Emotion recognition based on electroencephalography (EEG) signals has many potential applications. However, the development of an emotion recognition model for real-world deployment encounters significant setbacks in learning an invariant representation due to data distribution shift. Some prior studies adopted domain adaptation methods to tackle this issue. However, these studies perform marginally to tackle distribution shift in the source data and neglect to focus on the extraction of domain and target-specific features, which are crucial for domain invariant representation learning. In this study, we propose an assisted domain adaptive representation learning (ADARL) method that learns domain invariant and emotion-specific features for recognizing emotions across different subjects. The proposed architecture introduces novel attention modules to segregate domain-specific and emotion-specific feature maps by exploiting the inter-relationship between different channels and bands. The domain-specific residual feature maps (peripheral features) extracted by the proposed attention mechanism are used to assist the domain discriminator that forces the model to learn a generalized representation of EEG signals. To tackle the inter-subject variability in the source data, we introduce a novel class-aware compactness loss function to improve the domain adaptation transfer efficiency. The experiments conducted on three EEG datasets (SEED, SEED-IV, and DEAP) demonstrate that the ADARL method significantly improves emotion recognition performance and efficiently alleviates the domain shift problem by learning a generalized representation of EEG signals.

Extracting fine-grained OCR text from aged documents in diacritic languages remains challenging due to unexpected artifacts, time-induced degradation, and lack of datasets. While standalone spell correction approaches have been proposed, they show limited performance for historical documents due to numerous possible OCR error combinations and differences between modern and classical corpus distributions. We propose a method utilizing available content-focused ebooks as a reference base to correct imperfect OCR-generated text, supported by large language models. This technique generates high-precision pseudo-page-to-page labels for diacritic languages, where small strokes pose significant challenges in historical conditions. The pipeline eliminates various types of noise from aged documents and addresses issues such as missing characters, words, and disordered sequences. Our post-processing method, which generated a large OCR dataset of classical Vietnamese books, achieved a mean grading score of 8.72 on a 10-point scale. This outperformed the state-of-the-art transformer-based Vietnamese spell correction model, which scored 7.03, when evaluated on a sampled subset of the dataset. We also trained a baseline OCR model to assess and compare it with well-known engines. Experimental results demonstrate the strength of our baseline model compared to widely used open-source solutions. The resulting dataset will be released publicly to support future studies.

Segmenting biomarkers in medical images is crucial for various biotech applications. Despite advances, Transformer and CNN based methods often struggle with variations in staining and morphology, limiting feature extraction. In medical image segmentation, where datasets often have limited sample availability, recent state-of-the-art (SOTA) methods achieve higher accuracy by leveraging pre-trained encoders, whereas end-to-end methods tend to underperform. This is due to challenges in effectively transferring rich multiscale features from encoders to decoders, as well as limitations in decoder efficiency. To address these issues, we propose an architecture that captures multi-scale local and global contextual information and a novel decoder design, which effectively integrates features from the encoder, emphasizes important channels and regions, and reconstructs spatial dimensions to enhance segmentation accuracy. Our method, compatible with various encoders, outperforms SOTA methods, as demonstrated by experiments on four datasets and ablation studies. Specifically, our method achieves absolute performance gains of 2.76% on MoNuSeg, 3.12% on DSB, 2.87% on Electron Microscopy, and 4.03% on TNBC datasets compared to existing SOTA methods. Code: https://github.com/saadwazir/MCADS-Decoder

Video super-resolution remains a major challenge in low-level vision tasks. To date, CNN- and Transformer-based methods have delivered impressive results. However, CNNs are limited by local receptive fields, while Transformers struggle with quadratic complexity, posing challenges for processing long sequences in VSR. Recently, Mamba has drawn attention for its long-sequence modeling, linear complexity, and large receptive fields. In this work, we propose VSRM, a novel \textbf{V}ideo \textbf{S}uper-\textbf{R}esolution framework that leverages the power of \textbf{M}amba. VSRM introduces Spatial-to-Temporal Mamba and Temporal-to-Spatial Mamba blocks to extract long-range spatio-temporal features and enhance receptive fields efficiently. To better align adjacent frames, we propose Deformable Cross-Mamba Alignment module. This module utilizes a deformable cross-mamba mechanism to make the compensation stage more dynamic and flexible, preventing feature distortions. Finally, we minimize the frequency domain gaps between reconstructed and ground-truth frames by proposing a simple yet effective Frequency Charbonnier-like loss that better preserves high-frequency content and enhances visual quality. Through extensive experiments, VSRM achieves state-of-the-art results on diverse benchmarks, establishing itself as a solid foundation for future research.

Large language models (LLMs) often exhibit strong biases, e.g, against women or in favor of the number 7. We investigate whether LLMs would be able to output less biased answers when allowed to observe their prior answers to the same question in a multi-turn conversation. To understand which types of questions invite more biased answers, we test LLMs on our proposed set of questions that span 9 topics and belong to three types: (1) Subjective; (2) Random; and (3) Objective. Interestingly, LLMs are able to "de-bias" themselves in a multi-turn conversation in response to questions that seek a Random, unbiased answer. Furthermore, we propose B-score, a novel metric that is effective in detecting biases in Subjective, Random, Easy, and Hard questions. On MMLU, HLE, and CSQA, leveraging B-score substantially improves the verification accuracy of LLM answers (i.e, accepting LLM correct answers and rejecting incorrect ones) compared to using verbalized confidence scores or the frequency of single-turn answers alone. Code and data are available at: b-score.github.io.

Developing an efficient heartbeat monitoring system has become a focal point in numerous healthcare applications. Specifically, in the last few years, heartbeat classification for arrhythmia detection has gained considerable interest from researchers. This paper presents a novel deep representation learning method for the efficient detection of arrhythmic beats. To mitigate the issues associated with the imbalanced data distribution, a novel re-sampling strategy is introduced. Unlike the existing oversampling methods, the proposed technique transforms majority-class samples into minority-class samples with a novel translation loss function. This approach assists the model in learning a more generalized representation of crucially important minority class samples. Moreover, by exploiting an auxiliary feature, an augmented attention module is designed that focuses on the most relevant and target-specific information. We adopted an inter-patient classification paradigm to evaluate the proposed method. The experimental results of this study on the MIT-BIH arrhythmia database clearly indicate that the proposed model with augmented attention mechanism and over-sampling strategy significantly learns a balanced deep representation and improves the classification performance of vital heartbeats.

The majority of current Text-to-Speech (TTS) datasets, which are collections of individual utterances, contain few conversational aspects. In this paper, we introduce DailyTalk, a high-quality conversational speech dataset designed for conversational TTS. We sampled, modified, and recorded 2,541 dialogues from the open-domain dialogue dataset DailyDialog inheriting its annotated attributes. On top of our dataset, we extend prior work as our baseline, where a non-autoregressive TTS is conditioned on historical information in a dialogue. From the baseline experiment with both general and our novel metrics, we show that DailyTalk can be used as a general TTS dataset, and more than that, our baseline can represent contextual information from DailyTalk. The DailyTalk dataset and baseline code are freely available for academic use with CC-BY-SA 4.0 license1.

There are many real-life use cases such as barcode scanning or billboard reading where people need to detect objects and read the object contents. Commonly existing methods are first trying to localize object regions, then determine layout and lastly classify content units. However, for simple fixed structured objects like license plates, this approach becomes surplus and heavy to run. This work aims to solve this detect-and-read problem in a lightweight way by integrating multi-digit recognition into a one-stage object detection model. Our unified method not only eliminates the duplication in feature extraction (one for localizing, one again for classifying) but also provides useful contextual information around object regions for classification. Additionally, our choice of backbones and modifications in architecture, loss function, data augmentation and training make the method robust, efficient and speedy. Secondly, we made a public benchmark dataset of diverse real-life 1D barcodes for a reliable evaluation, which we collected, annotated and checked carefully. Eventually, experimental results prove the method's efficiency on the barcode problem by outperforming industrial tools in both detecting and decoding rates with a real-time fps at a VGA-similar resolution. It also did a great job expectedly on the license-plate recognition task (on the AOLP dataset) by outperforming the current state-of-the-art method significantly in terms of recognition rate and inference time.

