Nowadays, some decentralized anonymous payment (DAP) systems have emerged to protect transaction privacy. However, these systems may be strictly limited by certain governments due to the difficulty of tracking financial flow, which severely restricts their circulation. Although several auditable DAP (ADAP) systems permit authorities to monitor transactions, the potential for privacy violations instead makes users hesitate to apply them. As a result, it becomes a dilemma. Thus, this paper proposes an ADAP system that follows the least-privilege principle, called LP-ADAP, to strike a balance between auditability and privacy preservation. First, LP-ADAP employs a consensus audit mechanism to prevent transactions from being investigated by a single malicious auditor. In addition, the proposed method adheres to the least-privilege principle, strictly limiting auditors to only the minimum necessary permissions, thus limiting the privacy leakage of the senders and receivers. Finally, auditors in LP-ADAP can conduct offline audits without participating in the transaction phase, reducing their burden. Experiment results show that LP-ADAP outperforms existing mechanisms in transaction generation and verification. Moreover, the transaction size of LP-ADAP is mostly smaller than that of other methods, even when the number of auditors increases. Consequently, LP-ADAP is the most practical and trustworthy DAP system.

Fog computing is promising for Industrial Internet of Things (IIoT) which are typically large-scale, time-critical, and geo-distributed. However, it needs to address challenges such as scalability and heterogeneity in application provisioning with diverse and strict Quality-of-Service (QoS) requirements from IIoT devices. In this paper, a fog service orchestrator, Q-FSO, is proposed for IIoT application provisioning in which request satisfaction is maximized. With a two-level QoS model that considers key performance metrics for IIoT, including availability, reliability, response time, and cost, Q-FSO can realize large-scale service orchestration by exploiting the advantages of fog computing. Further with a harmonious local QoS assignment model, Q-FSO solves the workflow construction problem by multiple independent local optimization tasks, one for each stage, for the concurrent requests. Each local optimization task can be reduced from the multiple multidimensional knapsack (MMKP) problem. We then propose two practical MMKP heuristic algorithms, namely, Incremental Similarity Matching (ISM) and Greedy Multiple Matching (GMM), to tackle the local QoS assignment problem with polynomial time complexity. The scalability issue in orchestration is then handled by a decentralized workflow construction protocol. The simulation results validate their applicability, demonstrating that Q-FSO is indeed an efficient and effective IIoT application provision mechanism with significant performance benefits in terms of service processing throughput, resource utilization, and speed of service orchestration.

IoT data economy is an emerging concept since sensing data are widely applied in various IoT services. To maintain the stability of data economic system, this paper proposes the Reliable IoT Data Economic System (RIDES) to provide IoT data a reliable, automatic, and intelligent trading environment. Applying the smart contract technique, IoT data in RIDES can be automatically traded without human intervention. As the trading procedure is performed based on blockchain, all transactions can be protected against tampering. Also, the RIDES provides the Smart Decision Making Module for users to develop practical pricing strategies. Thus, a seller can derive a price with the highest expected revenue for its data and a buyer can purchase its desired data referring to cost-performance ratio or data owner's trustworthiness. Even if malicious sellers cheat buyers, the latter can invoke trade dispute procedure to refund their payment, which makes the trading environment more reliable. As the blockchain-based operations usually cost much, the RIDES adopts the edge computing architecture to alleviate the burdens of IoT devices. Finally, the RIDES is implemented in IoT environments. The case study demonstrates that the transactions on IoT data can be performed correctly by RIDES.

To promote independent living and elderly care in a smart living space, monitoring the activities of daily living and detecting upcoming critical situations are crucial. Thus, exploiting the spatial aspect related to the movement of the resident and their relations to a specific region is important. However, most of existing solutions do not consider related spatial aspect or it is incorporated in a very limited way, in which most solutions need to know the deployment location of sensors. To achieve this, a very time consuming sensor labeling process is needed. In this paper, an unsupervised spatial information extraction technique for smart living space is developed. This technique establishes an abstract spatial representation called virtual location for a physical living space without knowing the actual deployment location of the sensors in that space. A virtual location is achieved by utilizing our extended Louvain modularity method to extract key spatial features from the hidden correlations in the observations perceived through the sensors. By considering the characteristics of different sensors, the effectiveness of virtual location can be further improved. Two real world datasets are used to validate the applicability of the propose technique, and the experimental results for them achieve 90.3% and 71.7% V-measure values respectively.

