array storage hardware and relational database management system; which leads to poor system scalability, higher costs, and essentially, is difficult to adapt to the requirement of higher reliability on real-time state data of the smart city applications. Accurate, fast, open, shared information system is the basis for the smart city applications. In this paper, we proposed a scalable distributed cloud data storage and management platform which provides reliable data storage and management of the vast amount of information in the smart city.

Condition monitoring system of Ubin Island micro-grid

Singapore has embarked a project to build a remote monitoring system for an existing micro-grid located at Ubin Island. This paper introduces the design of the monitoring system. In this monitoring system, the module level of PV and cell level of battery conditions will be monitored. All the measurements within the micro-grid will be synchronized by using GPS modules for better understanding the micro-grid. And the resolution of the measurements can be as high as 10 samples per cycle. Because of the micro-grid is an isolated system, a remote monitoring operation center will be built for the users to monitor the micro-grid.

WSN-based energy-efficient data communication protocol for smart green building environment

One of the important design issues of smart green buildings is to make use of energy, water and other resources efficiently. Improving the energy efficiency of buildings is an important step towards a more sustainable lifestyle, leading to significant cost savings for the residents. Wireless sensor network will be responsible for the majority of the growth in smart green building systems over the next decade since it provides the necessary information to the smart green building system, allowing it to monitor and control the physical environment of the building. Since the energy saving plays the main challenge in the design and implementation of wireless sensor networks, we proposed an energy efficient data communication protocol for smart green building environment which is an integration of clustering and routing in sensor network. In our approach, we also performed network coding during the data routing in order to achieve the additional power saving in the cluster head nodes. Simulation results are provided to show the efficacy of the proposed method in terms of the throughput and end-to-end delay. In addition, the results show significant improvement of proposed method in the network lifetime over other technique.