Adversarial attack is aimed at fooling a target classifier with imperceptible perturbation. Adversarial examples, which are carefully crafted with a malicious purpose, can lead to erroneous predictions, resulting in catastrophic accidents. To mitigate the effect of adversarial attacks, we propose a novel purification model called CAP-GAN. CAP-GAN considers the idea of pixel-level and feature-level consistency to achieve reasonable purification under cycle-consistent learning. Specifically, we utilize a guided attention module and knowledge distillation to convey meaningful information to the purification model. Once the model is fully trained, inputs are projected into the purification model and transformed into clean-like images. We vary the capacity of the adversary to argue the robustness against various types of attack strategies. On CIFAR-10 dataset, CAP-GAN outperforms other pre-processing based defenses under both black-box and white-box settings.

This paper proposes integrating semantics-oriented similarity representation into RankingMatch, a recently proposed semi-supervised learning method. Our method, dubbed ReRankMatch, aims to deal with the case in which labeled and unlabeled data share non-overlapping categories. ReRankMatch encourages the model to produce the similar image representations for the samples likely belonging to the same class. We evaluate our method on various datasets such as CIFAR-10, CIFAR-100, SVHN, STL-10, and Tiny ImageNet. We obtain promising results (4.21% error rate on CIFAR-10 with 4000 labels, 22.32% error rate on CIFAR-100 with 10000 labels, and 2.19% error rate on SVHN with 1000 labels) when the amount of labeled data is sufficient to learn semantics-oriented similarity representation. The code is made publicly available at https://github.com/tqtrunghnvn/ReRankMatch.

Previous works on neural text-to-speech (TTS) have been addressed on limited speed in training and inference time, robustness for difficult synthesis conditions, expressiveness, and controllability. Although several approaches resolve some limitations, there has been no attempt to solve all weaknesses at once. In this paper, we propose STYLER, an expressive and controllable TTS framework with high-speed and robust synthesis. Our novel audio-text aligning method called Mel Calibrator and excluding autoregressive decoding enable rapid training and inference and robust synthesis on unseen data. Also, disentangled style factor modeling under supervision enlarges the controllability in synthesizing process leading to expressive TTS. On top of it, a novel noise modeling pipeline using domain adversarial training and Residual Decoding empowers noise-robust style transfer, decomposing the noise without any additional label. Various experiments demonstrate that STYLER is more effective in speed and robustness than expressive TTS with autoregressive decoding and more expressive and controllable than reading style non-autoregressive TTS. Synthesis samples and experiment results are provided via our demo page, and code is available publicly.

Explaining the decisions of an Artificial Intelligence (AI) model is increasingly critical in many real-world, high-stake applications.Hundreds of papers have either proposed new feature attribution methods, discussed or harnessed these tools in their work.However, despite humans being the target end-users, most attribution methods were only evaluated on proxy automatic-evaluation metrics (Zhang et al. 2018; Zhou et al. 2016; Petsiuk et al. 2018). In this paper, we conduct the first user study to measure attribution map effectiveness in assisting humans in ImageNet classification and Stanford Dogs fine-grained classification, and when an image is natural or adversarial (i.e., contains adversarial perturbations). Overall, feature attribution is surprisingly not more effective than showing humans nearest training-set examples. On a harder task of fine-grained dog categorization, presenting attribution maps to humans does not help, but instead hurts the performance of human-AI teams compared to AI alone. Importantly, we found automatic attribution-map evaluation measures to correlate poorly with the actual human-AI team performance. Our findings encourage the community to rigorously test their methods on the downstream human-in-the-loop applications and to rethink the existing evaluation metrics.

ChronoGraph is a novel system enabling temporal graph traversals. Compared to snapshot-oriented systems, this traversal-oriented system is suitable for analyzing information diffusion over time without violating a time constraint on temporal paths. The cornerstone of ChronoGraph aims at bridging the chasm between point-based semantics and period-based semantics and the gap between temporal graph traversals and static graph traversals. Therefore, our graph model and traversal language provide the temporal syntax for both semantics, and we present a method converting point-based semantics to period-based ones. Also, ChronoGraph exploits the temporal support and parallelism to handle the temporal degree, which explosively increases compared to static graphs. We demonstrate how three traversal recipes can be implemented on top of our system: temporal breadth-first search (tBFS), temporal depth-first search (tDFS), and temporal single source shortest path (tSSSP). According to our evaluation, our temporal support and parallelism enhance temporal graph traversals in terms of convenience and efficiency. Also, ChronoGraph outperforms existing property graph databases in terms of temporal graph traversals. We prototype ChronoGraph by extending Tinkerpop, a de facto standard for property graphs. Therefore, we expect that our system would be readily accessible to existing property graph users.

Barcodes are ubiquitous and have been used in most critical daily activities for decades. However, most traditional decoders require well-founded barcode under a relatively standard condition. While wilder conditioned barcodes such as underexposed, occluded, blurry, wrinkled and rotated are commonly captured in reality, those traditional decoders show weakness of recognizing. Several works attempted to solve those challenging barcodes, but many limitations still exist. This work aims to solve the decoding problem using deep convolutional neural network with the possibility of running on portable devices. Firstly, we proposed a special modification of inference based on the feature of having checksum and test-time augmentation, named Smart Inference (SI), in the prediction phase of a trained model. SI considerably boosts accuracy and reduces the false prediction for trained models. Secondly, we have created a large practical evaluation dataset of real captured 1D barcode under various challenging conditions to test our methods vigorously, publicly available for other researchers. The experiments' results demonstrated the SI effectiveness with the highest accuracy of 95.85% which outperformed many existing decoders on the evaluation set. Finally, we successfully minimized the best model by knowledge distillation to a shallow model which is shown to have high accuracy (90.85%) with a good inference speed of 34.2 ms per image on a real edge device.

Human pose estimation - the process of recognizing human keypoints in a given image - is one of the most important tasks in computer vision and has a wide range of applications including movement diagnostics, surveillance, or self-driving vehicle. The accuracy of human keypoint prediction is increasingly improved thanks to the burgeoning development of deep learning. Most existing methods solved human pose estimation by generating heatmaps in which the ith heatmap indicates the location confidence of the ith keypoint. In this paper, we introduce novel network structures referred to as multi-resolution representation learning for human keypoint prediction. At different resolutions in the learning process, our networks branch off and use extra layers to learn heatmap generation. We firstly consider the architectures for generating the multi-resolution heatmaps after obtaining the lowest-resolution feature maps. Our second approach allows learning during the process of feature extraction in which the heatmaps are generated at each resolution of the feature extractor. The first and second approaches are referred to as multi-resolution heatmap learning and multi-resolution feature map learning respectively. Our architectures are simple yet effective, achieving good performance. We conducted experiments on two common benchmarks for human pose estimation: MS-COCO and MPII dataset. The code is made publicly available at https://github.com/tqtrunghnvn/SimMRPose.

Nowadays, deep learning techniques show dramatic performance in computer vision areas, and they even outperform humans on complex tasks such as ImageNet classification. But it turns out a deep learning based model is vulnerable to some small perturbation called an adversarial attack. This is a problem in the view of the safety and security of artificial intelligence, which has recently been studied a lot. These attacks have shown that they can easily fool models of image classification, semantic segmentation, and object detection. We focus on the adversarial attack in semantic segmentation tasks since there is little work in this task. We point out this attack can be protected by denoise autoencoder, which is used for denoising the perturbation and restoring the original images. We build a deep denoise autoencoder model for removing the adversarial perturbation and restoring the clean image. We experiment with various noise distributions and verify the effect of denoise autoencoder against adversarial attack in semantic segmentation task.

