The authors have performed a simple computer simulation for a topography that models change in propagation characteristics due to change in traffic volume. The results of this simulation revealed that path loss for a traffic volume of about 2000 vehicles every 30 minutes in a typical urban environment exhibits a Rayleigh distribution. This result agrees well with that of actual measurements demonstrating that even a simple simulation can be a useful tool in system design.

In this paper, we propose an alternative method that does not generate a test for each path delay fault directly to generate tests for path delay faults. The proposed method generates an N-propagation test-pair set by using an Ni-detection test set for single stuck-at faults. The N-propagation test-pair set is a set of vector pairs which contains N distinct vector pairs for every transition faults at a check point. Check points consist of primary inputs and fanout branches in a circuit. We do not target the path delay faults for test generation, instead, the N-propagation test-pair set is generated for the transition (both rising and falling) faults of check points in the circuit. After generating tests, tests are simulated to determine their effectiveness for singly testable path delay faults and robust path delay faults. Results of experiments on the ISCAS'85 benchmark circuits show that the N-propagation test-pair sets obtained by our method are effective in testing path delay faults.

In this letter, we propose the Low-Density Parity-Check (LDPC) coded Orthogonal Frequency Division Multiplexing (OFDM) systems to improve the error rate performance of OFDM. We also evaluate the iterative decoding performance on both an AWGN and a frequency-selective fading channels. We show that when the energy per information bit to the noise power spectral density ratio Eb/N0 is not small, the LDPC coded OFDM (LDPC-COFDM) systems have the good error rate performance with a small number of iterations. We also show that when the Eb/N0 is small, the BER of the LDPC-COFDM systems is worse than that of the Turbo coded OFDM (TCOFDM) systems, while when the Eb/N0 is not small, the BER of the LDPC-COFDM systems is better with a small number of iterations.

In this letter, we show the effectiveness of a double-loop algorithm based on the concave-convex procedure (CCCP) in decoding linear codes. For this purpose, we numerically compare the error performance of CCCP-based decoding algorithm with that of a conventional iterative decoding algorithm based on belief propagation (BP). We also investigate computational complexity and its relation to the error performance.

Richardson and Urbanke developed a powerful method density evolution which determines, for various channels, the capacity of irregular low-density parity-check code ensembles. We develop generalized density evolution for minutely represented ensembles and show it includes conventional representation as a special case. Furthermore, we present an example of code ensembles used over binary erasure channel and binary input additive white Gaussian noise channel which have better thresholds than highly optimized ensembles with conventional representation.

Brain subsystems have a high degree of information processing ability using nonlinear dynamics and although various neuron models and artificial neural networks have been investigated, the information processing functions of biological neural networks have not yet been clarified. Recently, various research efforts have confirmed that dendrites perform an important role in brain information processing. In this paper, we discuss the nonlinear characteristics of a hardware active dendrite model, in order to clarify information encoding and transmission via action potentials. That is to say, we show that our proposed model can reproduce the nonlinear characteristics of a biologically active dendrite. First, the hardware active dendrite model we propose is described. We next discuss the response characteristics for pulse stimuli using the model. As a result, when input pulses are applied to an active line, which is the basic structure of the dendrite model, it is shown clearly that backpropagation characteristics are acquired and that the characteristics are qualitatively in agreement with the characteristics of biological dendrites. Furthermore, we verify that the ratio of input to output frequency at the cell body is influenced by the backpropagation characteristics with two branches, which is the simplest structure in the active dendrite model. Thus, with backpropagation characteristics, the possibility that the model can carry out clearly the information processing of biological neural networks, is suggested.

In this paper, a Compensatory Neuro-Fuzzy Network (CNFN) for nonlinear system control is proposed. The compensatory fuzzy reasoning method is using adaptive fuzzy operations of neural fuzzy network that can make the fuzzy logic system more adaptive and effective. An on-line learning algorithm is proposed to automatically construct the CNFN. They are created and adapted as on-line learning proceeds via simultaneous structure and parameter learning. The structure learning is based on the fuzzy similarity measure and the parameter learning is based on backpropagation algorithm. The advantages of the proposed learning algorithm are that it converges quickly and the obtained fuzzy rules are more precise. The performance of CNFN compares excellently with other various exiting model.

