In this paper, a semi-supervised transfer learning with dynamic associate domain adaptation is proposed for human activity recognition by using the channel state information (CSI) of the WiFi signal. We propose a dynamic associate domain adaptation (DADA), by modifying the existing associate domain adaptation algorithm, while the target domain can dynamically provide a different ratio of labelled data set/unlabelled data set. The advantage of DADA is that it provides a dynamic strategy to eliminate different effects under the different environments. We designed an attention-based DenseNet model (AD) as our training network, so our proposed scheme is simplified as DADAAD scheme. The experimental results illustrate that the accuracy of human activity recognition of the DADA-AD scheme is 97.4%. It also shows that DADA-AD has advantages over existing semi-supervised learning schemes.

Shot boundary detection (SBD) is an important and fundamental step in video content analysis such as content-based video indexing, browsing, and retrieval. In this paper, a hybrid SBD method is presented by integrating a high-level fuzzy Petri net (HLFPN) model with keypoint matching. The HLFPN model with histogram difference is executed as a predetection. Next, the speeded-up robust features (SURF) algorithm that is reliably robust to image affine transformation and illumination variation is used to figure out all possible false shots and the gradual transition based on the assumption from the HLFPN model. The top-down design can effectively lower down the computational complexity of SURF algorithm. The proposed approach has increased the precision of SBD and can be applied in different types of videos.

Mobile data offloading has become an important issue for mobile cellular network in recent years. Existing energy-aware data offloading scheme for mobile cellular network makes the offloading decision only according to the current local information of the user equipment (UE) and thus the UE cannot switch to a more efficient way to offload even other UEs have released their transmission resources later on. Hence, in this paper, a centralized offloading scheme based on cloud radio access networks (C-RAN) is proposed so as to make the offloading more efficient. The C-RAN based centralized offloading scheme considers all the situation of UEs of a baseband unit (BBU) at the same time and thus the BBU can do a better offloading decision for those UEs. At the beginning, each UE, who needs offloading, sends an offloading request to the BBU. When there is any UE released the transmission resource, the corresponding BBU will try to fulfill the pending offloading requests of the UEs. The corresponding BBU considers the pending UEs' transmission rate and energy consumption of the cellular network as well as the Wi-Fi network, and then makes offloading decisions for those UEs so as to save more energy and achieve higher throughput at the macroscopic level. Extensive simulations have been conducted to illustrate that the proposed energy-aware data offloading scheme can reduce the energy consumption and turnaround time, and enhance the transmission throughput.

Recently, depression become a general disease in the world due to the promotion of life quality and technology development. Most of people are not aware of the possibility of getting depressed himself in daily life. To accurately diagnose getting depressed becomes an important issue. In this paper, we utilize ontologies and Bayesian networks techniques to build the inference model for inferring the possibility of depression. We propose an ontology model to build the terminology of depression and utilize the Bayesian networks to infer the probability of depression. In addition, the paper also proposes an agent-based platform and addresses the implementation issue. The result shows that it can be well-inferring in the depression diagnosis.

With the advance of cloud computing, cloud service providers (CSP) provide increasingly diversified services to users. To utilize general search engine, such as Google, is not an effective and efficient if the services are similar but with different attributes. Therefore, an intelligent service discovery platform is necessary for seeking suitable services accurately and quickly. In the paper1, we propose a framework that integrates intelligent agent and ontology for service discovery in cloud environment. The framework contains some agents and mainly assists users to discovering suitable service according to user demand. User can submit his flat-text based request for discovering required service. We implement a cloud service discovery environment to demonstrate the concept and its application. We also utilize the Recall and Precision to evaluate the accuracy of the system.

Artificial neural networks (ANN) are developed for highly parallel and distributed systems. These systems are able to learn from experience and to perform inferences. Although Petri nets (PNs) were modified to be ANN-like multilayered architectures for fuzzy reasoning, some researchers have paid more attention to the PN-based learning so far. In this paper, we have developed supervised and unsupervised learning algorithms for the machine learning PN (MLPN) models in order to make them fully trainable and to remedy the difficulties encountered by ANN. When compared with ANN, the MLPN model shows some significant advantages. Main results are presented in the form of five observations and supported by some experiments.

In this paper, we, based on mobile agent technology, propose a resource-aware knowledge extraction architecture on mobile grid, named RAKER. RAKER can dynamically determine the processing and policy for achieving high-performance and high availability of knowledge extracting based on our previous proposed resource estimation model. On the RAKER, users can extract information or knowledge in efficient,effective and transparent way that kept on the mobile grid without caring about the energy consumption that is most important issue in mobile computing. We show the implementation and an example to demonstrate the use of RAKER. In addition, we also measure the latency and the energy consumption and simulate the system availability to show that the performance of RAKER.

In this paper, a new approach is proposed to reduce handoff operation in IEEE 802.16e network. Traditional mobile approaches, such as mobile IPv6 (MIPv6) and hierarchical MIPv6 (HMIPv6), can support smoothly handoff. These approach, ,unfortunately, suffer large handoff delay and packet lost in macro mobility for mobile users. With the aid of the partner node, DAD operation can be pre-executed by the partner node before the mobile node initialize the handoff request, we propose a cross-layer partner based fast handoff mechanism based on HMIPv6 in IEEE 802.16e network, called as P_HMIPv6 protocol. The P_HMIPv6 protocol is a cross-layer, layer 2 and layer 3, approach. The partner station (PS), which is a new component with relay ability and adopted by our protocol, is a static mobile station (MS). With the aid of the PSs, care-of address (CoA) can be pre-acquired and DAD operation can be pre-executed by the PS before the MS initials the layer 2 handoff. The simulation results show that P_HMIPv6 protocol actually achieves the performance improvements in the handoff delay time and the packet loss.