This paper presents a novel Echo State Network (ESN) model for chaotic time series prediction, which consists of three steps including input reconstruction, dimensionality reduction and regression. First, phase-space reconstruction is used to reconstruct the original ‘attractor’ of the input time series. Then, Independent Component Analysis (ICA) is used to identify independent components, reduce dimensionality and overcome multicollinearity problem of the reconstructed input matrix. Finally, Bayesian Ridge Regression provides accurate predictions thanks to its regularization effect to avoid over-fitting and its robustness to noise owing to its probabilistic strategy. Our experimental results show that our model significantly outperforms other ESN models in predicting both artificial and real-world chaotic time series.

ChronoGraph is a novel system enabling temporal graph traversals. Compared to snapshot-oriented systems, this traversal-oriented system is suitable for analyzing information diffusion over time without violating a time constraint on temporal paths. The cornerstone of ChronoGraph aims at bridging the chasm between point-based semantics and period-based semantics and the gap between temporal graph traversals and static graph traversals. Therefore, our graph model and traversal language provide the temporal syntax for both semantics, and we present a method converting point-based semantics to period-based ones. Also, ChronoGraph exploits the temporal support and parallelism to handle the temporal degree, which explosively increases compared to static graphs. We demonstrate how three traversal recipes can be implemented on top of our system: temporal breadth-first search (tBFS), temporal depth-first search (tDFS), and temporal single source shortest path (tSSSP). According to our evaluation, our temporal support and parallelism enhance temporal graph traversals in terms of convenience and efficiency. Also, ChronoGraph outperforms existing property graph databases in terms of temporal graph traversals. We prototype ChronoGraph by extending Tinkerpop, a de facto standard for property graphs. Therefore, we expect that our system would be readily accessible to existing property graph users.

Today, the Internet of Things (IoT) technology is applied everywhere, providing tremendous amounts of IoT service. The GS1, a non-profit international standards organization, has established standards for IoT technology. Especially, the GS1 standardized an Object Name Service (ONS) leveraging DNS's distributed and federated infrastructure, enables users to manage and discover IoT services such as the retail, food, healthcare, smart city, and so on. However, the ONS is vulnerable to the data tampering, privilege escalation, and service fault caused by DNS attacks including the man in the middle, cache poisoning, replay, hijacking, and denial of service attacks. Nowadays, IoT services are used in security-sensitive areas, such as finance and healthcare. Therefore, the security of ONS should be strengthened before causing severe problems such as data breach, economic loss, and social loss. In this paper, we propose a blockchain-based ONS with a tokenized authority called the BlockONS. The BlockONS provides strength in the data tampering and privilege escalation allowing a fault tolerance. The BlockONS consists of a content off-chain scaling, a data tampering validation method, a fault-tolerance method, and a Blockchain Token-Based Access Control (BTBAC) method. We designed the BlockONS into two parts: A BlockONS Node part manages the validation method and BTBAC model. A BlockONS Agent part manages the off-chain scaling and fault tolerance. Finally, we implemented the BlockONS leveraging a Hyperledger Sawtooth blockchain. We proved the proposed validation method, fault tolerance method, and BTBAC method through use cases and security analyses on attack situations. We deployed the BlockONS in the Daejeon city and evaluated the performance to show the feasibility of the BlockONS.

As supply chain has become extremely complicated, traditional supply chain management schemes reveal limitations in keeping track of resources for efficient risk management. Recently, blockchain-based solutions for supply chain have shown possibility to overcome those limits. However, numerous practical problems in this research domain are not fully explored up to date. Thus, we delve into one of the most significant problems among them: verification of business logic in transactions. In this paper, we propose a novel distributed ledger system for supply chain, enabled by anomaly detection framework that verifies semantic correctness of transactions based on business context data. Our blockchain data model is tailored to accurately represent events occurred in supply chain based on real-world business standards. In order to facilitate more efficient tracking of provenance, we leverage graph data model to represent supply chain network. On top of the data model, we present smart contract-based anomaly detection framework that verifies whether a transaction is anomalous. Generic rule-based and graph-based detection methods are devised. The feasibility of our proposed model is shown by the system implemented using Hyperledger Sawtooth. We show how our anomaly detection layer can be plugged into the system: how it interacts with other system components, and how overall system flow works with this new function. We evaluate our system with scenario-based simulations. For a number of use cases we synthesized, the correctness and effectiveness of our system are demonstrated.

Since the concept of Cognitive Radio (CR) was first introduced, it has been considered as a key technology of future wireless devices to better utilize radio spectrum. A number of CR-MAC protocols have been studied for Cognitive Radio Networks (CRNs), but most state-of-the-art frequencyhopping based protocols focus mainly on channel mobility and spectrum resource allocation. They lack integration of other essential features such as time synchronization and cooperative spectrum sensing, which are practically crucial, into a full-blown CR-MAC protocol. In this paper, we propose MSHCS-MAC: a mac protocol for Multi-hop CRNs based on Slow Hopping and Cooperative Sensing approach which is the cutting edge frequency hopping scheme. The contributions of MSHCS-MAC are threefold: (1) Support of multi-hop communication without dedicated control channel and multiple transceivers, (2) Integration of essential CR-MAC features such as bootstrapping, multi-channel operation, cooperative spectrum sensing and time synchronization, (3) Practical implementation and evaluation on commercial devices. The evaluation results show that MSHCS-MAC provides reasonable performance in the experimental testbed with supporting multihop communication and essential CR-MAC features.

In recent years, the connected automotive industry has grown explosively. Various connected car services are emerging, such as remote vehicle diagnostics, driver's health monitoring, infotainment, and vehicle safety management. As a result, the number and type of vehicle data are increasing tremendously day by day. However, existing connected automotive solutions have a limitation in that each company manages its own closed data silos. This restricts connected car services from using data sources in various domains. Hence, we propose the GS1 Connected Car, an integrated vehicle information platform, and its ecosystem. We suggest GS1-based automotive data standards for not only in-vehicle data but also all the automotive-related data generated during the lifecycle of vehicles. We provide standardized data collection to EPCIS, the discovery of global automotive services using ONS, IoT Mash-up service between an in-car dashboard platform and IoT devices, video infotainment called GS1 video, and automotive lifecycle management application. We have implemented our platform in a real car by developing an Android-based vehicle dashboard, including service discovery, Mash-up services, and GS1 Video. Also, a mobile application for lifecycle management and Amazon skills for collecting driver's information are developed. Our demonstration and case study show the feasibility of the proposed platform, widening the scope of future connected car services.

A new form of cloud computing, serverless computing, is drawing attention as a new way to design micro-services architectures. In a serverless computing environment, services are developed as service functional units. The function development environment of all serverless computing framework at present is CPU based. In this paper, we propose a GPU-supported serverless computing framework that can deploy services faster than existing serverless computing framework using CPU. Our core approach is to integrate the open source serverless computing framework with NVIDIA-Docker and deploy services based on the GPU support container. We have developed an API that connects the open source framework to the NVIDIA-Docker and commands that enable GPU programming. In our experiments, we measured the performance of the framework in various environments. As a result, developers who want to develop services through the framework can deploy high-performance micro services and developers who want to run deep learning programs without a GPU environment can run code on remote GPUs with little performance degradation.

In CR networks, to share the conditions whether the channel is occupied or not, many existing researches generally adopt common control channel(CCC) and multiple radio transceivers due to easy design and implementation for multi-hop communications. However, they are resource wasteful because one channel should be always occupied for CCC and the radio transceiver is additionally required. In this paper, we propose a new time synchronization method for multi-hop communication by extending the SHCS-MAC protocol. We implemented the proposed time synchronization method on IEEE 802.15.4 Zigbee PHY based USRP SDR devices using GNU Radio and conduct the experiment to confirm it works properly.

Nowadays, Graphics Processing Unit (GPU) is essential for general-purpose high-performance computing, because of its dominant performance in parallel computing compare to that of CPU. There have been many successful trials on the use of GPU in virtualized environment. Especially, NVIDIA Docker obtained a most practical way to bring GPU into the container-based virtualized environment. However, most of these trials did not consider sharing GPU among multiple containers. Without the above consideration, a system will experience a program failure or a deadlock situation in the worst case. In this paper, we propose ConVGPU, a solution to share the GPU in multiple containers. With ConVGPU, the system can guarantee the required GPU memory which the container needs to execute. To achieve it, we introduce four scheduling algorithms that manage the GPU memory to be taken by the containers. These algorithms can prevent the system from falling into deadlock situations between containers during execution.

