Tran Cong Hung

Improving handover performance in Mobile IPv6

A seamless handover scheme with low latency and low packet loss is important to maintain TCP performance of mobile users. Several solutions have been proposed such as hierarchical mobile IPv6 (HMIPv6), fast handover protocol for mobile IPv6 (FMIPv6), seamless handoff architecture for mobile IP (S-MIP), ..., however, the overall handover latency is still too high for time-sensitive services. This paper proposes a novel improvement to existing frameworks by optimizing the address configuration stage so that the handover latency is further reduced. The proposed framework is based on fast handover for hierarchical mobile IPv6 and optimistic duplicate address detection.

Advanced Routing Algorithms and Load Balancing on MPLS

In this paper, we will survey and classify researching advanced routing algorithms that take advantages of MPLS (multiprotocol label switch) network to extent routing algorithms, supporting QoS (quality of service) services and traffic engineering and load balancing. We also consider some recent projects that are researching and working with advanced routing algorithms. This paper include 6 parts. Part 1 presents introduction. Part 2 presents QoS based routing, Classification of QoS routing. Part 3 presents traffic based routing based on information of current network such as: minimum interference routing algorithm (MIRA), dynamic online routing algorithm (DORA) and based on profile information such as: profile based routing (PBR). Part 4 presents employing advanced routing algorithms part 5 presents load balancing. And last, simulation with the combination of essential packages, we build simulation environments for MPLS based on ns2, setup some advanced routing algorithms, and evaluate them with old routing protocols.

Interoperability between Mobile IPv4 and Mobile IPv6 based on MPLS core network

The wireless network with its rapid development and wide use, will be the next investigation of researchers because there are many obstacles that must be passed to completely deploy it at present and on the future. This paper focuses on one part of the wireless field, that is, the interoperability between mobile IPv4 and mobile IPv6. For better QoS services and how the proposal can be applied for large network on Internet, we develop our solution on an MPLS network that is now used for QoS and high-speed core network.

Mechanisms of Mobile IP in delivering packets and its trends for changing from IPv4 to IPv6

Mobile IP was produced by the IP Routing for Wireless/Mobile Hosts working group of the IETF, approved by the Internet Engineering Steering Group (IESG) in June 1996 and published as a proposed standard in November 1996. It gave solutions to solve problems that a mobility device obtains when moving from one link to another. The movement makes the network-prefix routing fail in delivering packets because the mobile nodes new network-prefix is no longer equal the network-prefix assigned to its current link. This paper describes mechanisms used to deliver IP packets to mobile nodes which allow them to maintain all ongoing communications while changing links. Then, mobile IPv6s mechanisms on MPLS network are briefly presented as solutions for mobility devices in the future. Finally, the conclusion gives the approach to do research on mobile IP in the trends of making it scalable, secure and adaptive with the future network, especially in the period of changing from IPv4 to IPv6.

Sequential multidimensional scaling with Kalman filtering for location tracking

Localization always plays a critical role in wireless sensor networks for a wide range of applications including military, healthcare, and robotics. Although the classical multidimensional scaling (MDS) is a conventionally effective model for positioning, the accuracy of this method is affected by noises from the environment. In this paper, we propose a solution to attenuate noise effects to MDS by combining MDS with a Kalman filter. A model is built to predict the noise distribution with regard to additive noises to the distance measurements following the Gaussian distribution. From that, a linear tracking system is developed. The characteristics of the algorithm are examined through simulated experiments and the results reveal the advantages of our method over conventional works in dealing with the above challenges. Besides, the method is simplified with a linear filter; therefore it suits small and embedded sensors equipped with limited power, memory, and computational capacities well.

Human Organ Classifications from Computed Tomography Images Using Deep-Convolutional Neural Network

Deep neural networks (DNNs) have recently indicated outstanding performance on image learning feature tasks while Convolutional neural networks (CNNs) have been applied for classification tasks by reducing spectral variations and the spectral correlations of the model which existed in images. In this paper, we independently approached our work as a sequence of steps. We first implemented sparse autoencoders as an unsupervised algorithm to obtain learned features in two hidden layers for the DNN model by evaluating the appropriate input features and the optimal number of hidden units, which allow us to validate basic capabilities of the dataset. Secondly, we trained a deep CNN which consisted of five main convolutional layers, followed by Rectified Linear Units (ReLUs) layers, max-pooling layers, three fully-connected layers and a final softmax probability layer, to classify the high-resolution medical images of Computed Tomography (CT) into five anatomical classes, corresponding to five organs in abdominal regions. As a result, we considerably achieved the classification accuracy of 83.74 ± 3.34% in testing. We also visualized the layer representations on CT datasets, where they indicated the state-of-the-art performance, and could hold much promise to initialize further research on computer-aided diagnosis.

Combined Plane and Point Registration of Sequential Depth Images for Indoor Localization

Indoor environment usually has complicated structures and contains planes which can be used as extra features for registration. This paper proposes a method to use planar features extracted from depth camera for indoor localization. Two consecutive depth images are converted to point clouds and segmented to planes. Two sets of planes are then matched together to estimate the rotation matrix which is used as initial guess of iterative closest point (ICP) of point-to-plane registration. Our experimental result shows that ICP of point-toplane algorithm with the extra step of estimating the rotation guess matrix performs faster than conventional ICP and overcomes the drawback of ICP point-to-point algorithm in indoor environment.

Human extraction from a sequence of depth images using segmentation and foreground detection

This paper investigates on an approach of how to extract and track multiple subjects from a sequence of depth images. The Kinect camera is used to obtain a depth image revealing the depth information of a scene. Our proposed system includes the object clustering module to segment different isolated regions correspondent to objects in a depth image and foreground detection module to find moving regions from a sequence of frames. The combination of the two modules let us know which object is moving within a sequence of frames to locate a human subject. In order to extract the depth silhouettes of multiple subjects during time, we propose the use of matching algorithm between two consecutive frames to track their movements. We evaluate the algorithm with a long sequence of frames within a complex environment containing backgrounds with furniture and show how the algorithm is able to precisely extract and separate different human subjects with a fast processing speed. Therefore, the proposed approach is suitable for widely practical applications working with human activity recognition, human pose estimation and human tracking from depth images.