A CMOS voltage-to-current converter in weak inversion is presented in this Letter. It can operate for low supply voltage and its power consumption is also low. As the input voltage varies from -0.15 V to 0.15 V, the measured maximum linearity error for the proposed voltage-to-current converter, is about 3.35%. Its power consumption is only 26 µW under the supply voltage of 2 V. The proposed voltage-to-current converter has been fabricated in a 0.5 µm N-well CMOS 2P2M process. The proposed circuit is expected to be useful in analog signal processing applications.

This paper takes full advantage of polarimetric scattering parameters and total power to classify polarimetric SAR image data. The parameters employed here are total power, polarimetric entropy, and averaged alpha angle (alphabar). Since these parameters are independent each other and represent all the scattering characteristics, they seem to be one of the best combinations to classify Polarimetric Synthetic Aperture Radar (POLSAR) images. Using unsupervised classification scheme with iterative Maximum Likelihood classifier, it is possible to decompose multi-look averaged coherency matrix with complex Wishart distribution effectively. The classification results are shown using Pi-SAR image data set comparing with other representative methods.

An RF CMOS active inductor with a novel loss compensation circuit network is proposed. Performance of this active inductor can be improved by adding a novel network, which simultaneously reduces parallel and series losses. Consequently, this technique not only increases Q value, inductance, and operating frequency, but also reduces power consumption and circuit complexity. Simulation results show that better performance indices can be achieved, such as minimum total equivalent loss of 1 mΩ, maximum Q value about 3E5, and inductance value from 20 nH to 45 nH in the RF range of 0.6 GHz to 1.6 GHz. Power dissipation is around 1.76 mW under 2.5 V dc supply voltage.

In this paper we give the necessary and sufficient conditions for 2-dimensional discrete-time systems described by the signum function to be stable.

This paper presents a network architecture with a dual interface IP handoff technique that allows smooth node mobility without using any intermediate proxy. The proposed architecture is suitable for low bit-rate time sensitive real time applications, where payload tends to be short and packet header overhead is particularly significant. Connections are established as per permanent addresses of the nodes but are carried on by the IP layer according to the temporary addresses by address translation within the end hosts. The mapping information is maintained by database servers, which can be placed in the Internet in a distributed manner. We describe the architecture and show its mobile capabilities by prototype implementation and performance evaluation. Furthermore a dual-interface handoff suitable to the proposed architecture is also introduced. Preliminary results show that the proposed architecture has significantly low overheads. It is compatible with the existing infrastructure and works fine in both IPv4 and IPv6 environments. Analysis also shows that with dual-interface handoff it is possible to achieve seamless handoff without any packet loss by exploiting overlapping coverage area and speed of the mobile node. Handoff latency is reduced significantly as compare to MIPv6. We believe that with more powerful network interface card drivers our concept of dual interface handoff can be realized.

In this paper, we propose a method of linear time-varying filtering of discrete time signals. The objective of this method is to derive a component, of an input signal, whose alias-free generalized discrete time-frequency distribution [Jeong & Williams 1992] concentrates on a specific region of a time-frequency plane. The method is essentially realized by computing an orthogonal projection of an input onto a subspace that is spanned by orthonormal signals, whose distributions concentrate on the region. We show that such orthonormal signals can be derived as eigenvectors of a matrix whose components are explicitly expressed by using the kernel of the distribution and the regions. This result shows that we can design such a filter prior to processing of the input if the specific region is given as a priori. This result is a generalization of [Hlawatsch & Kozek 1994], that is originally derived for the continuous Wigner distributions, to the discrete distributions.

The proposed comparator includes high gain preamplifier with a new swing limiter. It is shown that, for a given unity gain bandwidth, the high gain preamplifier of high output impedance can be made faster than the low gain one if properly combined with a high-speed low-level swing limiter.

Recently, Muramatsu proposed source coding algorithms that use the randomness of a past sequence. The technique of his source coding algorithms is one method of constructing codes from the technique of random coding. By using his technique, we propose a channel coding algorithm with random numbers which can be observed by both the encoder and the decoder where the random numbers are independent of the messages to be transmitted. Then the proposed coding algorithm can transmit messages over a discrete memoryless channel up to the channel capacity with an arbitrarily small decoding error rate and arbitrarily small bits of random numbers per message transmission asymptotically.