In the EPCglobal standards for RFID architecture frameworks and interfaces, the Electronic Product Code Information System (EPCIS) acts as a standard repository storing event and master data that are well suited to Supply Chain Management (SCM) applications. Oliot-EPCIS broadens its scope to a wider range of IoT applications in a scalable and flexible way to store a large amount of heterogeneous data from a variety of sources. However, this expansion poses data security challenge for IoT applications including patients' ownership of events generated in mobile healthcare services. Thus, in this paper we propose Secure-EPCIS to deal with security issues of EPCIS for IoT applications. We have analyzed the requirements for Secure-EPCIS based on real-world scenarios and designed access control model accordingly. Moreover, we have conducted extensive performance comparisons between EPCIS and Secure-EPCIS in terms of response time and throughput, and provide the solution for performance degradation problem in Secure-EPCIS.

Tactical Operations Center (TOC) system in military field is an advanced computer system composed of multiple servers and desktops to interlock internal/external weapon systems processing mission-critical applications in combat situation. However, the current TOC system has several limitations such as difficulty of integrating tactical weapon systems including missile launch system and radar system into the single TOC system due to the heterogeneity of HW and SW between systems, and an inefficient computing resource management for the weapon systems.

One of the major challenges in 6LoWPAN is to provide continuous services while mobile nodes' movements with minimizing network inaccessible time caused due to handoffs. Even though MIPv6, HMIPv6, and PMIPv6 are commonly accepted standards to address this in IP networks, they cannot inherently avoid the degradation in communication quality during handoff, since they are not designed with consideration of constrained node networks like 6LoWPAN. In this paper, we propose a new fast mobility management protocol for 6LoWPAN, named intra-MARIO. To minimize handoff delay and enhance service availability, intra-MARIO introduces three important components, which are a fast rejoin scheme for handoff management with an adaptive polling based movement detection and multi-hop pointer forwarding schemes for location management. To justify the effectiveness, we have conducted extensive simulations by comparing intra-MARIO with prior schemes like a basic mobility management scheme and a PMIPv6-based protocol. We then implement intra-MARIO on top of our 6LoWPAN platform (SNAIL) and evaluate the performance of intra-MARIO. The results highlight that intra-MARIO reduces overall handoff delay with low power consumption and minimizes packet losses during handoffs, compared to prior mobility protocols.

A fusion system, which collects air-tracks from distributed radars and eliminates duplicated tracks, is required to get a Single Integrated Air Picture (SIAP) for the wide surveillance area. We are developing a distributed radar system which consists of numerous mobile radars to cover wide surveillance area. Two-tier fusion system, which is the well-known solution for wide surveillance area, was adopted. Two-tier fusion system allows to put local fusion nodes between local radars and the central fusion node. We argue that the number of processed tracks of the two-tier fusion system is highly correlated with the fusion tree which decides the local fusion nodes and their child radars. We improved the number of processed tracks of our radar system by applying the fusion tree control which has been neglected in the most of research in this field. However it is hard to generate a proper fusion tree due to the possible number of fusion tree increases exponentially as the number of radars increase. To solve this problem, we propose a novel self-organized fusion tree generation algorithm. Especially, the proposed solution generates the fusion tree without any prior information, such as network-topology, and position of radars. We evaluate the performance of the proposed solution using the OPNET network simulator and show that the proposed solution performs better than the naive methods.

In the Internet of Things, tremendous data is generated by huge number of things and stored in globally distributed data repositories. In this respect, many IoT applications would need to find and get the data indirectly from the data repositories or directly from things. GS1, global leading standard organization, firstly presents the concept of Discovery Service (DS) to find the desired EPC Information Services (EPCIS) like data repositories for things. To realize GS1 DS, many works are proposed with considerations of performance and security issues. However, most works are only applied to the Inter-DS finding one of distributed Intra-DSes; thus deterioration in the Intra-DS actually finding desired data repositories may adversely affect entire DS. In this paper, we propose Oliot-Discovery Service (DS) which deals with performance as well as security issues by especially focusing on Intra-DS aspect. For this, we have designed and implemented the access control pursuing the fine-grained access control model and the two-layered storage architecture comprised of cache and main databases. Moreover, it supports not only indirect access to things' data through the data repositories but also direct access to things which would be also required for many IoT applications. In order to show feasibility of Oliot-DS, we have constructed experimental test-bed and evaluated the performance. The experiment results show that Oliot-DS provides durability against many user transactions, and the reasonable write and read performances by preventing unauthorized access.

A deep neural networks is proposed for the classification of premature ventricular contraction (PVC) beat, which is an irregular heartbeat initiated by Purkinje fibers rather than by sinoatrial node. Several machine learning approaches were proposed for the detection of PVC beats although they resulted in either achieving low accuracy of classification or using limited portion of data from existing electrocardiography (ECG) databases. In this paper, we propose an optimized deep neural networks for PVC beat classification. Our method is evaluated on TensorFlow, which is an open source machine learning platform initially developed by Google. Our method achieved overall 99.41% accuracy and a sensitivity of 96.08% with total 80,836 ECG beats including normal and PVC from the MIT-BIH Arrhythmia Database.

Recent years have seen the development of successful internet of things (IoT) technologies based on an IP-enabled 6LoWPAN, which enables the extensive message exchange of information generated from various applications such as healthcare, smart home, and factory automation. Because IoT devices are closely related to the human life, they generally handle critical data which must be protected from a malicious adversary. However, a majority of IoT devices are resource-constrained in terms of memory and computational ability, so that they cannot provide heavy security protocols and authentication. To protect connections over constrained networks, we introduce a back-end offload architecture which offloads the processing of a security protocol to a specified back-end offloader. The offloader assists constrained devices back-end by handling the packets of handshake procedure and encrypted application data. The load balancing of offloaders and the protection of offload messages are also provided in the design. The proposed architecture allows an extremely constrained device to establish a secure session by utilizing high-level authentications such as public key infrastructure or certificate, without the burden of deploying heavy security modules. This research would be advantageous for incompetent nodes to support security and reduce cost at the same time.

In food service, the transparency and reliability of product data are especially important requirements. But most of distributed information is non-trusted data without accurate source. To provide reliable data of food products, we propose the metadata service architecture for food service, applying GS1 Source defined by GS1 (Global Standard 1). GS1 Source is a standard for providing a communication of authentic and accurate product data between brand owners and consumers. We implemented the metadata service architecture as a java web service. The implemented architecture was applied in real food service and verified feasibility. We tested the performance of the web service by estimating request time, error rate, and response time, and conformed sufficient performance of the service. By applying GS1 Source to the food system, we can contribute to food service ecosystem with trusted data.

To ensure global interoperability in today's fast-moving trading networks, EPC global Network (EPC Network) architecture has been at the core of Electronic Product Code (EPC) as a universal identifier for objects' virtual representations. While Cloud technology has largely been used for EPC Network's real world deployment for thousands of vendors, an optimized Cloud solution has yet to be provided. Therefore, in this paper, we explore the challenges of EPC Network deployment, and propose our own Cloud solution with optimizations in load prediction and migration management. As proof of concept, we have conducted experiments with regard to prediction accuracy and migration performance.

The "Cloud of Things" (CoT) is a concept that provides smart things' functions as a service and allows them to be used by multiple applications. In the CoT, a single smart thing instance should efficiently host multiple applications, called multi-tenancy. Therefore, simultaneous accesses to shared smart things may lead to resource conflicts. Moreover, smart things inherently form complex dependencies on real-world. Since handling resource conflicts with complex dependencies at an application level is typically ad-hoc and error-prone, it increases the application development burden. To address these issues, we propose a middleware for Evolvable ClOut of things (ECO). The ECO middleware manages organizations to handle dependency among/between smart things and virtualizes physical smart things to enable isolation between/among multiple applications yet internally controls smart things' sharing to resolve resource conflicts and provides a consolidation framework for efficient utilization of the shared smart things. A lease-based sharing control is employed with two tenant switch schemes which are analyzed. All these features are hidden from application context, thus reducing complexities in developing applications. From implementation and evaluations with workload applications, we show that the ECO middleware provides efficient sharing controls and access controls with negligible virtualization overhead.