Climate control is vitally important to the operation of greenhouses. However, many conventional methods for controlling greenhouse climate are not efficient enough since they are either based on an on-off control method or a proportional control method. To maintain a steady climate, an intelligent switching control strategy for greenhouse climate control system was designed using time series to analyze historical data from greenhouse climate control system and predict target attribute's value to regulate climatic conditions such as suppress or raise the environmental factor based on prediction results. The simulated results showed that the greenhouse climate can be stabilized and maintained within the ideal range by the proposed control strategy.

Internet of Things (IoT) is an emerging network technology applied to provide various services in our daily life. Generally, IoT environments are composed of numerous heterogeneous devices with constrained resource. The limited capability of IoT devices makes it impractical to perform traditional security mechanisms, and thus IoT services are usually vulnerable to all kinds of security threats, such as impersonation and forgery attacks. Moreover, the inflexible protection provided by these security mechanisms leads to inefficiency because different services haves diverse requirements. To provide IoT services suitable security protection, Multi-graph Zero-knowledge-based Authentication System (M-ZAS), which is not only light-weight but also high-adaptive, is proposed. Compared to traditional authentication mechanisms as well as other Zero-knowledge-proof (ZKP) methods such as GMW-ZKP, M-ZAS provides higher performance and better security protection. In addition, M-ZAS has lower transmission overheads than GMW-ZKP does. Considering relevant contexts as parameters, M-ZAS provides adaptive protection to fulfill what users actually need. Experiment results show that M-ZAS is 3 times faster than GMW-ZKP and even 7 times than traditional authentication mechanisms in IoT devices. Also, M-ZAS reduces 3 times network traffic than GMW-ZKP. Thus, the proposed M-ZAS is the most practical authentication system in IoT environments.

A traffic‐aware routing protocol (TARCO) that considers integrated real‐time traffic conditions for constructing delivery paths over a vehicular environment is presented. This goal was achieved through three mechanisms: cooperative traffic information collection, lightweight real‐time road scoring and adaptable data route construction. An accurate overall view of the real‐time traffic conditions was obtained from the cooperation between information aggregators at junctions and information collectors on road segments. Each road segment was then assigned a weight according to the overall view of the traffic conditions and updated periodically to reflect traffic variations. Finally, the road segments providing efficient and reliable data routes were used to construct a routing path with guaranteed connectivity and a short delivery delay to the destination. Simulation results showed that the use of TARCO leads to high network performance in terms of the packet delivery ratio, end‐to‐end delay and communication overhead.

In this paper, we presents a routing protocol called Enhanced Hybrid Traffic-Aware Routing (EHTAR) to cope with the dynamics of Vehicular Ad Hoc Network (VANET) in urban environment. In EHTAR, an elected functional node called Junc-Tracker located at each junction is used to explore the real-time vehicular and network traffic information in each road segment. This information enables each Junc-Tracker to determine a reliability score for each road segment and assist with the selection of a more robust and efficient routing path during transmissions. The reliability score of each junction is updated periodically to adapt to the dynamics in the environment and recalculate these reliability scores if necessary. In addition, we also adopt an enhanced next forwarder selection scheme to make better decisions on how to select a more efficient and stable next forwarder by using location prediction and link duration estimation to enhance and maximize the performance of EHTAR. Simulation results shows EHTAR not only achieves a higher delivery ratio and transmission throughput, but also out performs other position-based routing protocols that adopt the traditional greedy forwarding strategy.

Data dissemination is an essential application in the opportunistic network which has received considerable attention recently. Delivering data effectively and rapidly is the major target of data dissemination. However, many scenarios ask data to be sent to specific locations at designated time. For example, the information of bargain news is useful in shopping streets during business hours. Unfortunately, existing dissemination routing protocols suffer from huge transmission overheads. Moreover, due to the limited capability of devices, context-based routing methods are also inapplicable since their computation complexity is high. This paper therefore presents a novel timesensitive data dissemination scheme (TSDD) composed of three major components including node location prediction, relay node selection, and successor selection. By exploiting some accessible information, TSDD can deliver time-sensitive data efficiently. Simulation results show that TSDD provides higher delivery rate with less transmission overheads than existing methods.

