Sung-Kee Noh

Energy-efficient predictive tracking for continuous objects in wireless sensor networks

Due to the severe resource constraints of sensor hardware, energy efficiency is one of critical factors for monitoring the movement of the large-scale phenomena such as wild fire and hazardous bio-chemical material, denoted by continuous objects. In order to save energy, most of existing research on tracking the continuous objects focuses on finding the ways to minimize the communication cost through the effective data delivery such as data aggregation and reducing the number of reporting nodes, and not much work has been done on sensor state scheduling. Energy efficiency is expected to improve if only sensor nodes near the boundary of continuous object actively participate in tracking process, while other sensor nodes stay on sleep state for energy saving. In this paper, we propose a predictive continuous object tracking scheme, called PRECO, which uses minimum set of active sensing nodes to reduce energy consumption. The proposed scheme predicts the future boundary line, which provides the knowledge for a wake-up mechanism to decide which sleeping nodes need to be activated for future tracking. The proposed algorithm is verified with simulation results that total energy consumption can be dramatically reduced under acceptable boundary detection accuracy.

A call admission control scheme for heterogeneous service considering fairness in wireless networks

Guaranteeing a handoff dropping probability below a predetermined threshold is essential for wireless mobile networks. Previous researches have proposed admission control policies for integrated voice/data traffic in wireless mobile networks. However, since QoS has been considered only in terms of CDP (call dropping probability), the result has been a serious CBP (Call Blocking Probability) unfairness problem between voice and data traffic. In this paper, we suggest a new admission control policy that treats integrated voice and data traffic fairly while maintaining the CDP constraint. For underprivileged data traffic, which requires more bandwidth units than voice traffic, the packet is placed in a queue when there are no available resources in the base station, instead of being immediately rejected. Furthermore, we have adapted the biased coin method concept to adjust unfairness in terms of CBP. We analyzed the system model of a cell using both a two-dimensional continuous-time Markov chain and the Gauss-Seidel method. Numerical results demonstrate that our CAC (call admission control) scheme successfully achieves CBP fairness for voice and data traffic.

Region Based Data Dissemination Scheme for Mobile Sink Groups in Wireless Sensor Networks

In many applications of Wireless Sensor Networks (WSNs), the sensing data are disseminated from a source to multiple mobile sinks. Since WSNs consists of a number of sensor nodes with limited capabilities, previous studies mainly discuss on how to send the data efficiently and do not consider the group mobility of mobile sinks that move together staying closely and randomly move within a geographically restricted region. Although the existing multicasting protocols could be applied, they suffer from high congestion and control overhead due to location updates by individual mobile sinks. Geocasting protocols are effective for data delivery to a sink group within a restricted region, but do not guarantee since they only focus on transmitting data to all nodes within a stationary region. Therefore, we propose Region Based Data Dissemination (RBDD) scheme to address these problems. RBDD provides efficient data dissemination scheme for mobile sink groups, so it guarantees data transmission when a sink group does not only move as a whole, but its member sinks also move inside of the region or outside of it. Simulation results show that RBDD guarantees data delivery to a mobile sink group.

Inter-Domain Roaming Mechanism Transparent to IPv6-Node among PMIPv6 Networks

Proxy Mobile IPv6 (PMIPv6) is designed to provide mobility service to IPv6-nodes in a network domain by network-based management which does not require the nodes to be involved in IP mobility management. In other words, the PMIPv6 can support roaming within a PMIPv6 domain, i.e. intra-domain roaming, transparent to mobile nodes. Recently, next generation wireless networks, such as 802.16e and Super 3G/3.9G, adopt the PMIPv6 as a management solution of the IP mobility. Such extensive adoption of the PMIPv6 might lead to frequent roaming of IPv6-nodes not only within a PMIPv6 domain but also between PMIPv6 domains. However, for the roaming between PMIPv6 domain, i.e. inter-domain roaming, existing solutions take interworking between PMIPv6 and MIPv6 into consideration. It means that every mobile node should involve IP mobility management although the nodes merely move around among PMIPv6 domains which provide the transparent node mobility service. Hence, we propose a network-based roaming mechanism between PMIPv6 domains to offer transparent mobility support to IPv6-nodes. The inter-domain mechanism takes into account concatenated tunnels between home domain and visited domain. The tunnels would be established by signaling among routers in the home and visited PMIPv6 network domains.

Large-Scale Phenomena Monitoring Scheme in Wireless Sensor Networks

This paper proposes a novel detection and tracking scheme for large-scale phenomena, called continuous objects, in wireless sensor networks, such as wild fire and bio-chemical material. Unlike previous works, the scheme takes into account a two-tier grid structure in order to achieve flexible and reliable detection and tracking. Firstly, for flexibility, the scheme constructs a coarse-grained grid structure and then, once a continuous object appears, fine-grained grid structures are established within coarse-grained grid cells only around the continuous object. The fine-grained grid structures could be constructed with dynamic grid cell size according to various requirements. Secondly, for reliability, minute grid cells of the fine-grained grid structures are able to provide detailed shape of boundary of the continuous object; to quickly deal with diffusion of the continuous object the scheme prepares the fine-grained grid structures into next coarse-grained grid cells toward diffusion direction of the continuous object.