In this paper, we propose a novel semi-directional flooding (SDF) algorithm for on-demand routing in IEEE 802.15.4-based low-rate wireless mesh networks (LRWMNs). The novelty of our work is that the proposed routing algorithm enables route discovery request (RREQ) packets to be semi-directionally flooded, with respect to a source-destination pair, in a fully distributed manner without either physical or virtual location information. The idea behind this is to exploit the hierarchical addressing structure that allows each router node to compute logical tree distances between source-destination pairs, which approximately mirrors relative hop distance between the pairs, without message exchanges. By exploiting tree distances with a given target address, the SDF algorithm enables RREQ packets to be flooded over a small area, directed towards the destination. We also apply SDF to a lightweight on-demand routing algorithm like AODVjr, and propose two techniques of adaptive timers and route repair that reduces path setup delays and overheads during route rediscovery, respectively. We carry out extensive simulations and quantitatively show that SDF significantly reduces flooding overhead of network-wide flooding (NWF). We compare SDF-based AODVjr with both NWF-based AODVjr and enhanced hierarchical routing protocol (EHRP) for on demand routing and demonstrate that SDF-based AODVjr drastically reduces route discovery overhead by up to about 95 percent while still providing comparable or better performance than NWF-based AODVjr and EHRP.

An increasing number of recent experimental works have demonstrated that the supposedly secure channels in the Internet are prone to privacy breaking under many respects, due to packet traffic features leaking information on the user activity and traffic content. We aim at understanding if and how complex it is to obfuscate the information leaked by packet traffic features, namely packet lengths, directions, and times: we call this technique traffic masking. We define a security model that points out what the ideal target of masking is, and then define the optimized traffic masking algorithm that removes any leaking (full masking). Further, we investigate the tradeoff between traffic privacy protection and masking cost, namely required amount of overhead and realization complexity/feasibility. Numerical results are based on measured Internet traffic traces. Major findings are that: 1) optimized full masking achieves similar overhead values with padding only and in case fragmentation is allowed, and 2) if practical realizability is accounted for, optimized statistical masking attains only moderately better overhead than simple fixed pattern masking does, while still leaking correlation information that can be exploited by the adversary.

The ZigBee tree routing is widely used in many resource-limited devices and applications, since it does not require any routing table and route discovery overhead to send a packet to the destination. However, the ZigBee tree routing has the fundamental limitation that a packet follows the tree topology; thus, it cannot provide the optimal routing path. In this paper, we propose the shortcut tree routing (STR) protocol that provides the near optimal routing path as well as maintains the advantages of the ZigBee tree routing such as no route discovery overhead and low memory consumption. The main idea of the shortcut tree routing is to calculate remaining hops from an arbitrary source to the destination using the hierarchical addressing scheme in ZigBee, and each source or intermediate node forwards a packet to the neighbor node with the smallest remaining hops in its neighbor table. The shortcut tree routing is fully distributed and compatible with ZigBee standard in that it only utilizes addressing scheme and neighbor table without any changes of the specification. The mathematical analysis proves that the 1-hop neighbor information improves overall network performances by providing an efficient routing path and distributing the traffic load concentrated on the tree links. In the performance evaluation, we show that the shortcut tree routing achieves the comparable performance to AODV with limited overhead of neighbor table maintenance as well as overwhelms the ZigBee tree routing in all the network conditions such as network density, network configurations, traffic type, and the network traffic.

This paper proposes an interesting yet practical use case of CoAP, an important application protocol in realizing the Internet of Things (IoT) vision. Originally, CoAP protocol is designed to communicate between embedded devices, however its applications can spread well over the Internet by exposing the devices' capabilities to the Web as Web resources. Thus, we claim that CoAP protocol could also be used in web-based application to leverage CoAP-based resources. Unfortunately, the design of CoAP protocol, for example, bind to UDP socket to reduce the package size or bidirectional communication to cope with duty-cycle power scheme, prevents it from being supported on conventional web browsers. Additionally, translation of CoAP into HTTP protocol via HTTP/CoAP proxy as recommended by IETF is not the best solution since many CoAP features are limited. Instead, new bidirectional web protocol such as HTML5 Web Socket would help to preserve the protocol's characteristics. We propose a solution called SCoAP that facilitates true CoAP communication in conventional web browsers by applying HTML5 Web Socket protocol. Experiment results have shown significant advantages of SCoAP solution over standard HTTP/CoAP proxy in term of network traffic and computational demand.

Tree based hierarchical routing is widely used in 6LoWPAN due to advantages of low control overhead and low memory resources. However, it generates a triangular detour path considering point-to-point (P2P) traffic. Reactive routing provides a reasonable P2P path by conducting flooding-based route discovery, but the flooding increases packet interference due to huge control overhead. Multipath routing can improve the reliability by preparing for the alternative path. Nevertheless, it needs an extra table for storing the alternative path and it is not appropriate to diverse traffic patterns. In this paper, we propose a traffic-aware multipath routing (TAMR). By managing an overlaid topology comprised of two different topologies without additional overhead, TAMR can dynamically choose a proper routing protocol according to traffic patterns and it can improve network reliability. Besides, TAMR is suitable for 6LoWPAN, because it does not need to store global routing state and generates low control overhead. In order to evaluate the performance, we have conducted the simulation by comparing with representative routing protocols in 6LoWPAN. The results show the overall performance of TAMR surpasses others in terms of hop count, packet delivery ratio, number of control packets, and memory usage.

Many wireless sensor network applications require sensor nodes to be deployed on the ground or other surfaces. However, there has been little effort to characterize the large- and small-scale path loss for surface-level radio communications. We present a comprehensive measurement of path loss and fading characteriztics for surface-level sensor nodes in the 400 MHz band in both flat and irregular outdoor terrain in an effort to improve the understanding of surface-level sensor network communications performance and to increase the accuracy of sensor network modeling and simulation. Based on our measurement results, we characterize the spatial small-scale area fading effects as a Rician distribution with a distance-dependent K-factor. We also propose a new semi-empirical path loss model for outdoor surface-level wireless sensor networks called the Surface-Level Irregular Terrain (SLIT) model. We verify our model by comparing measurement results with predicted values obtained from high-resolution digital elevation model (DEM) data and computer simulation for the 400 MHz and 2.4 GHz band. Finally, we discuss the impact of the SLIT model and demonstrate through simulation the effects when SLIT is used as the path loss model for existing sensor network protocols.

One of the important challenges in the Internet of Things (IoT) is how to acquire the physical context of things. IP-based wireless sensor networks (IP-WSNs) could be a promising approach to collecting the physical context of things and to integrating WSNs to the Internet. However, realizing IP-WSNs in IoT exposes two major challenges. One is how to embed the Internet Protocol (IP) in resource-constrained sensor nodes. The other is how to achieve real-world deployment of WSNs and its integration with the Internet on the fly and on the cheap. In this paper, we present the SNAIL (Sensor Networks for All-IP World) project and introduce a new type of IP-WSN gateway, which supports dual wireless access points for WiFi and IP-WSN, enabling deployment of SNAIL nodes in an easy and rapid manner as for the solution. To show the proof-of-concept, we implement a new SNAIL platform from tiny sensor nodes to a gateway.