Type reduction of interval type-2 (IT2) fuzzy sets is essential in conducting the type-2 fuzzy sets expressed with the resolution forms of IT2 fuzzy sets. Several type reduction methods, such as KM, EKM, and centroid flow, have been proposed. These methods are relatively easy to implement but still computation-intensive because they need to invoke an iterative switching point finding procedure. This study derives a theorem and proposes a heuristic algorithm, which can fast and accurately identify the minimum and maximum switching points of a piecewise smooth IT2 fuzzy set. It also demonstrates that it is easy to derive the close-form expressions of the switching points of a piecewise smooth IT2 fuzzy set if both of its upper and lower membership functions can be parameterized as LR fuzzy sets, which are defined in this paper. Then, the type reduction of piecewise smooth IT2 fuzzy sets can be simplified to solve the close-form expressions of their switching points in terms of LR parameters. Experiments on IT2 fuzzy sets with various piecewise smooth membership functions, including linear, Gaussian, and hybrid-shaped ones were made. The results showed that the proposed type reduction method can obtain solutions which accurately approximate to the desired switching points with much lower computational overhead than the Karnik-Mondel (KM) and enhanced KM (EKM) methods.

Rendezvous is an essential operation for secondary users communicating in a cognitive radio network (CRN). Because it is infeasible to employ a common control channel in a licensed spectrum, blind rendezvous is an expected solution and generally estimated by the criteria of asynchronous rendezvous support, guaranteed rendezvous, asymmetric model support, heterogeneity of roles, and short time-to-rendezvous (TTR). Most existing blind rendezvous methods fail to fully satisfy these criteria or have considerable TTR significantly increased with the number of channels. This paper presents a new alternate hop-and-wait channel hopping method, called E-AHW, which can support fast blind rendezvous. By assigning each secondary user a unique alternate channel hopping sequence composed of HOP/WAIT-mode subsequences, all blind rendezvous criteria can be satisfied with shorter TTR. According to simulation results, the performance of E-AHW is better than existing methods. Moreover, using E-AHW, the average value and standard deviation of time intervals between two rendezvous are also less than other methods. Thus, E-AHW can provide an efficient and stable communication environment.

In this paper, we present a novel multimodal emotion recognition technique that automatically learns decision-making parameters customized for each modality. Specifically, the process of decision-making is implemented in a multi-stage and collaborative fashion: Given a classifier for single modality, the classifier is regarded as a virtual expert since classification methods can make emotion recognition in accordance with certain expertise. Then, in the reputation equalization, the expert's classification capability is then quantitatively equalized to assure the reputation and/or confidence for each expert. To compromise decisions among experts, the final decision is obtained by calculating the weighted-sum of all the equalized reputation quantities, in such a way that the decision of one expert can be made in collaboration with that of the others. Moreover, to learn the proposed model parameters, the genetic algorithm is tailored and applied to alleviate the local minima problem during the process of finding an optimal solution. The experimental results have shown that the proposed collaborative decision-making model is effective in multimodal emotion recognition.

Cognitive Radio (CR) is an emerging technology developed to improve the utilization of licensed spectra, and a promising solution to alleviate the spectrum shortage problem. In cognitive radio networks (CRNs), if users want to establish communication links with neighbors, they need to rendezvous on an available channel. However, it is infeasible to employ a common control channel (CCC) in a licensed spectrum, because the CCC will be congested by primary users and degrade the performance of CRNs by the single point of failure problem. Therefore, the channel hopping (CH) based methods without CCC support are usually adopted to establish rendezvous in CRNs. This kind of rendezvous is called blind rendezvous and generally estimated by the following criteria: 1) asynchronous rendezvous support, 2) guaranteed rendezvous, 3) asymmetric model support, 4) multi-user/multi-hop scenario support, and 5) short time-to-rendezvous (TTR). Most existing blind rendezvous methods fail to fully satisfy these criteria or have considerable TTR which is significantly increased with the number of available channels. In this paper, the alternate hop-and-wait CH method is proposed to solve these problems. Furthermore, each user has a unique alternating sequence of HOP and WAIT modes. The theoretical analysis results have confirmed that the proposed method satisfies all evaluation criteria mentioned above and provides shorter TTR. According to simulation results, the performance of the proposed method even is two times better than Jump-Stay algorithm (JS), which is the most efficient blind rendezvous method as we know, under the asymmetric model which is much critical for CRNs.

In this paper, we present a routing scheme called Junction-Based Traffic-Aware Routing (JTAR). This scheme dynamically elects a non-static vehicle at each junction to function as a temporary access point and adopt a traffic information collection mechanism to obtain real-time traffic and network information. Based on the collected information and the degree by which the outgoing roads of each junction are diverged from the destination, we are able to determine a weighted score for each road and select the most reliable and efficient route for data packets to reach the destination. In addition, we also introduce a probe-merge method in the traffic information collection mechanism. This method not only reduces the overheads produced by the active probing method used in the mechanism, but also decrease the probing distance required to obtain information from neighboring junctions. The simulation results show that the proposed scheme achieves a higher delivery ratio and lower end-to-end delay.

