In this letter, an acoustic environment classification algorithm based on the 3GPP2 selectable mode vocoder (SMV) is proposed for context-aware mobile phones. Classification of the acoustic environment is performed based on a Gaussian mixture model (GMM) using coding parameters of the SMV extracted directly from the encoding process of the acoustic input data in the mobile phone. Experimental results show that the proposed environment classification algorithm provides superior performance over a conventional method in various acoustic environments.
Min Li HUANG Hyung-Joun YOO Sin-Chong PARK
This paper presents a reconfigurable multi-band class E power amplifier designed in CMOS technology. The proposed class E power amplifier operates efficiently over sparsely separated frequency bands by switching the capacitance of the load network. Simulation results showed a stable and high power added efficiency of 60% with 18.5 dB gain, and 83% with 14.5 dB gain for 2.4 GHz and 5 GHz WLAN applications, respectively.
Jiancheng SUN Chongxun ZHENG Xiaohe LI
With a Gaussian kernel function, we find that the distance between two classes (DBTC) can be used as a class separability criterion in feature space since the between-class separation and the within-class data distribution are taken into account impliedly. To test the validity of DBTC, we develop a method of tuning the kernel parameters in support vector machine (SVM) algorithm by maximizing the DBTC in feature space. Experimental results on the real-world data show that the proposed method consistently outperforms corresponding hyperparameters tuning methods.
Jungsuk SONG Hiroki TAKAKURA Yasuo OKABE Yongjin KWON
Intrusion detection system (IDS) has played an important role as a device to defend our networks from cyber attacks. However, since it is unable to detect unknown attacks, i.e., 0-day attacks, the ultimate challenge in intrusion detection field is how we can exactly identify such an attack by an automated manner. Over the past few years, several studies on solving these problems have been made on anomaly detection using unsupervised learning techniques such as clustering, one-class support vector machine (SVM), etc. Although they enable one to construct intrusion detection models at low cost and effort, and have capability to detect unforeseen attacks, they still have mainly two problems in intrusion detection: a low detection rate and a high false positive rate. In this paper, we propose a new anomaly detection method based on clustering and multiple one-class SVM in order to improve the detection rate while maintaining a low false positive rate. We evaluated our method using KDD Cup 1999 data set. Evaluation results show that our approach outperforms the existing algorithms reported in the literature; especially in detection of unknown attacks.
Internet routers need to classify incoming packets quickly into flows in order to support features such as Internet security, virtual private networks and Quality of Service (QoS). Packet classification uses information contained in the packet header, and a predefined rule table in the routers. Packet classification of multiple fields is generally a difficult problem. Hence, researchers have proposed various algorithms. This study proposes a multi-dimensional encoding method in which parameters such as the source IP address, destination IP address, source port, destination port and protocol type are placed in a multi-dimensional space. Similar to the previously best known algorithm, i.e., bitmap intersection, multi-dimensional encoding is based on the multi-dimensional range lookup approach, in which rules are divided into several multi-dimensional collision-free rule sets. These sets are then used to form the new coding vector to replace the bit vector of the bitmap intersection algorithm. The average memory storage of this encoding is θ (LNlog N) for each dimension, where L denotes the number of collision-free rule sets, and N represents the number of rules. The multi-dimensional encoding practically requires much less memory than bitmap intersection algorithm. Additionally, the computation needed for this encoding is as simple as bitmap intersection algorithm. The low memory requirement of the proposed scheme means that it not only decreases the cost of packet classification engine, but also increases the classification performance, since memory represents the performance bottleneck in the packet classification engine implementation using a network processor.
Tae DEMPSEY Gokhan SAHIN Yu T. (Jade) MORTON
Wireless ad hoc networks have fundamentally altered today's battlefield, with applications ranging from unmanned air vehicles to randomly deployed sensor networks. Security and vulnerabilities in wireless ad hoc networks have been considered at different layers, and many attack strategies have been proposed, including denial of service (DoS) through the intelligent jamming of the most critical packet types of flows in a network. This paper investigates the effectiveness of intelligent jamming in wireless ad hoc networks using the Dynamic Source Routing (DSR) and TCP protocols and introduces an intelligent classifier to facilitate the jamming of such networks. Assuming encrypted packet headers and contents, our classifier is based solely on the observable characteristics of size, inter-arrival timing, and direction and classifies packets with up to 99.4% accuracy in our experiments.
In LTE, AC barring check is performed before RRC connection. In some cells with a low access probability, the UEs keep retrying access which results in higher connection failure and longer access delay. We therefore propose balancing the UEs by adjusting the cell reselection criteria based on the access probability, so that the UEs shall be more encouraged to reselect a cell with a higher access probability.
