In this paper, we design an efficient P2P based mobile social network to facilitate contents search over mobile ad hoc networks. Social relation is established by considering both the locations and interests of mobile nodes. Mobile nodes with common interests and nearby locations are recommended as friends and are connected directly in a mobile social network. Contents search is handled by using social relationships of the mobile social network rather than those of the whole network. Since each mobile node manages only neighboring nodes that have common interests, network management overhead is reduced. Results of experiments have shown that our proposed method outperforms existing methods.

The estimation of the power spectral density (PSD) of noise is crucial for retrieving speech in noisy environments. In this study, we propose a novel method for estimating the non-white noise PSD from noisy speech on the basis of a generalized gamma distribution and the minimum mean square error (MMSE) approach. Because of the highly non-stationary nature of speech, deriving its actual spectral probability density function (PDF) using conventional modeling techniques is difficult. On the other hand, spectral components of noise are more stationary than those of speech and can be represented more accurately by a generalized gamma PDF. The generalized gamma PDF can be adapted to optimally match the actual distribution of the noise spectral amplitudes observed at each frequency bin utilizing two real-time updated parameters, which are calculated in each frame based on the moment matching method. The MMSE noise PSD estimator is derived on the basis of the generalized gamma PDF and Gaussian PDF models for noise and speech spectral amplitudes, respectively. Combined with an improved Weiner filter, the proposed noise PSD estimate method exhibits the best performance compared with the minimum statistics, weighted noise estimation, and MMSE-based noise PSD estimation methods in terms of both subjective and objective measures.

This paper proposes and investigates a multiplexing and error control scheme for Body Area Network (BAN). In February 2012, an international standard of WBAN, IEEE802.15.6, was published and it supports error control schemes. This standard also defines seven different QoS modes however, how to utilize them is not clearly specified. In this paper, an optimization method of the QoS is proposed. In order to utilize the QoS parameters, a multiplexing scheme is introduced. Then, the Hybrid ARQ in IEEE 802.15.6 is modified to employ decomposable codes and Weldon's ARQ protocol for more associations with channel conditions and required QoS. The proposed scheme has higher flexibility for optimizing the QoS parameters according to the required QoS.

Suppose that each arc in a digraph D = (V,A) has two costs of non-negative integers, called a spine cost and a leaf cost. A caterpillar is a directed tree consisting of a single directed path (of spine arcs) and leaf vertices each of which is incident to the directed path by exactly one incoming arc (leaf arc). For a given terminal set K ⊆ V, we study the problem of finding a caterpillar in D such that it contains all terminals in K and its total cost is minimized, where the cost of each arc in the caterpillar depends on whether it is used as a spine arc or a leaf arc. In this paper, we first show that the problem is NP-hard for any fixed constant number of terminals with |K| ≥ 3, while it is solvable in polynomial time for at most two terminals. We also give an inapproximability result for any fixed constant number of terminals with |K| ≥ 3. Finally, we give a linear-time algorithm to solve the problem for digraphs with bounded treewidth, where the treewidth for a digraph D is defined as the one for the underlying graph of D. Our algorithm runs in linear time even if |K| = O(|V|), and the hidden constant factor of the running time is just a single exponential of the treewidth.

By the motivation to discover patterns in massive structured data in the form of graphs and trees, we study a special case of the k-subtree enumeration problem with a tree of n nodes as an input graph, which is originally introduced by (Ferreira, Grossi, and Rizzi, ESA'11, 275-286, 2011) for general graphs. Based on reverse search technique (Avis and Fukuda, Discrete Appl. Math., vol.65, pp.21-46, 1996), we present the first constant delay enumeration algorithm that lists all k-subtrees of an input rooted tree in O(1) worst-case time per subtree. This result improves on the straightforward application of Ferreira et al.'s algorithm with O(k) amortized time per subtree when an input is restricted to tree. Finally, we discuss an application of our algorithm to a modification of the graph motif problem for trees.

This letter proposes a noise spectrum estimation algorithm for speech enhancement. The algorithm incorporates the speech presence probability, which is calculated from SNR (signal-to-noise ratio) discrepancy. The discrepancy is measured based on the estimation of the a priori and a posteriori SNR. The proposed algorithm is found to be effective in rapidly switched noise environments. This is confirmed by the experimental results which indicate that the proposed algorithm when integrated in a speech enhancement scheme performs better than conventional noise estimation algorithms.

We propose a fast and resource-efficient agreement protocol on a request set, which is used to realize Byzantine fault tolerant server replication. Although most existing randomized protocols for Byzantine agreement exploit a modular approach, that is, a combination of agreement on a bit value and a reduction of request set values to the bit values, our protocol directly solves the multi-valued agreement problem for request sets. We introduce a novel coin tossing scheme to select a candidate of an agreed request set randomly. This coin toss allows our protocol to reduce resource consumption and to attain faster response time than the existing representative protocols.

