Many services on the Internet are provided by multiple identical servers in order to improve performance and robustness. The number, the location and the distribution of servers affect the performance and reliability of a service. The server placement is, however, often determined based on the empirical knowledge of the administrators. This paper investigates issues of the server placement in terms of the service performance and the server load. We identify that a server selection mechanism plays an important role in server placement, and thus, evaluate different server selection algorithms. The result shows that it is essential to the robustness of a service to employ a mechanism which distributes service requests to the servers according to the measured response time of each server. As a case study, we evaluate the server selection mechanisms employed by different DNS (Domain Name System) implementations. Then, we show the effects of the different server selection algorithms using root-server measurements taken at different locations around the world.

This paper defines the distributed arrangement selection problem in a line network in a distributed context and describes the design of a strictly-time-optimal algorithm which solves the problem with a limited local memory space. The problem is regarded as a combined distributed sorting and k-selection problem, namely a problem of sorting elements that are not larger than the kth minimum element in predetermined processes. The algorithm also provides a solution to a resource allocation problem in a line network in a strictly-optimal time.

This paper describes a stock trading system based on reinforcement learning, regarding the process of stock price changes as Markov decision process (MDP). The system adopts two popular reinforcement learning algorithms, temporal-difference (TD) and Q, for selecting stocks and optimizing trading parameters, respectively. Input features of the system are devised using technical analysis and value functions are approximated by feedforward neural networks. Multiple cooperative agents are used for Q-learning to efficiently integrate global trend prediction with local trading strategy. Agents communicate with others sharing training episodes and learned policies, while keeping the overall scheme of conventional Q-learning. Experimental results on the Korean stock market show that our trading system outperforms the market average and makes appreciable profits. Furthermore, we can find that our system is superior to a system trained by supervised learning in view of risk management.

This paper proposes and investigates a coding and decoding scheme to achieve adaptive channel coding using a Finite State Machine (FSM) for Software Defined Radio (SDR). Adaptive channel coding and decoding systems that can switch between different coding rates and error correcting capabilities in order to adapt to changing applications and environments, are effective for SDR. However, in these systems, a receiver cannot always select the correct decoder which causes decoding errors, usually referred to as Decoder-Selection-Errors (DSE). We propose a trellis encoder estimation scheme that compensates for this problem. This scheme uses the circuit of FSM to limit the encoder transition and the Viterbi algorithm for maximum likelihood trellis encoder estimation. Computer simulations are applied for evaluating the DSE rate, the Bit Error Rate (BER) and Throughput of the proposed scheme in comparison with a conventional scheme.

This paper presents a practical fault model for delay testing, called a multiple-threshold gate-delay fault model, to obtain high quality tests that guarantee the detection of delay faults for various extra-delays. Fault efficiencies for multiple thresholds of the extra-delay are introduced as a coverage metric that describes the quality of tests. Our approach guarantees that each gate-delay fault is tested on the path that is almost the longest one passing through the faulty line by using two-pattern tests with pattern-independent timing. We present the procedures of the path selection, fault simulation, and the test generation, where the path-status graph technique is used as not to rely on the enumeration of paths. Experimental results for benchmark circuits demonstrate that the proposed metric gives useful information that transition fault efficiency cannot, and that the proposed test generation can achieve high fault efficiencies for multiple-threshold gate-delay faults.

In this paper, a multi-coded variable spreading gain (MC-VSG) CDMA system employing a binary transmission of MC signals by introducing a level clipper, termed MC-VSG BNet system, is proposed for a possible candidate of wireless personal area network (WPAN) and 3 G cellular applications. With an emphasis on the MC-VSG BNet physical layer and the system performance, we address the concise specification of the MC-VSG BNet system including the spreading code, level clipping, modulation, coding, and frame format. Especially, we focus on the level clipping of multi-level MC signals for both power- and cost-efficient implementation and the VSG code design fir high-rate transmissions. From the receiver performance based on simulation results, in addition to simple receiver structure, an acceptable performance degradation of the MC-VSG BNet system over the existing DS/CDMA system is observed, while guaranteeing an high bit rate transmission.

In this paper, we propose a new neural filter to which the features related to a given task are input, called a neural filter with features (NFF), to improve further the performance of the conventional neural filter. In order to handle the issue concerning the optimal selection of input features, we propose a framework composed of 1) manual selection of candidates for input features related to a given task and 2) training with automatically selection of the optimal input features required for achieving the given task. Experiments on the proposed framework with an application to improving the image quality of medical X-ray image sequences were performed. The experimental results demonstrated that the performance on edge-preserving smoothing of the NFF, obtained by the proposed framework, is superior to that of the conventional neural and dynamic filters.