Many opportunistic spectrum access MAC protocols for cognitive radio networks have been proposed to allow coexistence of both primary and secondary users in licensed bands. However, most of the proposed approaches are based on common control channel that may become a bottleneck for data transfer, as well as be prone to jammer or primary user activity. We propose a coordinator-based cognitive radio network and an opportunistic spectrum access MAC protocol using slow hopping, SH-MAC, to achieve robustness to primary user or jammer activity as well as aggregate throughput improvement. The coordinator considers primary user activity identification, spectrum sensing, channel hopping scheduling, and network time synchronization. The proposed MAC protocol includes mechanisms such as network joining, two-level CS-MACA operation, on-time channel switching, returning to common hop to run in the slow hopping and coordinator-based network environment. Furthermore, the proposal is devised for secondary users to operate with only one transceiver that has non-negligible switching overhead. We validate our protocol using simulation. Simulation results show SH-MAC efficiently increases network capacity regardless of PU activity over the licensed channels.

Wi-Fi interface is one of the predominant energy consumers in Wi-Fi stations. Despite many researches on Wi-Fi energy management, energy wastage of Wi-Fi stations resulting from network contention among multiple access points(APs) has not been widely investigated. In this paper, we analyze the network contentions occur among multiple APs, and show that Wi-Fi power save mode performance could be severely affected by network contentions. In order to overcome the network contention problem, we propose a scheduling policy, Tame, to assign access points into different sub clusters, in each of which none of the access points have network contentions and data can be transmitted simultaneously without collision. Access points properly control Wi-Fi stations to switch states between sleep and active to avoid stations' packet receiving time overlapping, while the energy wastage from network contentions is mitigated and network performance is guaranteed. We simulate Tame in Qualnet and conclude that Tame, compared with the related work Sleep Well, enhances the average throughput of Wi-Fi stations by an average of 13% for CBR traffic. Simultaneously, the corresponding Wi-Fi interface energy is consumed more efficiently.

Recently, a pointer forwarding scheme (PFS) was proposed to reduce location update overhead in wireless mesh networks. Using the PFS, location update is replaced with a simple forwarding pointer setup between two neighboring MRs until the forwarding chain K is less than or equal to the configurable parameter Kmaxfor each MC. However, in PFS, if the two MRs are not one hop neighbors, the PFS fails to set up a forwarding pointer, and a location update event must be triggered, increasing location update overhead. To improve PFS, we present a novel location update scheme called a multi-hop pointer forwarding scheme (MPFS). The MPFS allows forwarding pointers to be constructed over multi-hop at once even if MRs are not one hop neighbor with each other. The key to achieving this lies in the logical tree distance constructed during network formation. The tree distance is used to relay forwarding pointer packets over multi-hop links without additional control overhead and to estimate hop distance between two MRs. By doing so, the MPFS improves the probability of success in forwarding pointer setup while ensuring K ≤ Kmax, resulting in lowering the location update overhead. To evaluate our scheme we implement both the PFS and the MPFS in ns-2 and compare them. As a result, we show that the MPFS significantly reduces the number of location update events, resulting in reduction of location update delay and signaling overhead, and packet losses during location updates.

Tree-based routing protocols in low-rate wireless mesh networks usually have the detour problem in return for the no route discovery overhead. In this paper, we propose a novel algorithm, named Incremental Path Reduction (IPR), which incrementally shortens inefficient detoured path as more data packets are delivered. In IPR, data packets are delivered along the tree route in use by using 1-hop broadcast, enabling neighbor nodes to learn about the data packets' hop count. Using the hop counts, each node estimates their residual hop count to destination. As a result, each forwarder selects next hop node that has small residual hop count. In this way, IPR incrementally shortens the detoured route as more data packets are delivered. To verify our algorithm, we applied IPR to the representative tree routing protocols, and evaluated the path stretch and packet delivery ratio as well as control packet overhead. Simulation results show that IPR significantly enhances the overall routing metrics for any types of tree-based routing protocols.

In this paper, we propose a novel on-demand routing protocol for IEEE 802.15.4-based low rate wireless mesh networks (LRWMNs). The novelty of our work is that the proposed routing algorithms enable route discovery packets (RREQs) to be semi-directionally flooded with respect to a source-destination pair in a distributed manner without either physical or virtual location systems. Instead, we use the logical tree distance derived from the hierarchical addressing structure. The idea behind this is that nodes with longer tree distance to the destination than that of the source-destination pair are not involved in flooding RREQs. To evaluate the algorithms, we apply our algorithms into AODVjr and carry out extensive simulations while comparing the algorithms with AODVjr using network-wide flooding (NWF). Results show that the proposed algorithms drastically reduce RREQ overhead while providing comparable or better performance than NWF in AODVjr.

To realize the Internet of Things (IoT), an important step would be to allow things and information about them to be accessible in an easy way and a platform-independent manner. As World Wide Web (WWW, Web) shows explosive growth over the last decade, the web is the most popular and familiar user interface to acquire knowledge about everyday life. In this respect, web browsing through standard protocols like HTTP over TCP could be a promising approach to access things and to collect things' information. In future, IP-based wireless sensor networks (IP-WSNs) are expected to be integrated into the IoT. However, due to the characteristics of IP-WSN such as resource limited and low data-rate, the challenges that we confronted are 1) how to make clients access IP-WSNs as if they use regular web browsers with rich presentation and acceptable response time and 2) how to embed the standard web in such resource-constrained sensor nodes. In this paper, to address these challenges, we introduce a new browsing architecture, which enables the exploration of the IP-WSNs through a conventional web browser and the navigation with rich user interfaces. We also design lightweight web protocols, lwHTTP and lwTCP, with header compression in order for HTTP and TCP to run on the resource-constrained IP-WSNs. To show the feasibility, we implement the proposed architecture in SNAIL (Sensor Networks for an All-IP world), our IP-WSN platform, and we evaluate the performance of proposed browsing architecture.

As wireless sensor networks are becoming more and more commercialized in many applications such as smart home network, building automation system, the needs for network scalability are increasing to support several hundreds of nodes and massive amount of data from them. In this paper, we propose the hierarchical network protocol (HNP) to provide the efficient communication, network reliability, and network management as well as network scalability.

For adopting wireless sensor network in consumer networks, easy configuration and standard based interoperable operations are important. General device-controller approach in service discovery protocols like UPnP allows sensor networks to be discovered and accessed as general UPnP devices via gateways. However, users also need standard capability information that fully describes sensor types, attributes, operations, and calibration to utilize sensor nodes. Integration of IEEE1451 architecture and UPnP network enables self-identification of sensor nodes and "plug and play" capability. In this short paper, we present a management system based on integration of IEEE1451 sensor networks and UPnP.

Service gateways (SGs) are a promising approach to integrate wireless sensor and actuator networks (SANETs) with pervasive infrastructure over Internet by encapsulating distributed physical sensor and actuators into regular objects. Typically, SGs for SANETs are assumed to know operational environment in advance at design time with limited awareness of operational SANET conditions and, in the meanwhile, provide a single particular form of access without knowledge of application activities. As a SANET evolves, end users will frequently add sensor and actuator nodes (SANs) for additional services, remove defective or obsolescent SANs, and upgrade SANs in deployed SANETs. Therefore, traditional SGs are infeasible to integrate the future evolvable SANET (ESANET) with the pervasive infrastructure. In this paper, we introduce the reflective service gateway (RSG) architecture combined with ESANET management functions. RSG enables heterogeneous middleware systems to concurrently access ESANETs. In addition, by reflection mechanism where network management functions are explicitly reified and integrated into RSG so that it can learn the changing ESANET runtime environment, RSG self-adapts to ESANET dynamics and enables ESANETs to customize its activities according to application services involved. To validate the RSG, we have implemented the RSG and conducted empirical performance measurements with home sensor and actuator applications using ZigBee.

Mobility management is one of the most important research issues in 6LoWPAN, which is a standardizing IP-based Wireless Sensor Networks(IP-WSN) protocol. Since the IP-WSN application domain is expanded to real-time applications such as healthcare and surveillance systems, a fast and seamless handover becomes an important criterion for mobility support in 6LoWPAN. Unfortunately, since existing mobility protocols for 6LoWPAN have not solved how to reduce handover delay, we propose a new fast and seamless mobility protocol to support inter-PAN handover in 6LoWPAN, named inter-MARIO. In our protocol, a partner node, which serves as an access point for a mobile node, preconfigures the future handover of the mobile node by sending mobile node's information to candidate neighbor PANs and providing neighbor PAN information like channel information to the mobile node. Also, the preconfigured information enables the foreign agent to send a surrogate binding update message to a home agent instead of the mobile node. By the preconfiguration and surrogate binding update, inter-MARIO decreases channel scan delay and binding message exchange delay, which are elements of handover delay. Additionally, we define a compression method for binding messages, which achieves more compression than existing methods, by reducing redundant fields. We compare signaling cost and binding message exchange delay with existing mobility protocols analytically and we evaluate handover delay by simulation. Analysis and simulation results indicate that our approach has promising fast, seamless, and lightweight properties.