This paper proposes a new optimization problem and several implementation algorithms for energy-efficient clouds where energy efficiency is measured by the number of physical machines that can be removed from operation and turned off. The optimization problem is formulated is such a way that solutions are considered favorable not only when the number of migrations is minimized but also when the resulting layout has more free physical machines which can therefore be turned off to save electricity.

This letter proposes a mobile application framework named erasable photograph tagging (EPT) for photograph annotation and fast retrieval. The smartphone owner's voice is employed as tags and hidden in the host photograph without an extra feature database aided for retrieval. These digitized tags can be erased anytime with no distortion remaining in the recovered photograph.

In this paper, we present a new four parameter estimator of sampled sinusoidal signals that does not require iteration. Mathematically, the four parameters (frequency, phase, magnitude, and dc offset) of sinusoidal signals can be obtained when four data points are given. In general, the parameters have to be calculated with iteration since the equations are nonlinear. In this paper, we point out that the four parameters can be obtained analytically if the four data points given are measured using a fixed sampling interval. Analytical expressions for the four parameters are derived using the signal differences. Based on this analysis, we suggest an algorithm of estimating the four parameters from N data samples corrupted by noise without iteration. When comparing with the IEEE-1057 method which is based on the least-square method, the proposed algorithm does not require the initial guess of the parameters for iteration and avoid the convergence problem. Also, the number of required numerical operations for estimation is fixed if N is determined. As a result, the processing time of parameter estimation is much faster than the least-square method which has been confirmed by numerical simulations. Simulation results and the quantitative analysis show that the estimation error of the estimated parameters is less than 1.2 times the square root of the Cramer-Rao bounds when the signal to noise ratio is larger than 20dB.

This paper presents a novel acoustic modeling technique of large vocabulary automatic speech recognition for under-resourced languages by leveraging well-trained acoustic models of other languages (called source languages). The idea is to use source language acoustic model to score the acoustic features of the target language, and then map these scores to the posteriors of the target phones using a classifier. The target phone posteriors are then used for decoding in the usual way of hybrid acoustic modeling. The motivation of such a strategy is that human languages usually share similar phone sets and hence it may be easier to predict the target phone posteriors from the scores generated by source language acoustic models than to train from scratch an under-resourced language acoustic model. The proposed method is evaluated using on the Aurora-4 task with less than 1 hour of training data. Two types of source language acoustic models are considered, i.e. hybrid HMM/MLP and conventional HMM/GMM models. In addition, we also use triphone tied states in the mapping. Our experimental results show that by leveraging well trained Malay and Hungarian acoustic models, we achieved 9.0% word error rate (WER) given 55 minutes of English training data. This is close to the WER of 7.9% obtained by using the full 15 hours of training data and much better than the WER of 14.4% obtained by conventional acoustic modeling techniques with the same 55 minutes of training data.

Workflow nets are a standard way for modeling and analyzing workflows. There are two aspects in a workflow: definition and instance. In form of workflow nets, a workflow definition and a workflow instance are respectively represented as a net structure and a marking. The correctness of the workflow definition can be checked by using a workflow nets' property, called soundness. On the other hand, the correctness of the workflow instance can be checked by using a Petri nets' well-known property, called reachability. The reachability problem is known to be intractable. In this paper, we have shown that the reachability problem for (i) sound extended free-choice workflow nets with a marking representing one workflow instance or (ii) acyclic well-structured workflow nets with a marking representing one or more workflow instances can be solved in polynomial time.

This paper summarizes the current status of regulations, standardization efforts and trials around the world regarding white space (WS) communications, especially television band WS (TVWS). After defining WS communication systems configurations and function and the categories of white space database, the TVWS regulations in United States, United Kingdom, and Japan are summarized. Then regarding status of standardization for TVWS devices, IEEE 802 and IEEE 1900 standards are summarized. Finally ongoing pilot projects and trials of WS communications in the world are summarized, and trends and future direction of research on WS communication systems are summarized.

In this paper, we propose a robust speech/music classification algorithm to improve the performance of speech/music classification in the selectable mode vocoder (SMV) of 3GPP2 using deep belief networks (DBNs), which is a powerful hierarchical generative model for feature extraction and can determine the underlying discriminative characteristic of the extracted features. The six feature vectors selected from the relevant parameters of the SMV are applied to the visible layer in the proposed DBN-based method. The performance of the proposed algorithm is evaluated using the detection accuracy and error probability of speech and music for various music genres. The proposed algorithm yields better results when compared with the original SMV method and support vector machine (SVM) based method.