It is easy for a multi-layered perception (MLP) to fit a stratified spatial interpolation pattern whose form is close to open surface; while it is easy for a radial basis function network (RBFN) to fit a pocket (radial) spatial interpolation pattern whose form is close to closed surface. However, in the real world, the spatial interpolation pattern may consist of stratified and pocket patterns. Neither MLP nor RBFN can fit the pattern easily. To combine their advantages to fit the complex hybrid spatial interpolation patterns, in this article we propose a novel neural network, MLP–RBFN hybrid network (MRHN), whose hidden layer contains sigmoid and Gaussian units at the same time. Although there are two kinds of processing units in MRHN, in this study we used the principle of minimizing the error sum of squares to derive the supervised learning rules for all the network parameters. This research took rainfall distribution in Taiwan as a case study. The results show that (1) the prediction error of the testing dataset outside the training dataset demonstrated that MRHN was the most accurate among the three networks, RBFN was the next best, and MLP was the worst; (2) the MLP model seriously underestimated the values of high observed rainfall; (3) over-learning may be a serious shortcoming of using RBFN in spatial interpolation applications; (4) MRHN may have better generalization learning capacity than RBFN in spatial interpolation applications.

Motion capture (mocap) data normally used to produce a three-dimensional (3D) mesh animation is exploited for watermarking. Copyright protection of the 3D mesh animation obtained is not guaranteed since identical joints in the mocap data are not distinguished easily. A novel robust blind watermarking method is proposed for the copyright protection of 3D mesh animations in the paper. At the beginning, the shape diameter function (SDF) method segments one frame of a 3D animation into several meaningful parts. A watermark is then embedded into these parts by adjusting the geodesic distances. Vertex trajectories are further exploited to modify the geodesic distribution of residual frames to be the same as that of the watermarked frame. Our experimental result demonstrates that the proposed method is able to resist the noise addition, similarity transformation, quantization, cropping, simplification, smoothing, and resampling attacks.

In this paper, a series of novel secure dynamic routing protocols are proposed to provide the end-to-end security. Unlike the existing dynamic routing protocols with security consideration which focus on a certain specific network threat, in the paper the proposed approach is based on a comprehensive network security evaluation framework. By analyzing the threats, vulnerabilities, countermeasures, and security requirements of networks, the danger intensity of threats is systematically quantified. Then, the risk score of a link can be obtained by integrating danger intensity of all threats on this link. Since a routing path is composed of several links, three secure dynamic routing protocols, with specific security and performance requirements, are proposed to find a secure routing path according to the link risk scores. The simulation results show that the proposed secure dynamic routing protocols are able to derive a more secure routing path than traditional dynamic routing protocols, even with considering routing performance.

In this paper, we present an enhanced next forwarder selection scheme based on our previous work CODS-AD [4]. It enable vehicles to cooperatively assist source vehicle to discover the position of the destination without the help of location services and to adjust the forwarding strategy according to the road density. The proposed scheme overcomes the disadvantage of the greedy forwarding strategy used in CODS-AD where route stability is not considered and thus link-breakage probability is increased. In addition, proposed scheme make better decisions on how to select a more efficient and stable next forwarder by using location prediction and link duration estimation. The simulation result shows that the proposed scheme not only achieves a higher delivery ratio, but also significantly reduces control overhead and end-to-end delay.

A novel feature region selection method for robust digital image watermarking is proposed in this paper. This method aims to select a nonoverlapping feature region set, which has the greatest robustness against various attacks and can preserve image quality as much as possible after watermarked. It first performs a simulated attacking procedure using some predefined attacks to evaluate the robustness of every candidate feature region. According to the evaluation results, it then adopts a track-with-pruning procedure to search a minimal primary feature set which can resist the most predefined attacks. In order to enhance its resistance to undefined attacks under the constraint of preserving image quality, the primary feature set is then extended by adding into some auxiliary feature regions. This work is formulated as a multidimensional knapsack problem and solved by a genetic algorithm based approach. The experimental results for StirMark attacks on some benchmark images support our expectation that the primary feature set can resist all the predefined attacks and its extension can enhance the robustness against undefined attacks. Comparing with some well-known feature-based methods, the proposed method exhibits better performance in robust digital watermarking.