Hiroto NAGAYOSHI Yoshitaka HIRAMATSU Hiroshi SAKO Mitsutoshi HIMAGA Satoshi KATO
A system for detecting fundus lesions caused by diabetic retinopathy from fundus images is being developed. The system can screen the images in advance in order to reduce the inspection workload on doctors. One of the difficulties that must be addressed in completing this system is how to remove false positives (which tend to arise near blood vessels) without decreasing the detection rate of lesions in other areas. To overcome this difficulty, we developed classifier selection according to the position of a candidate lesion, and we introduced new features that can distinguish true lesions from false positives. A system incorporating classifier selection and these new features was tested in experiments using 55 fundus images with some lesions and 223 images without lesions. The results of the experiments confirm the effectiveness of the proposed system, namely, degrees of sensitivity and specificity of 98% and 81%, respectively.
Aamir Saeed MALIK Tae-Sun CHOI
A classification method is presented for differentiating honeycombed High Resolution Computed Tomographic (HRCT) images from normal HRCT images. For successful classification of honeycombed HRCT images, a complete set of methods and algorithms is described from segmentation to extraction to feature selection to classification. Wavelet energy is selected as a feature for classification using K-means clustering. Test data of 20 patients are used to validate the method.
A 900 mV single-stage class-AB amplifier suitable for the Switched-Opamp technique is presented. To improve the slew-limited characteristics, a Dynamic Current Source (DCS) circuit which boosts the tail currents of the amplifier is proposed. The tail current of the proposed circuit is well defined and independent of technology parameters and supply variations. The tail current of the amplifier is 40 µA with zero differential voltages, while the maximum output current is nearly 900 µA. A single-loop 3rd order Σ-Δ modulator with the proposed amplifier was designed. For a 260 mV 15.625 kHz sinusoidal input signal, the simulated dynamic range of the modulator is 89 dB.
Fair allocation of bandwidth and maximization of channel utilization are two important issues when designing a contention-based wireless medium access control (MAC) protocol. However, fulfilling both design goals at the same time is very difficult. Considering the problem in the IEEE 802.11 wireless local area networks (WLANs), in this work we propose a method using a p-persistent enhanced DCF, called P-IEEE 802.11 DCF, to achieve weighted fairness and efficient channel utilization among multiple priority classes in a WLAN. Its key idea is that when the back-off timer of a node reaches zero, the transmission probability is properly controlled to reflect the relative weights among data traffic flows so as to maximize the aggregate throughput and to minimize the frame delay at the same time. In particular, we obtain the optimal transmission probability based on a theoretical analysis, and also provide an approximation to this probability. The derived optimal and approximation are all evaluated numerically and simulated with different scenarios. The results show that the proposed method can fulfill our design goals under different numbers of priority classes and different numbers of nodes.
MPLS-based path technology shows promise as a means of realizing reliable IP networks. Real-time services such as VoIP and video-conference supplied through a multi-domain MPLS network must be able to guarantee end-to-end QoS of the inter-domain paths. Thus, it is important to allocate an appropriate QoS class to the inter-domain paths in each domain traversed by the inter-domain paths. Because each domain has its own policy for QoS class allocation, it is necessary to adaptively allocate the optimum QoS class based on estimation of the QoS class allocation policies in other domains. This paper proposes two kinds of adaptive QoS class allocation schemes, assuming that the arriving inter-domain path requests include the number of downstream domains traversed by the inter-domain paths and the remaining QoS value toward the destination nodes. First, a measurement-based scheme, based on measurement of the loss rates of inter-domain paths in the downstream domains, is proposed. This scheme estimates the QoS class allocation policies in the downstream domains, using the measured loss rates of path requests. Second, a state-dependent type scheme, based on measurement of the arrival rates of path requests in addition to the loss rates of paths in the downstream domains, is also proposed. This scheme allows an appropriate QoS class to be allocated according to the domain state. This paper proposes an application of the Markov decision theory to the modeling of state-dependent type scheme. The performances of the proposed schemes are evaluated and compared with those of the other less complicated non-adaptive schemes using a computer simulation. The results of the comparison reveal that the proposed schemes can adaptively increase the number of inter-domain paths accommodated in the considered domain, even when the QoS class allocation policies change in the other domains and the arrival pattern of path requests varies in the considered domain.
Yingjun TANG De XU Guanghua GU Shuoyan LIU
We present a novel model, named Category Constraint-Latent Dirichlet Allocation (CC-LDA), to learn and recognize natural scene category. Previous work had to resort to additional classifier after obtaining image topic representation. Our model puts the category information in topic inference, so every category is represented in a different topics simplex and topic size, which is consistent with human cognitive habit. The significant feature in our model is that it can do discrimination without combined additional classifier, during the same time of getting topic representation. We investigate the classification performance with variable scene category tasks. The experiments have demonstrated that our learning model can get better performance with less training data.
In this letter, we propose a novel approach to speech/music classification based on the support vector machine (SVM) to improve the performance of the 3GPP2 selectable mode vocoder (SMV) codec. We first analyze the features and the classification method used in real time speech/music classification algorithm in SMV, and then apply the SVM for enhanced speech/music classification. For evaluation of performance, we compare the proposed algorithm and the traditional algorithm of the SMV. The performance of the proposed system is evaluated under the various environments and shows better performance compared to the original method in the SMV.