In this paper, we propose an advanced structure of the interface circuit, called a wrapper, for Globally Asynchronous Locally Synchronous (GALS) systems. The proposed wrapper is composed of a sender module and a receiver module. The sender module carries out data transfers in an efficient way by decoupling dependency between an external handshake protocol and an internal clock. The decoupling effect allows the external handshake protocol and the internal clock to be executed in a concurrent way and hence allows the wrapper to show better performance. We have designed our wrapper at the transistor level with 0.35-µm technology. When we compare our decoupled wrapper with two conventional wrappers based on pausible clocking scheme, our simulation results show that performance improvement is about 8-13% and 13-56%, respectively.

Applying the formerly proposed classification framework for first order line search optimization techniques we introduce novel superlinear first order line search methods. Novelty of the methods lies in the line search subproblem. The presented line search subproblem features automatic step length and momentum adjustments at every iteration of the algorithms realizable in a single step calculation. This keeps the computational complexity of the algorithms linear and does not harm the stability and convergence of the methods. The algorithms have none or linear memory requirements and are shown to be convergent and capable of reaching the superlinear convergence rates. They were practically applied to artificial neural network training and compared to the relevant training methods within the same class. The simulation results show satisfactory performance of the introduced algorithms over the standard and previously proposed methods.

The goal of source localization in the brain is to estimate a set of parameters for representing source characteristics; one of such parameters is the source number. We here propose a method combining the Powell algorithm with the information criterion method for determining the optimal dipole number. The potential errors can be calculated by the Powell algorithm with the concentric 4-sphere head model and 32 electrodes, then the number of dipoles is determined by the information criterion method with the potential errors mentioned above. This method has the advantages of a high identification accuracy of dipole number and a small number of EEG data because in this method: (1) only one EEG topography is used in the computation, (2) 32 electrodes are used to obtain the EEG data, (3) the optimal dipole number can be obtained by this method. In order to prove our method to be efficient, precise and robust to noise, 10% white noise is introduced to test this method theoretically. Some investigations are presented here to show our method is an advanced approach for determining the optimal dipole number.

We propose a new data link protocol with an adaptive frame length control scheme for satellite networks. The wireless communication channel in satellite networks is subject to errors that occur with time variance. The frame length of the data link layer is another important factor that affects throughput performance in dynamic channel environments. If the frame length could be chosen adaptively in response to changes in the dynamically varying satellite channel, maximum throughput could be achieved under both noisy and non-noisy error conditions. So, we propose a frame length control scheme that acts adaptively to counter errors that occur with time variance. We model the satellite channel as a two-state Markov block interference (BI) model. The estimation of the channel error status is based on the short-term bit error rate and the duty cycle of noise bursts. Numerical and computer simulation results show that the proposed scheme can achieve high throughput for both dense and diffuse burst noise channels.

DNA Sequence Design Problem is a crucial problem in information-based biotechnology such as DNA computing. In this paper, we introduce a powerful design strategy for DNA sequences by refining Random Generator. Random Generator is one of the design strategies and offers great advantages, but it is not a good algorithm for generating a large set of DNA sequences. We propose a Two-Step Search algorithm, then show that TSS can generate a larger set of DNA sequences than Random Generator by computer simulation.

In this letter, a new extended recursive least squares (RLS) algorithm is proposed for the identification of fading channels. We extend the standard RLS algorithm by converting the linear regression model into a state-space model. The unknown terms of the extended model are obtained by estimating the values which minimize the mean squared error (MSE). The proposed algorithm has lower computational complexity than the Kalman filter combined with the hypermodel described in, and exhibits superior performance in simulation than the existing RLS algorithms, namely the exponentially weighted RLS algorithm with a fixed forgetting factor (EW-RLS), and the RLS algorithm with a variable forgetting factor (VFF-RLS).