Pose editing, changing the poses of three-dimensional (3D) mesh model, frequently results in de-synchronization effect which most 3D watermarking methods are sensitive to and normally regard it as an attack. A novel pose-invariant blind watermarking algorithm for 3D triangular mesh model is proposed in this paper to resist the pose changing attack. In this algorithm, the geodesic distance of each vertex pair on the 3D mesh model to be watermarked is calculated, and then this model is segmented into meaningful pose-invariant components based on the calculation result. Each meaningful component is further divided into non-overlapping cells according to the distribution of geodesic distances from its vertices to the cutting boundary. Finally the defined watermark is embedded into these cells by adjusting the mean value of geodesic distances in order to achieve the invariance of pose changing, and later detection follows the same concept. Our experimental result demonstrates that the proposed algorithm has the ability to resist the attacks of pose changing, similarity transformations, vertex reordering, local deformation, cropping, simplification, additive random noise, and smoothing.

Recognizing human emotions from facial expressions is highly dependent on the quality of the referred facial expression features. Conventional methods often suffer from high computation time and serious influence of environment variations. In this paper, a triangular facial feature extraction method based on a Modified Active Shape Model (MASM) is proposed. This method features considering the interactions of all facial features, escaping from the affection of environment variations as well as noisy facial features, and reducing feature dimensions. MASM adopts the same shape representation and shape training procedures as ASM, but executes a different landmark searching procedure without using the gray level training procedure to avoid the affection from environment variations. Using the feature points extracted by MASM, two methods, one is based on statistical analysis and another one is derived from the genetic algorithm, are proposed to extract an optimal set of triangular facial features for emotion recognition. In the experiments with JAFFE database, a neural network classifier is employed to recognize emotions with those extracted triangular facial features. The experimental results show that based on the statistical analysis 65.1% recognition rate is achieved, and based on the genetic algorithm 70.2% recognition rate is achieved.

Negative emotions may affect the quality of life and the function of the cardiovascular system. The purpose of this study is to detect negative emotion symptoms using the heart rate recovery and decay time constant. Twenty seven subjects were recruited for this study with informed consent and Institutional review board approval. The subjects were divided into the negative emotion group and normal group based on tests of Hamilton's Depression and Anxiety Scales. During the experiment, the electrocardiograms (ECGs) for all subjects were continuously recorded, and the related RR intervals were measured for later analysis. Nine subjects having the HAM-D ≥ 10 or HAM-A ≥ 18 are defined and collected to be the negative emotion group. From the experimental results, the heart rate recovery of 1 min (HRR1min), 2 min (HRR2min), and 3min (HRR3min), chronotropic response, and the time of twelve-beats heart rate decay (T12beat) after maximal exercise test are statistically significantly different between the groups of normal and negative emotion symptoms. It is demonstrated that the heart rate recovery and time of twelve-beats heart rate decay are the useful and objective index for negative emotions detection.

As cross-lingual information retrieval is attracting increasing attention, tools that measure cross-lingual semantic similarity between documents are becoming desirable. In this paper, two aspects of cross-lingual semantic document similarity measures are investigated: One is document representation, and the other is the formulation of similarity measures. Fuzzy set and rough set theories are applied to capture the inherently fuzzy relationships among concepts expressed by natural languages. Our approach first develops a language-independent sense-level document representation based on the fuzzy set model to reduce the barrier between different languages and further explores the fuzzy-rough hybrid approach to obtain a more robust macrosense-level document representation through the partitioning of the integrated sense association network of the document collection into macrosenses. Then, Tversky's notion of similarity and the F1 measure on information retrieval are adopted to formulate, respectively, two document similarity measures with fuzzy set operations on the two proposed document representations. The effectiveness of our approach is demonstrated by its success rate in identifying the English translations to their corresponding Chinese documents in a collection of Chinese-English parallel documents. Moreover, the proposed approach can be easily extended to process documents in other languages. It is believed that the proposed representations, along with the similarity measures, will enable more effective text mining processes.

In order to stimulate the development and research in computer Go, several Taiwanese Go players, including three professional Go players and four amateur Go players, were invited to play against the famous computer Go program, MoGo, in the Taiwan Open 2009. The MoGo program combines the online game values, offline values extracted from databases, and expert rules defined by Go expert that shows an excellent performance in the games. The results reveal that MoGo can reach the level of 3 Dan in Taiwan amateur Go environment. But there are still some drawbacks for MoGo that should be solved, for example, the weaknesses in semeai and how to flexibly practice the human knowledge through the embedded opening books. In this paper, a new game record ontology for computer Go knowledge management is proposed to solve the problems that MoGo is facing. It is hoped that the advances in intelligent agent and ontology model can provide much more knowledge to make a progress in computer Go and achieve as much as computer chess or Chinese chess in the future.