In this paper, we propose a novel dynamic phase shift (DPS) scheme for energy efficiency in multi-hop wireless sensor networks. The DPS algorithm allows not only a receiver to shift its wake-up schedule dynamically to a sender's wake-up schedule, but also a sender to shift its wake-up schedule dynamically to a receiver's wake-up schedule to reduce the energy waste that occurs during a rendezvous period of asynchronous duty cycle scheme. Based on the DPS algorithm, we also propose DPS-MAC, in which a collision avoidance scheme and a delay reduction technique have been included. Furthermore, we design the DPS-MAC to cooperate with the network layer for energy saving of the overall network. Simulation results show that the proposed DPS-MAC achieves better energy efficiency and improved latency compared to a conventional asynchronous MAC protocol.

Service-oriented computing which provides flexible composition of various applications using multiple reusable services has getting more attractive. We propose a service oriented wireless sensor networks toolbox, which encapsulates complexities of sensor networks and enables simple service compositions using an intuitive GUI. We utilize sensor networks as a collection of services that are discovered and coordinated by users. Service metadata which are self-describing service capabilities and interfaces are defined and exposed via our service metadata repositories. A service oriented sensor network toolbox consisting of sensor networks, an intuitive GUI application that retrieves metadata for discovered services and helps users to composite various roles are presented.

Wireless Sensor Networks (WSNs) enable the surveillance and reconnaissance of a particular area with low cost and less manpower. However, the biggest problem against the commercialization of the WSN is the limited lifetime of the battery-operated sensor node. Taking this problem into account, a mobile sink is deployed as a robot, vehicle or portable device to only activate the sensor nodes that are interesting to the sink and leaves other nodes deactivated for. This can considerably extend the lifetime of the sensor nodes compared to existing power management algorithm of using all nodes. However, in this environment, the mobility of the sink raises new issues of energy efficiency and connectivity in communications. To solve these issues, we propose a DRMOS (Dynamic Routing protocol for Mobile Sink) method that includes a designated wake-up-zone to make sensor nodes prepare for an incoming sink. The shape of the wake-up-zone is dynamically changing to reflect the past moving patterns of the sinks. Moreover, we present the extensive simulation results and recommend parameters for practical use of DRMOS from the simulation analysis.

Established service discovery protocols such as UPnP allow control points to find devices and services and to retrieve descriptions about them in order to learn all about the device and services. However, general device-control point approaches are not suitable for wireless sensor networks due to se

Context-aware systems have traditionally used distributed sensors to gather context information. The unique identity provided by radio frequency identification (RFID) tags could provide additional information to the sensor data. However, the task of matching identity and sensor information in the same context is not trivial. By placing wireless sensor nodes in the same RFID tagged objects, we can build distributed wireless networks that collaborate to produce uniquely identified context. In this paper, we introduce a set of network protocols that dictate the formation of context-specific wireless sensor networks (WSNs). One of the main goals of our protocols is to maintain an energy-efficient WSN, in which network members can join and leave the network in a dynamic, transparent way. Simulations show that our design provides important improvements in network lifetime and that it is specially suited for dynamic environments. Additionally, we describe our implementation experience and provide a set of tools for creating and evaluating our application.

In an integration system of RFID and wireless sensor network (WSN), RFID is used to identify objects while WSN can provide context environment information of these objects. Thus, it increases system intelligent in pervasive computing. We propose the EPC sensor network (ESN) architecture as an integration system of RFID and WSN. This ESN architecture is based on EPCglobal architecture, the de facto international standard for RFID. The core of ESN is the middleware part which is also implemented in our work. In this paper, complex event processing (CEP) technology is used in our ESN middleware which can handle large volume of events from distributed RFID and sensor readers in real time. Through filtering, grouping, aggregating and constructing complex event, ESN middleware provides a more meaningful report for the clients and increases system automation.

In this paper, we present RFMS, the real-time flood monitoring system with wireless sensor networks, which is deployed in two volcanic islands Ulleung-do and Dok-do located in the East Sea near to the Korean peninsula and developed for flood monitoring. RFMS measures river and weather conditions through wireless sensor nodes equipped with different sensors. Measured information is employed for early-warning via diverse types of services such as SMS (short message service) and a Web service.

A body sensor network (BSN) is a network of sensors that sense vital signs of a human body. The vital signs can be divided into two categories based on significance to the life: critical and non-critical signs. The critical vital signs, or real-time messages (RTMs), should be transmitted within the deadlines, and they must not be collided. On the other hand, the non-critical signs, or non-real-time messages (NRTMs), are required to be delivered with the best effort. In this paper, an on-demand MAC (OD-MAC) is proposed for BSN to provide real-time transmission, collision avoidance, and energy efficiency. The protocol is developed on IEEE 802.15.4 standard, and some specifications are modified to support those requirements. The OD-MAC changes the superframe structure dynamically to schedule RTMs and NRTMs. Those messages request bandwidth, and the network coordinator adaptively changes beacon interval and allocates slots to the messages. For evaluating the performance, duty cycle according to required utilization, delivery ratio, and energy efficiency are analyzed with simulations using NS2.

This paper shows IEEE 802.15.4a CSS-based localization system for wireless sensor networks. IEEE 802.15.4a CSS technology can provide high accurate ranging functionality to a sensor node. However, as it measures a distance based on Time-of-Flight(TOF) of RF signal, the system needs well designed ranging and report protocol. In this paper we show our ranging protocol and location calculation server architecture.

Recently, the RFID technology has become essential for ubiquitous computing. As the deployment of mega SCM (supply chain management) environments starts at the largest companies in the distribution industry, efforts tend to concentrate on a variety of performance improvements for the RFID middleware aiming to give quality of service for large RFID-data transmission. Web services, cornerstone of the RFID networks, require high performance, security and extensibility. Since SOAP (simple object access protocol) inherits the poor performance of XML, it is not easy for the RFID middleware to support high performance web services. The ALE (application level event) communication interface from the RFID middleware sends a response message. Its serialization, which includes the conversion of common language runtime objects to the XML documents and streams and packing of this data into a message buffer, has been proven as the bottleneck for SOAP's poor performance. In this paper, we propose ALE-TSOAP, based on ALE templates, that provides an increase in the performance of the RFID middleware when generating response messages. We analyze SOAP messages to classify the various ALE template formats, to design and implement our ALE templates and, finally, to evaluate the performance of the ALE-TSOAP processing. The use of ALE-TSOAP does not change the SOAP protocol or the ALE communication interface of the RFID middleware. Through our experiments, we observed that our approach obtains up to a 197.8% performance gain by only using ALE templates for the serialization of SOAP message.

Unlike relative location information, obtaining absolute location in multi-hop way usually requires anchor nodes to periodically flood the network. However, due to the limited resource of sensor nodes, this causes significant reduction in sensor nodes lifetime. In this paper, we propose a localization system called Mosaic. As the name implies, by matching pieces of the puzzle, which plays similar role with the relative locations, Mosaic provides absolute location with significantly reduced packet transmissions. Our simulation results show significantly reduced packet transmissions of Mosaic while not losing accuracy compared with flooding-based localization systems.

WLAN-based mesh networking is being developed with the benefit of easy deployment and easy configuration to overcome the limitation of a single-hop communication in terms of coverage, cost, capacity, scalability, etc. Wireless mesh networks(WMN)' distributed and dynamic characteristic needs an distributed authentication scheme which supports the mobility of the mesh points(MP). In this paper, we propose an efficient distributed authentication method which significantly eases the management burden, reduces the storage space on mesh points, and support the mobility of the MP.