We had previously proposed a fuzzy logic-based buffer management scheme (BMS) called fuzzy early detection (FED), which was designed to improve transmission control protocol (TCP) performance over the unspecified bit rate (UBR) service of asynchronous transfer mode (ATM) networks. Since a weakness in FED was discovered later, we present a refined version of it named FED+ here. Maintaining the design architecture and the algorithm of FED, FED+ further adopts a specific per virtual connection accounting algorithm to achieve its design goals. The effects of TCP implementation, TCP maximum segment size, switch buffer size and network propagation delay on FED+ performance are studied through simulation. Its performance is then compared with those of pure early packet discard (EPD), P-random early detection (P-RED) and FED. Our evaluations show that FED+ is superior to the others if the issues of efficiency, fairness, robustness, buffer requirement and the ease of tuning control parameters of a BMS are considered collectively.

An automatic-repeat-request (ARQ) scheme for improving the system throughput efficiency is evaluated in coded multiple-input multiple-output (MIMO) transmissions. Supplementary trellis-coded modulation (TCM) code has been proposed for hybrid ARQ schemes. The free distance of the TCM after code combining can be increased by employing different TCM codes for retransmissions. The MIMO scheme offers additional flexibility in preventing successive frame errors by changing the connections between transmitters and transmit antennas upon retransmission. In this paper, an ARQ strategy employing both TCM reassignment and antenna permutation technique is investigated. It is shown through computer simulations that this ARQ scheme achieves high throughput even in severe conditions of low signal-to-noise ratio and high Rician factor over spatially and temporally correlated Nakagami-Rice fading MIMO channels.

We numerically and experimentally investigated the upgradability of the longest and the typical segments of the JIH system. Through these studies, we confirmed that a 100 GHz-spaced 25 42.7 Gbit/s transmission with the total capacity of 1 Tbit/s can be attainable even by using NRZ signal and standard FEC for the typical segments. We also found that RZ signal format was desirable for the longest segment and a further wide system margin could be expected by using adjacent channel polarization control and advanced FEC technologies.

In this paper, we consider QoS-guaranteed wavelength allocation for WDM networks with limited-range wavelength conversion. In the wavelength allocation, the pre-determined number of wavelengths are allocated to each QoS class depending on the required loss probability. Moreover, we consider two wavelength selection rules and three combinations of the rules. We analyze the connection loss probability of each QoS class for a single link using continuous-time Markov chain. We also investigate the connection loss probability for a uni-directional ring network by simulation. In numerical examples, we compare connection loss probabilities for three combinations of selection rules and show how each combination of selection rules affects the connection loss probability of each QoS class. Furthermore, we show how wavelength conversion capability affects the connection loss probability. It is shown that the proposed allocation with appropriate wavelength selection rule is effective for QoS provisioning when the number of wavelengths is large. We also show the effective combination of wavelength selection rules for the case with small wavelength conversion capability.

An information retrieval (IR) system with query expansion on a low-cost high-performance PC cluster environment is implemented. We study how query performance is affected by query expansion and two declustering methods using two standard Korean test collections. According to the experiments, the greedy method shows about 20% enhancement overall when compared with the lexical method.

In this article we analyze the performance of a space division output buffered multichannel switch operating in an ATM multimedia environment as follows. Fixed size packets belonging to two classes arrive onto the switch inputs in each time slot. Class-1 packets, representing real time traffic such as live audio and live video communications, are sensitive to delay but insensitive to loss and have their own service time needs. Class-2 packets, representing nonreal time communications such as file transfers, are insensitive to delay but sensitive to loss and have their different service time needs. To respond to the class-1 delay sensitivity, the switch gives class-1 packets higher service priority over class-2. And to respond to the difference in service time needs, the switch operates at two service rates, one for each class. This latter assumption is the major feature of the article, as previous studies have usually assumed that the two classes have the same service needs and thus the same service rate. For the purpose of the analysis, the switch is modelled as a priority, discrete time, batch arrival, multiserver queueing system, with infinite buffer and two geometric service times with two parameters. Performance measures analyzed are system occupancy and packet waiting time.