In smart living spaces, a reliable context-aware application must adapt to the variable environment and should function to complete the user's requirements. Therefore, if a context-aware application is to achieve the user's requirements smoothly, it must understand the environment status according to trustworthy context information. However, the context information influenced by the environment status may be imperfect. If a context-aware application responds to the user's requirements using the incorrect context information, this will lead to undesirable result. To resolve these problems, this paper proposes a Context Model (CM) to evaluate the reliability of context information and predicted the context information that context-aware application needs. When CM approves of the context information, it considers both the key features of context information, and also the different requirements of the context-aware application. Finally, an experiment is proposed to verify the reliability of CM. The experiment verifies that the reliability of context information recommended by CM can represent the real status. In addition, the experiment verifies whether CM can provide appropriate context information according to the particular requirements of a context-aware application.

In this paper, the service-oriented device composer (SODC) is proposed to support ubiquitous service automation with considering QoS and resource management. A ubiquitous service is regarded as a distributed application carried out via the cooperation among a selected set of devices. Thus, it is essential to arrange the resource-limited devices for all requested ubiquitous services in a bandwidth-limited network and to consider service quality simultaneously. To achieve this goal, we first formulate the resource-constrained device composition problem (RDCP) with the objective of optimizing the service quality of all requested ubiquitous services. However, RDCP is subject to the resource and bandwidth constraints. In our work, RDCP is transformed into a multi-dimension multi-choice knapsack problem (MMKP) by dimension reduction technique. Thus, any efficient MMKP approximation solution is easily applied to solve RDCP. Finally, the experimental result shows the proposed approach is capable of satisfying various user-specific service qualities in resource-constrained ubiquitous environments.

In this paper, a feature-based robust digital image watermarking algorithm is proposed to achieve the goal of image authentication and protection simultaneously. The Hessian-Affine feature detector is at first adopted to extract characteristic regions of an image in the proposed watermark embedding scheme. Then the copyright watermark is embedded into these characteristic regions according to the local orientation of each pixel. Moreover, the remainder regions are applied for image authentication by using block-wise fragile watermarking method. Similarly, the proposed watermark detection scheme follows the above procedure to extract the copyright watermark and the authentic information independently and blindly from watermarked images. Various attacks are also applied to the watermarked images in order to examine the robustness of our algorithm. The experimental results in this paper show that the proposed watermarking algorithm can resist most removal and geometric attacks. Besides, changes or modifications of an image will be reflected in our hidden watermarks.

A resource-efficient key management protocol is essential for security-sensitive applications in wireless sensor networks (WSN). Moreover, the dynamic pair-wise key and group key management protocols are also important for long-lived and mobile WSN. In this paper, a two-layered dynamic key management (TDKM) approach for cluster-based WSN (CWSN) is proposed. Both pair-wise key and group key are distributed in three rounds for key material exchange without encryption/decryption and exponentiation operations in TDKM. In theoretical analysis, TDKM is compared with other key management protocols to show its efficiency. Finally, the relationships between the number of groups and the system performance including key generation overhead, network security, and secured data transmission overhead in CWSN are analyzed.

In this paper a two-stage mode selection (TSMS) algorithm is presented to speed up the H.264/AVC video encoding process with rate distortion optimization (RDO). However, lots of additional computing power is required and this makes the realization of H.264/RDO in a resource-limited system very difficult. The proposed TSMS employs a two-stage decision process: the first stage is to predict some probable encoding modes according to the information when one encodes the preceding macroblocks and video frames. The second stage refines the decision with techniques based on Baye's probability rule and Back-Propagation neural network (BPN). According to the experiment results, over 50% of the computation time is reduced with very slight loss in peak signal-to-noise ratio (PSNR) and a slightly increment in bit rate when TSMS is applied. The TSMS is even faster than the encoding program with running RDO part. All programs are based on the H.264/AVC standard reference software (JM 9.2).

In this paper, we propose two approaches to improve the security of clustering-based sensor networks: authentication-based intrusion prevention and energy-saving intrusion detection. In the first approach, different authentication mechanisms are adopted for two common packet categories in generic sensor networks to save the energy of each node. In the second approach, different monitoring mechanisms are also needed to monitor cluster-heads (CHs) and member nodes according to their importance. When monitoring CHs, member nodes of a CH take turns to monitor this CH. This mechanism reduces the monitor time, and therefore saves the energy of the member nodes. When monitoring member nodes, CHs have the authority to detect and revoke the malicious member nodes. This also saves the node energy because of using CHs to monitor member nodes instead of using all the member nodes to monitor each other. Finally, simulations are performed and compared with LEACH, based on an ns2 LEACH CAD tool. The simulation result shows that the proposed approaches obviously extend the network lifetime when the clustering-based sensor network is under attack.