A Novel Continuous Object Tracking Scheme for Energy-Constrained Wireless Sensor Networks

In order to monitor continuously moving phenomena such as wile fire and hazardous bio-chemical material in wireless sensor network, boundary tracking approach has been widely used by reason of its huge scale and extensive diffusion property. With the boundary tracking scheme, the energy efficiency is expected to improve if only sensor nodes near the boundary of continuous object actively participate in tracking process, while other sensor nodes stay in sleep mode for energy saving. In this paper, we propose a predictive continuous object tracking scheme, which uses minimum set of active sensing nodes to reduce energy consumption. The proposed scheme predicts the future boundary line, which provides the knowledge for a wake-up mechanism to decide which sleeping nodes need to be activated for future tracking. The proposed algorithm is verified with simulation results that total energy consumption can be dramatically reduced under acceptable boundary detection accuracy.

New call admission control mechanisms considering adaptive modulation control

Adaptive modulation control (AMC) has been proposed as the next generation modulation method for increasing network performance in cellular networks. Adaptive modulation is a powerful technique to improve the spectral efficiency in wireless transmission over fading channels. Similarly, it is possible to apply the adaptive modulation technique to call admission control (CAC) scheme in order to enhance network performance and satisfy quality of service (QoS) requirements. In this paper, we investigate a novel call admission control (CAC) mechanism in the cellular networks using AMC. First, we build the system model in which takes into account to both CAC and AMC. Second, we verify that the CAC model can offer better performance by using adaptive modulation technique. Finally we prove our claim by numerical analysis

Communication Strategy to Reflect Group Mobility Issue of Mobile Sinks in Wireless Sensor Networks

In wireless sensor networks, many studies on data dissemination to individual mobile sinks traditionally rely on a strategy that consists of a virtual infrastructure to serve current location of the mobile sinks and a per-sink foot-print chaining mechanism to support local mobility after location update. To adapt the strategy for supporting mobile sink groups, it might be simply considered to exploit a representative per a mobile sink group in order to representatively register location of the mobile sink group and representatively make the foot-print chain of the sink group. However, the representative manner may lead to considerable problems with respect to efficiency and robustness of data dissemination: inefficient data collection problem, needless location update problem, and relay path loss problem. These problems could be caused from dependency between the group and the representative, i.e. a member sink. Therefore, we propose a novel strategy for data dissemination decoupled with any member sink of a mobile sink group. In order to independently deal with a mobile sink group, the novel strategy is composed of three major mechanisms: 1) representative location update, 2) distributed data collection, and 3) per-group foot-print chaining.

Reliable continuous objects detection algorithm in wireless sensor networks

In wireless sensor networks, reliable event detection is one of the most important research issues. For the reliable event detection, existing studies usually assume that events are individual objects such as tanks, people, and animals. The existing studies estimate detection reliability of an individual object by the ratio of the number of successfully received data packets at a sink to the number of data packets determined by applications for reliable detection. Recently, detection of the other one kind of events, called the continuous object, such as wild fire and bio-chemical material has been receiving increased attention. Unlike the individual objects, a continuous object might cover a wide area and it could dynamically alter its own shape according to physical environments, e.g. wind, geographical features, and so on. Hence, the continuous object detection reliability can not be estimated by the indicator for the individual objects. In this paper we newly define the reliability indicator for continuous object detection. Then, we propose error recovery and revision scheme based on the estimation result from the novel indicator. Simulation results prove that the proposed scheme provides high reliability with respect to continuous object detection.

Effective retransmission scheme for supporting Communication Reliability in sensor networks

Wireless sensor networks may be utilized for various application areas such as health, military, home, and so on. In the wireless sensor networks, data transmission typically relies on multi-hop communication with high packet loss rates on wireless links so that reliable end-to-end data transmission is desirable. Recently, in order to achieve desired Communication Reliability (CR) levels of the various sensor network applications, a flexible loss recovery mechanism, called Active Caching (AC) is proposed. With respect to a desired CR of an application, when the packet delivery rate during multi-hop transmission from a source to an intermediate node decreases below the CR, AC retransmits lost packets from the source to the intermediate node so that the intermediate node has all data packets just like the source node. Then, during the data transmission from the intermediate node to the destination, if the packet delivery rate at another intermediate node falls below the CR, the new intermediate node recovers the lost packets from the previous intermediate node, not from the source node. Namely, AC flexibly caches all data packets at some intermediate nodes on the routing path in order to achieve the CR. However, AC does not consider the loss rate of each wireless link on the multi-hop routing path in terms of the loss recovery. Since the lost packets are requested and retransmitted via the multi-hop routing path, AC may cause a lot of retransmissions of the lost packets and thus increasing the number of totally transmitted packets. Therefore, we propose a strategy to minimize the number of totally transmitted packets when all data packets from a source are delivered to a destination with a desired CR. In order to minimize the number of totally transmitted packets, the strategy allocates an optimized one-hop packet transmission rate per each wireless link calculated by geometric programming.