This paper investigates the effect of fast cell selection (FCS) associated with fast packet scheduling methods and hybrid automatic repeat request (HARQ) with Chase combining, in which the optimum cell (or sector) transmitting a slot-assigned downlink shared channel (DSCH) is selected based on the received signal-to-interference power ratio (SIR), in high-speed downlink packet access (HSDPA). The Round robin (RR), Proportional fairness (PF) and Maximum carrier-to-interference power ratio (CIR) schedulers are used as the scheduling algorithm. The simulation results elucidate that although almost no additional diversity gain through FCS is obtained for the PF and Maximum CIR schedulers, the improvement in throughput by FCS coupled with the RR scheduler is achieved. Furthermore, we elucidate that the effect of FCS is small when only inter-sector FCS is performed; however, inter-cell FCS is effective in improving the radio link throughput for the access users with a lower received SIR near the cell edge. The radio link throughput at the cumulative distribution of 20% of soft handover users when both inter-sector and inter-cell FCS are performed is increased by approximately 20% and 60% for PF and RR schedulers, respectively, compared to that without FCS, i.e. with hard handover. We also show that when a traffic model such as the modified ETSI WWW browsing model is taken into account, the effect of FCS associated with the decreasing effect of fast packet scheduling is greater than that assuming continuous packet transmission. The user throughput at the cumulative distribution of 20% employing both inter-sector and inter-cell FCS is increased by approximately 60% compared to that without FCS.

Feature subset selection is an important preprocessing task for pattern recognition, machine learning or data mining applications. A Genetic Algorithm (GA) with a fuzzy fitness function has been proposed here for finding out the optimal subset of features from a large set of features. Genetic algorithms are robust but time consuming, specially GA with neural classifiers takes a long time for reasonable solution. To reduce the time, a fuzzy measure for evaluation of the quality of a feature subset is used here as the fitness function instead of classifier error rate. The computationally light fuzzy fitness function lowers the computation time of the traditional GA based algorithm with classifier accuracy as the fitness function. Simulation over two data sets shows that the proposed algorithm is efficient for selection of near optimal solution in practical problems specially in case of large feature set problems.

Terrestrial radio links with sparsely distributed multipath delays can be represented by a tapped-delay line with a few significant tap coefficients. This letter presents criteria and performance of identification methods that determine channel taps with significant power. In particular, a tap identification method derived from the maximum-likelihood criterion and its closed form error probabilities are presented. Performance improvement over a previously reported scheme is quantified using the derived error probabilities.

This paper presents novel flooding schemes for wireless mobile ad hoc networks. Clustering of nodes is assumed as a basic ad hoc network structure. GWF (Gateway Forwarding) and SGF (Selected Gateway Forwarding) are presented based on clustering. A new protocol, termed FGS (Flooding Gateway Selection) protocol, between a cluster head and its gateways to realize SGF is presented. It is shown that SGF significantly improves the packet delivery performance in ad hoc networks by reducing flooding traffic.

Feature subset selection basically depends on the design of a criterion function to measure the effectiveness of a particular feature or a feature subset and the selection of a search strategy to find out the best feature subset. Lots of techniques have been developed so far which are mainly categorized into classifier independent filter approaches and classifier dependant wrapper approaches. Wrapper approaches produce good results but are computationally unattractive specially when nonlinear neural classifiers with complex learning algorithms are used. The present work proposes a hybrid two step approach for finding out the best feature subset from a large feature set in which a fuzzy set theoretic measure for assessing the goodness of a feature is used in conjunction with a multilayer perceptron (MLP) or fractal neural network (FNN) classifier to take advantage of both the approaches. Though the process does not guarantee absolute optimality, the selected feature subset produces near optimal results for practical purposes. The process is less time consuming and computationally light compared to any neural network classifier based sequential feature subset selection technique. The proposed algorithm has been simulated with two different data sets to justify its effectiveness.

This paper presents a novel design method for reducing the complexity of the design procedure for diversity antennas on the hand-held phone. Recently, antenna selection diversity has been widely used for hand-held phones in order to overcome a problem of fading. A monopole antenna and an inverted-F antenna are the typical combination for this purpose. In the case of the conventional design method, the mutual coupling between two antennas are used for improving the diversity performance. However, strong mutual coupling often makes the diversity antenna design difficult and degrades the radiation performance. The proposed design method suppresses this coupling by tuning the terminating impedance on the unselected antenna and improves the diversity performance by modifying the shape of inverted-F antenna. The validity of the proposed method is investigated under the effect of the user's hand and head by FDTD simulation.