A novel neural network based method for halftoning and inverse halftoning of digital images is presented. We first start from inverse half-toning of images produced from error diffusion methods using an RBF network plus an MLP network. The restored contone images have had good quality already. Then, an SLP neural network is used to refine the halftoning processing and the training process of the inverse half-toning network is also involved. The combined training procedure produces half-tone images and the corresponding continuous tone images at the same time. It is found that these contone images have even better PSNR performance. Furthermore, the resulted half-tone images are visually sharper and clearer, too. The proposed inverse half-toning method is also compared to the well-known LUT method.

A dynamic slot allocation scheme, called as DSA, to control packet delay in TDMA base station is presented in this paper. DSA applies multi-queue architecture and use the queue status of base station, corresponding to each mobile host, as a feedback to determine the required data rate of hosts. Adaptive rate control, dynamics priority and multi-queuing system are three pillars of DSA. In analysis, we illustrate how the proposed DSA controls the target queue dynamics. Then we investigate the relationship between queue dynamics and packet delay. As a result, the packet delay of each host is effectively controlled by the parameters. Simultaneously, the controlled target queue alleviates the buffer management of a base station. The simulation results show the well-controlled queue dynamics and delay of the proposed scheme, even under a condition of high traffic variation.

This paper proposes an intelligent fuzzy image filter (FIF) to remove impulse noise. The filter includes two processes, the intelligent fuzzy number deciding (IFND) process and fuzzy inference process, to filter impulse noise from heavily corrupted images efficiently. IFND can automatically decide the number of fuzzy number based on image features to overcome the drawbacks of adaptive weighted fuzzy mean (AWFM) filter that must be defined by domain expert. Moreover, the fuzzy inference process refers the knowledge base produced by IFND and fuzzy rule base that can improve the weakness of conventional filters in heavily corrupted condition. The intelligent FIF achieves better performance than the other filters based on the criteria of mean absolute error (MAE), and mean square error (MSE). By the experiments, FIF still keeps the high performance to filtering impulse noise from color image.

We describe the construction of a picture archiving and communication system (PACS) in the Tri-Service General Hospital (TSGH) of Taiwan, and apply a fuzzy technique to process medical images taken from this system. A PACS consists of image and data acquisition, storage, and display subsystems integrated by various digital networks. Based on PACS, we combine fuzzy logic with a knowledge base to carry out medical image processing. The fuzzy image processing is used at the display workstation to improve the diagnostic value of the images. Experimental results show that the method can detect image edges successfully.

The important properties and applications of the adaptive weighted fuzzy mean (AWFM) filter are presented in this paper. AWFM is an extension of the weighted fuzzy mean (WFM) filter to overcome the drawback of WFM in fine signal preservation. It not only preserves the high performance of WFM on heavy additive impulse noise, but also improves the efficiency of WFM on removing light additive impulse noise. Some deterministic and statistical properties of the AWFM filter are analyzed, and the main characteristic of the AWFM filter that maps the input signal space into a root signal space, where a root signal is an invariant signal to the filter, is also discussed. Compared with the other filters, AWFM exhibits better performance in the criteria of mean absolute error and mean square error. On the subjective evaluation of those filtered images, AWFM also results in a higher quality of global restoration. ©1999 John Wiley & Sons, Inc.

A novel primitive cell structure for high-performance hardware realization of fuzzy computations is proposed in this paper. Such a cell structure is called generic LR fuzzy cell because it is an integral unit that encapsulates an LR fuzzy set and a basic fuzzy operation such as implication or arithmetic operation. Based on the proposed cell structure, we can develop two major kinds of fuzzy cell-LR fuzzy implication cell and LR fuzzy arithmetic cell-for the systematic synthesis of fuzzy application specific integrated circuits (ASICs) or general purposed fuzzy processors. The fuzzy systems synthesized with LR fuzzy cells possess the characteristics of decentralized knowledge manipulations and massively parallel inference. The system expandability and reconfigurability are also warrantable. This paper emphasizes on the design and application of fuzzy implication cell. The LR fuzzy implication cell is implemented with analog current mode technology. By this technology, an implication cell has the characteristics including small circuit area, high performance, low-power dissipation, etc. Moreover, the implication cell manipulates continuous data so that the realization of a pure fuzzy system is possible. In this paper, the key circuit characteristics of fuzzy implication cell are evaluated in details and there are two cases-fuzzy knowledge system and fuzzy mean filter-implemented to confirm the effectiveness and efficiency of fuzzy hardware synthesis by the LR fuzzy cells.