As a promising lamination-loss-free fabrication technique, diffusion bonding of etched thin metal plates is used to realize double-layer waveguide slot antennas. Alternating-phase feed is adopted in this paper to reduce the number of laminated plates to simplify fabrication as well as to reduce cost. A 20 × 20-element double-layer waveguide slot antenna with a bottom partially-corporate feed circuit is designed for 39GHz band operation as an example. The adjacent radiating waveguides as well as the 2 × 2 sub-arrays fed in an alternating-phase manner eliminate the need for complete electrical contact in the top layer. However, the feed circuit in the bottom layer has to be completely diffusion-bonded. These two layers are simply assembled by screws. An antenna laminated by only diffusion bonding is also fabricated and evaluated for comparison. The comparison proved that the simply fabricated antenna is comparable in performance to the fully diffusion-bonded one.

Web page segmentation has a variety of benefits and potential web applications. Early techniques of web page segmentation are mainly based on machine learning algorithms and rule-based heuristics, which cannot be used for large-scale page segmentation. In this paper, we propose a formulated page segmentation method using visual semantics. Instead of analyzing the visual cues of web pages, this method utilizes three measures to formulate the visual semantics: layout tree is used to recognize the visual similar blocks; seam degree is used to describe how neatly the blocks are arranged; content similarity is used to describe the content coherent degree between blocks. A comparison experiment was done using the VIPS algorithm as a baseline. Experiment results show that the proposed method can divide a Web page into appropriate semantic segments.

Two classes of 3rd order correlation immune symmetric Boolean functions have been constructed respectively in [1] and [2], in which some interesting phenomena of the algebraic degree have been observed as well. However, a good explanation has not been given. In this paper, we obtain the formulas for the degree of these functions, which can well explain the behavior of their degree.

Coupled with the discrete wavelet transform, SPIHT (set partitioning in hierarchical trees) is a highly efficient image compression technique that allows for progressive transmission. One problem, however, is that its decoding can be extremely sensitive to bit errors in the code sequence. In this paper, we address the issue of transmitting SPIHT-encoded images via noisy channels, wherein errors are inevitable. The communication scenario assumed in this paper is that the transmitter cannot get any acknowledgement from the receiver. In our scheme, the original SPIHT code sequence is first segmented into packets. Each packet is classified as either a CP (critical packet) or an RP (refinement packet). For error control, cyclic redundancy check (CRC) is incorporated into each packet. By checking the CRC check sum, the receiver is able to tell whether a packet is correctly received or not. In this way, the noisy channel can be effectively modeled as an erasure channel. For unequal error protection (UEP), each of those packets are repeatedly transmitted for a few times, as determined by a process called diversity allocation (DA). Two DA algorithms are proposed. The first algorithm produces a nearly optimal decoded image (as measured in the expected signal-to-noise ratio). However, its computation cost is extremely high. The second algorithm works in a progressive fashion and is naturally compatible with progressive transmission. Its computation complexity is extremely low. Nonetheless, its decoded image is nearly as good. Experimental results show that the proposed scheme significantly improves the decoded images. They also show that making distinction between CP and RP results in wiser diversity allocation to packets and thus produces higher quality in the decoded images.

In this paper, we investigate how to achieve call admission control (CAC) for guaranteeing call dropping probability QoS which is caused by handoff timeout in cognitive radio (CR) networks. When primary user (PU) appears, spectrum handoff should be initiated to maintain secondary user (SU)'s link. We propose a novel virtual queuing (VQ) scheme to schedule spectrum handoff requests sent by multiple SUs. Unlike the conventional first-come-first-served (FCFS) scheduling, resuming transmission in the original channel has higher priority than switching to another channel. It costs less because it avoids the cost of signaling frequent spectrum switches. We characterize the handoff delay on the effect of PU's behavior and the number of SUs in CR networks. And user capacity under certain QoS requirement is derived as a guideline for CAC. The analytical results show that call dropping performance can be greatly improved by CAC when a large amount of SUs arrives fast as well as the VQ scheme is verified to reduce handoff cost compared to existing methods.

An amplitude shift keying transmitter and receiver chipset with low power consumption using 40nm CMOS technology for wireless communication systems is described, in which a maximum data rate of 10Gbps and power consumption of 98.4mW are obtained with a carrier frequency of 135GHz. A simple circuit and a modulation method to reduce power consumption are selected for the chipsets. To realize multi-gigabit wireless communication, the receiver is designed considering the group delay optimization. In the receiver design, the low-noise amplifier and detector are designed considering the total optimization of the gain and group delay in the millimeter-wave modulated signal region.