Most of the image restoration filters proposed so far include parameters that control the restoration properties. For bringing out the optimal restoration performance, these parameters should be determined so as to minimize a certain error measure such as the mean squared error (MSE) between the restored image and original image. However, this is not generally possible since the unknown original image itself is required for evaluating MSE. In this paper, we derive an estimator of MSE called the subspace information criterion (SIC), and propose determining the parameter values so that SIC is minimized. For any linear filter, SIC gives an unbiased estimate of the expected MSE over the noise. Therefore, the proposed method is valid for any linear filter. Computer simulations with the moving-average filter demonstrate that SIC gives a very accurate estimate of MSE in various situations, and the proposed procedure actually gives the optimal parameter values that minimize MSE.

Using moment generating function (MGF) of generalized selection combining (GSC) output signal-to-noise ratio (SNR), we derive closed-form expressions for average combined SNR at the output of GSC, which optimally combines the N largest out of L available diversity signals, over Nakagami-m fading channels for N = 2, 3 and L = 4. The Nakagami-m fading statistics on each diversity branch are assumed to be independent and identically distributed (i.i.d.). The average combined SNRs at the outputs of GSC receivers are also compared with the average combined SNRs at the outputs of conventional maximal ratio combining (MRC) and selection combining (SC).

Optimum antenna sub-set selection in MIMO systems is an attractive cost reducing technique. In this paper we develop an optimal antenna sub-set selection technique to be used in conjunction with space-time block codes over a MIMO link to optimize link error performance over a fading channel. We study the case when antenna sub-set selection is applied either at the transmitter or the receiver. We provide analytical results for substantial improvement in average SNR and outage capacity when antenna sub-set selection is used. Simulation results that verify our analytical prediction are also presented.

Orthogonal Frequency Division Multiplexing-Direct Sequence/Code Division Multiple Access (OFDM-DS/CDMA) systems provide frequency diversity gain avoiding inter symbol interference (ISI) in a frequency selective fading channel. However, path diversity gain can not be obtained by using conventional OFDM-DS/CDMA schemes. This paper proposes a new multiple antenna transmission system with combined path diversity and frequency diversity. Signal of each antenna is delayed by several chips to create artificial path diversity as well as frequency diversity of multi-carrier transmission in which can then be combined by using a RAKE receiver. Therefore multiple antenna transmission scheme creates a path diversity effect on uncorrelated signals in multi-carriers from each antenna. The received uncorrelated signals can be processed by Maximum Ratio Combining (MRC) diversity without ISI at a RAKE receiver even when we use FFT modulation. As a result, we can obtain combined path diversity and frequency diversity gain effectively by the RAKE system with the combination of multiple antennas.

Since the traffic of NetNews is increasing, keeping all articles becomes serious problem from a viewpoint of waste of network bandwidth and the amount of disk usage. In addition, users read not all incoming articles. We have proposed several caching algorithms to overcome this problem and shown that a selective prefetch scheme gives the best system performance among the proposed ones. However, since the selective prefetch scheme employed a simple selecting policy, the scheme gave low hit ratio in some cases. Therefore, this paper intends to improve the selective prefetch scheme from a viewpoint of the amount of disk usage as well as hit ratio. In this paper, we divide the scheme into three factors: reference span, criterion, and threshold in criterion. Through simulation experiments using actual NetNews logs, we investigate the influence of the factors of the reference span and the threshold to system performance. As a result, it is shown that the reference span is more significant factor than the threshold, the selective prefetch scheme with a value around the seven days reference span keeps high hit ratio and reduces the amount of disk usage.

Many experimentally and theoretically based models have been proposed to predict, quantitatively evaluate, and combat the fading phenomenon in mobile communication systems. However, to the best of the authors' knowledge, up to now there is no objective method to determine which is the most suitable distribution to model the fading phenomenon based on experimental data. In this work, the Minimum Description Length (MDL) criterion for model selection is proposed for that purpose. Furthermore, the MDL analysis is performed for some of the most widely used fading models based on measurements taken in a sub-urban environment.

Recently, antenna selection diversity has been widely used for hand-held phones to overcome a fading problem. A monopole antenna (MPA) and an inverted-F antenna (IFA) are the typical antennas used for this purpose. However, strong mutual coupling generally appears between these two antennas and often makes the diversity antenna design difficult. In particular, in the case that the MPA is unselected antenna the mutual coupling can be minimized using the open terminating impedance. On the other hand, in the case that the IFA is unselected antenna the terminating impedance, which can minimize the mutual coupling, has not been clarified. This paper presents a novel analytical method for optimizing the terminating impedance of the IFA. The method exploits the Z-matrix, and the final expression of the terminating impedance is expressed by self- and mutual-impedance. The numerical and experimental results confirm that the proposed optimization method is effective for minimizing the mutual coupling.