This paper proposes a three-layered parallel fuzzy inference model called reinforcement fuzzy neural network with distributed prediction scheme (RFNN-DPS), which performs reinforcement learning with a novel distributed prediction scheme. In RFNN-DPS, an additional predictor for predicting the external reinforcement signal is not necessary, and the internal reinforcement information is distributed into fuzzy rules (rule nodes). Therefore, using RFNN-DPS, only one network is needed to construct a fuzzy logic system with the abilities of parallel inference and reinforcement learning. Basically, the information for prediction in RFNN-DPS is composed of credit values stored in fuzzy rule nodes, where each node holds a credit vector to represent the reliability of the corresponding fuzzy rule. The credit values are not only accessed for predicting external reinforcement signals, but also provide a more profitable internal reinforcement signal to each fuzzy rule itself. RFNN-DPS performs a credit-based exploratory algorithm to adjust its internal status according to the internal reinforcement signal. During learning, the RFNN-DPS network is constructed by a single-step or multistep reinforcement learning algorithm based on the ART concept. According to our experimental results, RFNN-DPS shows the advantages of simple network structure, fast learning speed, and explicit representation of rule reliability.

Applying a fuzzy reasoning concept in an expert system can improve its abilities in approximate reasoning. Embedding an object oriented paradigm into the organization of an expert system can additionally provide richer knowledge representation and reasoning power. A novel fuzzy linguistic object (FLO) model is developed, by which the traditional rule base is replaced by a corpus of fuzzy linguistic objects, and the problem solving process is realized by interactions and inheritances among those objects. Based on the proposed model, a three layered structure including knowledge acquisition layer, object design layer and program design layer is adopted to implement a development tool for fuzzy knowledge systems. In this three layered structure, the user can proceed his design work at any layer, and design information is automatically transformed among layers. An object simulator is also developed to enhance the ability of rapid prototyping. Two case studies were performed to confirm the functions of the proposed development tool.

The cerebellar model articulation controller (CMAC) neural network has the advantages of fast convergence speed and low computation complexity. However, it suffers from a low storage space utilization rate on weight memory. In this paper, we propose a direct weight address mapping approach, which can reduce the required weight memory size with a utilization rate near 100%. Based on such an address mapping approach, we developed a pipeline architecture to efficiently perform the addressing operations. The proposed direct weight address mapping approach also speeds up the computation for the generation of weight addresses. Besides, a CMAC hardware prototype used for color calibration has been implemented to confirm the proposed approach and architecture.

A hardware design method for a high-speed weighted fuzzy mean (WFM) filter is proposed in this paper. The WFM filter is powerful for removing heavy additive impulse noises from images. When the probability of occurrence of mixed impulse noises is over 0.3, the WFM filter can recover the noise-corrupted image quite well in contrast with several conventional filters. For the dedicated hardware implementation, the WFM filter is synthesized with the generic LR fuzzy cells which realize high speed fuzzy inference. The hardware complexity corresponds to the amount of input samples which is also much simpler than the conventional median filter. The simulation and chip layout exhibit that up to 90 256/spl times/256 images can be filtered per second by the very small-sized WFM filter.

The process of eliminating color errors from the gamut mismatch, resolution conversion, quantization and non-linearity between scanner and printer is an essential issue of color reproduction. To efficiently calibrate the non-linear color distortion characteristic between color scanner and printer and to enhance the reproduction quality of color documents, we develop a hardware processing module, which realizes the operation of a higher order CMAC neural network model with linear systolic array architecture, to on-line calibrate the color during the reproducing session. This mapping scheme exhibits fast computation speed in evaluating the output responses of higher-order CMAC model.

A fuzzy classifier based on a four-layered feedforward neural network model is proposed. This connectionist fuzzy classifier, called CFC, realizes the weighted-Euclidean-distance fuzzy classification concept in a massively parallel manner to recognize input patterns. CFC employs a hybrid supervised/unsupervised learning scheme to organize referenced pattern vectors. This scheme not only overcomes the major drawbacks of multilayer perceptron models using the backpropagation algorithm, i.e., the local minimal problem and long training time, but also avoids the disadvantage of the huge storage space requirement of the probabilistic neural network. According to experimental results, CFC shows better accuracy for speech recognition than several existing methods, even in a noisy environment. Moreover, it has higher stability of recognition rates for different environmental conditions. A massively parallel hardware architecture has been developed to implement CFC. A bus-oriented multiprocessor, systolic processor structure, and pipelining are used to obtain low-cost, high-performance fuzzy classification.>