In this paper, we propose the cluster-based hierarchical time synchronization protocol (CHTS) for wireless sensor networks (WSN). To couple the network with real world tightly, we provide a network-wide time synchronization through the introduction of abstractions for cluster and hierarchy. Generally the error of time synchronization has the tendency of being accumulated as the hop count along the synchronization path increases in multi-hop wireless sensor network. Thus we cluster the randomly deployed nodes to decrease the hop count along the synchronization path. We then apply different optimal synchronization mechanisms to the cluster heads and cluster members for the sake of power efficiency through an explicit broadcast manner. We prove the performance enhancements with simulation comparisons.

Recently, the availability of cheap and small micro sensor node and low power wireless communication give a contribution of enhanced developments of wireless sensor network application in real society. Furthermore, middlewares and operating systems for convenience on development of sensor network application are essentially needed. In this paper, we introduce a sensor network operation system, Nano-Qplus platform, which is flexible, dynamic, and easy manageable for sensor network application programmers. Furthermore, for the purpose of performance evaluation, we compare Nano-Qplus to other sensor network operating systems related to memory read/write time and task creation latency. The results of performance analysis shows that Nano-Qplus offers enhanced advantages that other sensor network operating systems, so we can notice Nano-Qplus is easily applied to real sensor network applications

The wireless and resource-constraint nature of a sensor network makes it an ideal medium for attackers to do any kinds of vicious things. In this paper, we describe PLUS, a parameterized and localized trust management scheme for sensor networks security, where each sensor node maintains highly abstracted parameters, rates the trustworthiness of its interested neighbors to adopt appropriate cryptographic methods, identify the malicious nodes, and share the opinion locally. Results of a serious of simulation experiments show that the proposed scheme can maximize security as well as minimize energy consumption for sensor networks. And also, the secure routing proposed based on PLUS shows its benefit and feasibility

Applications and services in ubiquitous environment are not static but are joined and separated dynamically. Hence dependency of applications and authentication of users to services are significant. Existing ubiquitous middleware are structured for applications to easily cope with dynamic contexts but do not authenticate users; leaving services exposed to security threats. This paper suggests platform requirements that satisfy security features in ubiquitous environment and means to regulate service authorizations using RBAC.

Existing range-based localization algorithms are superior only when a high accuracy node-to-node measured distance exists. This assumption is actually difficult to satisfy with current ranging techniques used in tiny sensor nodes. Meanwhile, range-free localization algorithms work independently of ranging error but can only produce limited node accuracy. In this paper, we propose a novel localization scheme that uses a learning-based distance function to estimate distances. The adaptation of distance function to ranging error and other network conditions, i.e., network density, number of anchor, results in better estimated distances. This leads to more accurate position calculation comparing to existing works, especially when ranging error is high.

We propose the special kind of the mobility so called sub-network mobility for wireless sensor networks. In the sub-network mobility, the strongly coupled nodes, which are attached to the same object, move together passively as the object moves. Strongly coupled nodes require a different mobility scheme from the conventional mobility mechanism due to the characteristic that the relative positions among them are not changed while they are moving following the object. Thus, we research into how to manage strongly coupled nodes and suggest a sub-network mobility scheme which uses a sub-network gateway to relay the packets from the nodes to the main-network. With mathematical modeling, we found the condition in which the proposed method outperforms the method which doesn't support sub-network mobility. The overheads of the proposed algorithm and the packet transmission performance are evaluated with ns-2 simulator.

Since the lifetime of a battery directly impacts the lifetime of sensor networks, one of the key considerations in the design of sensor networks is the ability to maximize battery lifetime. In this paper, we present (1) a task modeling methodology and (2) a battery-aware real-time task scheduling technique for sensor networks. The task modeling is achieved based on task classification in terms of the usage of resources on a micro-sensor system. For exploiting the recovery effect of battery, the battery-aware task scheduling algorithm composed of three phases is designed to maximize the lifetime of a battery while meeting the timing constraint of each task.

There are many trials to provide an optimal route for minimum energy consumption in a wireless sensor network. Currently, however, the mechanisms to find minimum energy property graph (MEPG) do not properly take into account the efficiency in time and storage, the optimality in results, and the feasibility in real systems. In this paper, we propose an efficient and first optimal algorithm to find the MEPG, in which all minimum energy paths are included, not only significantly reducing its total number of edges, but also obtaining an optimal result in O(VlogV+E). We also develop power aware data-centric routing protocol characterized by minimum energy consumption and longer lifetime.

We propose a real-time message scheduling algorithm which is applied to schedule periodic real-time messages in IEEE 802.15.4 for LR-WPAN(Low Rate Wireless Personal Area Network). The standard allows GTSs (Guaranteed Time Slots) in the optional use of a superframe structure in a beacon-enabled network to be used to exchange real-time messages. To utilize these features of the standard efficiently, a proper message scheduling algorithm is needed. The proposed off-line message scheduling algorithm which is based on a distance constrained scheduler generates the standard specific parameters including BO, SO, and GTS information to schedule the given message set. The algorithm is evaluated by simulation and the guaranteed time service using the schedule is implemented and evaluated on CC2420DB which is a prototyping platform including an IEEE 802.15.4 compliant transceiver of Chipcon AS.

With low-power consumption, small code and data size, evolvability as design criteria, we develop an evolvable operating system (EOS) for wireless sensor network applications. The EOS provides memory space efficient thread management, collaborative thread communication model and network stack. It also supports power management of microcontroller and radio transceiver, and network wide time synchronization function. Above all, the most important feature is the concept of evolvability with which the operating system itself can be easily configurable and upgradeable.

To facilitate the integration of real-time applications in a common platform, temporal and spatial partitioning should be provided. A strongly partitioned integrated real-time system (SPIRIT) is reported in this paper that adopts a two-level hierarchical scheduling mechanism to ensure temporal partitioning. At the lower level, multiple partitions (applications) are dispatched under a cyclic scheduling, whereas, at the higher level, multiple periodic tasks of a partition are scheduled within the partition according to a fixed priority algorithm. The proposed Distance Constraint guaranteed Dynamic Cyclic (DC2( scheduler applies three basic operations, left-sliding, right-putting and compacting, to dynamically schedule aperiodic tasks and, in the meantime, guarantees the distance constraint characteristics of a partition cyclic schedule. In addition, the slack time calculation of these dynamic operations can be applied for scheduling hard aperiodic tasks. With simulation studies, we observe that the DC2 algorithm can result in a significant performance enhancement in terms of the average response time of soft aperiodic tasks and the acceptance rate for hard aperiodic tasks.

Ethernet technology has received much attention in embedded system industries because of its cost efficiency, high availability, and popularity. This trend is not an exception even in safety critical real-time systems such as integrated modular avionics systems. To overcome the lack of deterministic characteristics in the Ethernet protocol, we propose a software-oriented approach based on table-driven proportional access. In addition to the protocol details, performance and schedulability analyses, as well as a prototype platform, are described.

In the recent development of avionics systems, integrated modular avionics (IMA) is advocated for next generation architecture that needs integration of mixed criticality real-time applications. These integrated applications meet their own timing constraints while sharing avionics computer resources. To guarantee timing constraints and dependability of each application, an IMA-based system is equipped with the schemes for spatial and temporal partitioning. We refer the model as SP-RTS (strongly partitioned real-time system), which deals with processor partitions and communication channels as its basic scheduling entities. This paper presents a partition and channel-scheduling algorithm for the SP-RTS. The basic idea of the algorithm is to use a two-level hierarchical schedule that activates partitions (or channels) following a distance-constraints guaranteed cyclic schedule and then dispatches tasks (or messages) according to a fixed priority schedule. To enhance schedulability, we devised heuristic algorithms for deadline decomposition and channel combining. The simulation results show the schedulability analysis of the two-level scheduling algorithm and the beneficial characteristics of the proposed deadline decomposition and channel combining algorithms.