As expert-system technology gains broader acceptance, the need to build and maintain large-scale knowledge-based systems (KBSs) will assume greater importance. Traditional approaches to KBS verification generally contain no predicate/transition (PrT) net models, thus making them slow for the large-scale KBS with chained errors. This paper proposes an attractive alternative to KBS verification, in which the KBS is modeled as a PrT-net model. Then, the least fixpoint semantics of the PrT-net model can be introduced into the KBS for the purpose of speeding up the computations of the KBSs. The significance of this paper is that seven propositions are formulated to detect errors of redundancy, subsumption, unnecessary condition, circularity, inconsistency, dead end, and unreachable goal. Thus, the performance of a computer-aided-design tool for KBSs can be improved to some extent. Meanwhile, specification languages, including Programming in Logic, Frame-and-Rule-Oriented Requirements Specification Language, and the like, are suitable to this approach.

The arrangement graph A/sub n,k/ is not only a generalization of the star graph (n-k=1), but is also more flexible. Designing an efficient routing algorithm on a regular interconnection network is a fundamental issue for parallel processing techniques. We elucidate the multi-node broadcasting problem in an all-port communication model on the arrangement graph. Our routing strategy is proposed by construction of 2(n-k) spanning trees, where the height of each spanning tree is 2k-1. We also extend the routing strategy to a one-to-all broadcasting algorithm. Using 2(n-k) spanning trees allows us to present efficient (one/multi)- node broadcasting algorithms in the arrangement graph. The arrangement graph is assumed to use one-port and all-port models and packet-switching (or store-and forward) technique. Moreover we show that our (one/multi)-node broadcasting algorithms outperform previous results in the literature. This is justified by our performance analysis.

Trees are a common structure to represent the inter-task communication pattern of a parallel algorithm. In this paper, we consider the embedding a complete binary tree in a star graph with the objective of minimizing congestion and dilation. We develop two embeddings: i) a congestion-free, dilation-2, load-1 embedding of a level- p binary tree, and ii) a congestion-free, dilation-2, load-2 k embedding of a level-(p+k) binary tree, into an n-dimensional star graph, where p = \\Sigma n i=2 blog ic = \\Omega\\Gamma n log n) and k is any positive integer. The first result offers a tree of size comparable or superior to existing results, but with less congestion and dilation. The second result provides more flexibility in the embeddable tree sizes compared to existing results. Keywords: Graph embedding, interconnection network, complete binary tree, star graph, parallel processing. This research was supported in part by the National Science Council, R.O.C., under grant numbers NS...

Recently, wormhole routers with multidestination capability have been proposed to support fast multicast in a multicomputer network. To avoid communication deadlock, existing results have proposed to construct a Hamilton path, Euler path, trip, or their variants in the network, perhaps with some degree of support of virtual channels. In this paper, we identify that a network which is itself Eulerian or is Eulerian after some links are removed, can enjoy the multidestination capability without support of virtual channels. From this definition, we then develop several techniques to achieve fault-tolerant multicast in a torus/mesh of any dimension with regular fault patterns (such as single node, block, L-shape, T-shape, +-shape, U-shape, and H-shape) and even irregular fault patterns. The result improves over existing results on the requirement of support of virtual channels and fault-tolerant capability. Simulation results on tori are presented.

The arrangement graph A/sub n,k/ is a generalization of star graph (n-k=1) and more flexible than the star graph. In this paper we consider the embedding of multiple spanning trees in an arrangement graph with the objective of being congestion-free. This is first result to exploit multiple spanning trees in the arrangement graphs. We develop a congestion-free embedding of n-k spanning trees with height 2k-1 in an (n, k)-dimensional arrangement graph.

Recently, wormhole routers with multi-destination capability have been proposed to support fast multicast in a multi-computer network. In this paper, we develop a new multicasting model for such networks based on the concept of Euler path/circuit in graph theory. The model can support multiple concurrent multicasts freely from deadlock and can be applied to any network which is Eulerian or is Eulerian after some links being removed. No virtual channels are needed. In particular, we demonstrate the potential of this model by showing its fault-tolerant capability in supporting multicasting in the currently popular torus/mesh topology of any dimension with regular fault patterns (such as single node, block, L-shape, +-shape, U-shape, and H-shape) and even irregular fault patterns. The result has improved over existing fault-tolerant routing algorithms for meshes/tori in at least one of the following aspects: the number of faults tolerable, the shape of fault patterns, the number of deactivated healthy nodes, the requirement of support of virtual channels, and the range of network topology acceptable.

We present an optimistic crash recovery technique without any communication overhead during normal operations of the distributed system. Our technique does not append any information to the application messages, it does not suffer from the domino effect, and each processor rolls back at most once during recovery. We present three distributed rollback algorithms, their complexities, and correctness proofs. Their performances are measured through extensive simulations.

Recovering from processor failures in distributed systems is an important problem in the design and development of reliable systems. Two solutions to this problem which involve very little overhead are presented. Without appending any information to the messages of the application program, it is shown that it is possible to recover from failures using O( mod V mod mod E mod ) messages where mod V mod is the number of processors and mod E mod is the number of communication links in the system. The second algorithm can be used to recover from processor failures without forcing nonfaulty processors to roll back under certain conditions.>