Wonjong Noh

Multi-path ad hoc routing considering path redundancy

This paper proposes a new on-demand multi-path routing protocol considering path redundancy as one of route selection criteria. Path redundancy implies how many possible redundant paths may exist on a route that contains more redundant paths toward the destination by involving intermediate nodes with relatively mode adjacent nodes in a possible route. Our approach can localize the effects of route failures, and reduce control traffic overhead and route reconfiguration time by enhancing the reachability to the destination node without source-initiated route rediscoveries at route failures. We have evaluated the performance of our routing scheme through a series of simulation using the network simulator 2 (ns-2).

Node Synchronization Based Redundant Routing for Mobile Ad-Hoc Networks

This paper proposes a new on-demand redundant-path routing protocol considering node synchronization based path redundancy as one of route selection criteria. Path redundancy implies how many possible redundant paths may exist on a route to be built up. Our proposal aims to establish a route that contains more redundant paths toward the destination by involving intermediate nodes with relatively mode adjacent nodes in a possible route. Our approach can localize the effects of route failures, and reduce control traffic overhead and route reconfiguration time by enhancing the reachability to the destination node without source-initiated route re-discoveries at route failures. We have evaluated the performance of our routing scheme through a series of simulations using the Network Simulator 2 (ns-2).

Time-Synchronized Neighbor Nodes Based Redundant Robust Routing (TSN2R3) for Mobile Sensor Networks

This paper proposes a new on-demand redundant-path routing protocol considering time synchronization based path redundancy as one of route selection criteria. Path redundancy implies how many possible redundant paths may exist on a route to be built up. Our proposal aims to establish a route that contains more redundant paths toward the destination by involving intermediate nodes with relatively mode adjacent nodes in a possible route. Our approach can localize the effects of route failures, and reduce control traffic overhead and route reconfiguration time by enhancing the reachability to the destination node without source-initiated route re-discoveries at route failures. We have evaluated the performance of our routing scheme through a series of simulations using the Network Simulator 2 (ns-2).

A Multi-path Support for Mobile IP with the Hierarchical Architecture

The explosive growth of the Internet combined with the widespread availability of highly mobile small hosts in the form of laptop and palmtop computers, and personal digital assistants, has created a big demand for the concept of Mobile IP. In the base Mobile IP, if the distance between the visited network and the home network of the mobile node is large, the signaling delay for these registrations may be long. Also the two main problems faced are the loss of registration packets and crashes of entities that maintain location caches. Therefore to solve this problem, the solution for performing registrations locally in the visited domain in order to reduce registration delay and to support micro mobility is proposed. In this paper, we propose that a hierarchical structure may be helpful to fast handoff in mobile environments. In company with the fast handoff, reliability is a great concern to the designers of mobile computing systems.

Multi-path routing using local virtual infrastructure for large-scale mobile ad-hoc networks: stochastic optimization approach

In this paper, we proposed new ad-hoc optimal redundant routing schemes for large-scale wireless mobile ad-hoc networks. First, we proposed locally and partially clustered network model and the concepts of synchronization. Second, we formulated a mathematical optimization model for the selection of an optimal redundant synchronous routing path. The performance evaluation results showed that the our optimal redundant routing scheme had better performance than SMR and TORA with respect to average throughput, average end-to-end delay and average control traffic overhead in large scale ad hoc networks.

Adaptive bus model for distributed multimedia stream in mobile computing environments

The main concern of this research is to design an adaptive bus model supporting distributed multimedia computing effectively in wireless or mobile networks. Key features of the model are as follows: first, the bus model has an ability, to control the binding at various levels explicitly. In addition to typical explicit binding characteristics, it has made possible a more flexible bus configuration from the application level to the transport level by hierarchical explicit binding control and controllable visibility of explicitness for security and integrity of computing. It enables the bus to adapt more flexibly in response to fluctuation of the environments. Second, the bus model has two-layer meta-object protocols (MOP), which are the component layer MOP and binding layer MOP, for implementation of adaptation. The component MOP plays the roles of inspecting and manipulating each component in the function and resource aspect. The binding MOP plays the roles of inspecting and manipulating each binding in the topology, access and flow aspect. Its architecture makes possible adaptations in integrity. Third, the bus model makes adaptations based on profiles, which is some collected information from the user, host and networks. The profile-based adaptation has made possible self-adaptation to some degree at each level and some prediction on usage pattern or resources.

Dynamic path control scheme in mobile ad hoc networks using on-demand routing protocol

In existing on-demand routing protocols, the generated paths cannot dynamically adapt to the change of network topology even if another route with less hop is available. Routing efficiency affects network performance and lifetime. In this paper, we propose a dynamic path control (DPC) scheme that can actively adapt to network topology change due to node mobility. Each node monitors its active neighboring information and tries to meet the situation when a better local link is available. We evaluate the proposed DPC scheme using analytical modeling and prove that the path available time of existing routing schemes can be significantly improved by the application of the proposed scheme.

Core nodes based stable routing for large-scale ad-hoc networks

In this paper, we propose new core nodes based stable ad-hoc routing scheme using synchronizations, as a critical factor of optimal routing path selection and its management, for stable and robust routing in large-scale wireless mobile ad-hoc networks. First, we proposed an ad-hoc network model including the concepts of hub and bridge node, spatial, time and mobility synchronization. Next, we proposed optimal core nodes based stable routing protocols using these network models. For the optimal core routing path selection, we derived a mathematical formulation using reconstructed core network topology. The performance of the proposed routing algorithm is evaluated in comparison with legacy ad-hoc routing algorithms such as DSR and CBRP with respect to average throughput, average normalized control overhead and average end-to-end delay jitter. The performance evaluation results show that our routing protocol has better performance than DSR and CBRP in large scale ad hoc networks.

Location Management with Dynamic Anchor Scheme in Wireless ATM Networks

This paper presents a dynamic anchor scheme in wireless ATM (asynchronous transfer mode) networks, which improves the location registers (LR) scheme, in which LR’s are hierarchically structured. This dynamic anchor scheme introduces the CCR (communication-to-computation cost ratio) term in the LR scheme. The proposed scheme reduces computation cost by introducing the active LR. There are at most two active LR’s in a hierarchical arrangement of LR’s and the location of one active LR is static and the location of another active LR called as anchor LR is dynamically selected among the LRs in the hierarchy in order to minimize a total location management cost. The total location management cost is calculated depending on the CCR as well as the CMR (call-to-move ratio). Numerical results show that the proposed scheme can reduce the location management cost compared with the original LR scheme for the different values of the CCR and the CMR.