Jae Soong LEE Jae Young LEE Soobin LEE Hwang Soo LEE
Although each application has its own quality of service (QoS) requirements, the resource allocation for multiclass services has not been studied adequately in multiuser orthogonal frequency division multiplexing (OFDM) systems. In this paper, a total transmit power minimization problem for downlink transmission is examined while satisfying multiclass services consisting of different data rates and target bit-error rates (BER). Lagrangian relaxation is used to find an optimal subcarrier allocation criterion in the context of subcarrier time-sharing by all users. We suggest an iterative algorithm using this criterion to find the upper and lower bounds of optimal power consumption. We also propose a prioritized subcarrier allocation (PSA) algorithm that provides low computation cost and performance very close to that of the iterative algorithm. The PSA algorithm employs subcarrier selection order (SSO) in order to decide which user takes its best subcarrier first over other users. The SSO is determined by the data rates, channel gain, and target BER of each user. The proposed algorithms are simulated in various QoS parameters and the fading channel model. Furthermore, resource allocation is performed not only subcarrier by subcarrier but also frequency block by frequency block (comprises several subcarriers). These extensive simulation environments provide a more complete assessment of the proposed algorithms. Simulation results show that the proposed algorithms significantly outperform existing algorithms in terms of total transmit power consumption.
Reversal complexity has been studied as a fundamental computational resource along with time and space complexity. We revisit it by contrasting with access complexity which we introduce in this study. First, we study the structure of space bounded reversal and access complexity classes. We characterize the complexity classes L, P and PSPACE in terms of space bounded reversal and access complexity classes. We also show that the difference between polynomial space bounded reversal and access complexity is related with the L versus P problem. Moreover, we introduce a theory of memory access patterns, which is an abstracted structure of the order of memory accesses during a random access computation, and extend the notion of reversal and access complexity for general random access computational models. Then, we give probabilistic analyses of reversal and access complexity for almost all memory access patterns. In particular, we prove that almost all memory access patterns have ω(log n) reversal complexity while all languages in L are computable within O(log n) reversal complexity.
Ji-Yeoun LEE Sangbae JEONG Hong-Shik CHOI Minsoo HAHN
This work proposes new features to improve the pathological voice quality classification performance. They are the means, the variances, and the perturbations of the higher-order statistics (HOS) such as the skewness and the kurtosis. The HOS-based features show meaningful differences among normal, grade 1, grade 2, and grade 3 voices classified in the GRBAS scale. The jitter, the shimmer, the harmonic-to-noise ratio (HNR), and the variance of the short-time energy are utilized as the conventional features. The performances are measured by the classification and regression tree (CART) method. Specifically, the CART-based method by utilizing both the conventional features and the HOS-based ones shows its effectiveness in the pathological voice quality measurement, with the classification accuracy of 87.8%.
Ruicong ZHI Qiuqi RUAN Jiying WU
This paper proposes a novel algorithm for image feature extraction-the dual two-dimensional fuzzy class preserving projections ((2D)2FCPP). The main advantages of (2D)2FCPP over two-dimensional locality preserving projections (2DLPP) are: (1) utilizing the fuzzy assignation mechanisms to construct the weight matrix, which can improve the classification results; (2) incorporating 2DLPP and alternative 2DLPP to get a more efficient dimensionality reduction method-(2D)2LPP.
Youngsoo KIM Sangbae JEONG Daeyoung KIM
In this paper, an efficient node-level target classification scheme in wireless sensor networks (WSNs) is proposed. It uses acoustic and seismic information, and its performance is verified by the classification accuracy of vehicles in a WSN. Because of the hard limitation in resources, parametric classifiers should be more preferable than non-parametric ones in WSN systems. As a parametric classifier, the Gaussian mixture model (GMM) algorithm not only shows good performances to classify targets in WSNs, but it also requires very few resources suitable to a sensor node. In addition, our sensor fusion method uses a decision tree, generated by the classification and regression tree (CART) algorithm, to improve the accuracy, so that the algorithm drives a considerable increase of the classification rate using less resources. Experimental results using a real dataset of WSN show that the proposed scheme shows a 94.10% classification rate and outperforms the k-nearest neighbors and the support vector machine.
Chien-Tsun CHEN Yu Chin CHENG Chin-Yun HSIEH
Design by Contract (DBC), originated in the Eiffel programming language, is generally accepted as a practical method for building reliable software. Currently, however, few languages have built-in support for it. In recent years, several methods have been proposed to support DBC in Java. We compare eleven DBC tools for Java by analyzing their impact on the developer's programming activities, which are characterized by seven quality attributes identified in this paper. It is shown that each of the existing tools fails to achieve some of the quality attributes. This motivates us to develop ezContract, an open source DBC tool for Java that achieves all of the seven quality attributes. ezContract achieves streamlined integration with the working environment. Notably, standard Java language is used and advanced IDE features that work for standard Java programs can also work for the contract-enabled programs. Such features include incremental compilation, automatic refactoring, and code assist.