This paper proposes an analytical expression for the maximum operating frequency of an emitter-coupled-logic master-slave toggle flip-flop (ECL MS TFF) based on an impulse response method. The analytical expression was in good agreement with not only SPICE simulations, but also experimental values. The analytical expression also indicated that state-of-the-art InP-based heterojunction bipolar transistors have potential to achieve over 100-GHz operation in ECL MS TFFs. Also, the proposed method was applied to the maximum operating frequency of a source-coupled FET logic (SCFL) MS TFF.

This paper describes a new learning method for Multiple-Value Logic (MVL) networks using the local search method. It is a "non-back-propagation" learning method which constructs a layered MVL network based on canonical realization of MVL functions, defines an error measure between the actual output value and teacher's value and updates a randomly selected parameter of the MVL network if and only if the updating results in a decrease of the error measure. The learning capability of the MVL network is confirmed by simulations on a large number of 2-variable 4-valued problems and 2-variable 16-valued problems. The simulation results show that the method performs satisfactorily and exhibits good properties for those relatively small problems.

A new multilayer artificial neural network learning algorithm based on the pattern search method is proposed. The learning algorithm is designed to provide a very simple and effective means of searching the minima of an objective function directly without any knowledge of its derivatives. We test this algorithm on benchmark problems, such as exclusive-or (XOR), parity and alphabetic character learning problems. For all problems, the systems are shown to be trained efficiently by our algorithm. As a simple direct search algorithm, it can be applied to hardware implementations easily.

Probabilistic inference by means of a massive probabilistic model usually has exponential-order computational complexity. For such massive probabilistic model, loopy belief propagation was proposed as a scheme to obtain the approximate inference. It is known that the generalized loopy belief propagation is constructed by using a cluster variation method. However, it is difficult to calculate the correlation in every pair of nodes which are not connected directly to each other by means of the generalized loopy belief propagation. In the present paper, we propose a general scheme for calculating an approximate correlation in every pair of nodes in a probabilistic model for probabilistic inference. The general scheme is formulated by combining a cluster variation method with a linear response theory.

Radio access networks (RANs) for new generation mobile communication systems are required to construct economical high capacity networks. An optical wireless link is expected as a solution to the linking method in the construction of these networks. The optical wireless link provides high bit-rate transmission and allows for easy installation. However, optical-waves are severely attenuated in free-space transmission due to weather conditions and cannot provide a high level of link availability in long-distance communications. This paper describes the applicability of an optical wireless link to the RAN based on obtained atmosphere propagation properties from experiments on an 800-nm band optical wireless link from the viewpoint of link availability. The experimental results show that the BER performance of the optical wireless link exhibits a sufficiently low power penalty of less than 1 dB compared to that of the optical fiber link. Moreover, the results confirm that atmosphere attenuation of an 800-nm band optical wave due to fog and rainfall can be estimated by using a simple estimation equation. The optical wireless link could be used for RAN approach links with the range of less than 350 m to achieve the link outage rate of 0.004% from the viewpoint of the link budget design.

Wall shadow removal problems differ due to configuration of walls in different propagation scenarios. Applying Weiler-Atherton polygon clipping method helps calculate illuminated regions on the building walls, but unfortunately the technique has limitation when there are many walls and the walls configuration is complex. The modified Weiler-Atherton polygon clipping method proposed can solve the problem by regarding all vertices of the subject polygon or clipping polygon, that are also intersection points as simply intersection points. In the case that a certain vertex of the subject polygon is a vertex of the clipping polygon, this vertex of the subject polygon is still considered a vertex. It is found that wall shadow removal using the proposed modified Weiler-Atherton polygon clipping method is more efficient.

Though analysis of an indoor propagation channel has been conventionally used the ray-tracing method, in this paper using three dimension of finite difference time domain methods can easily and exactly be obtained three-dimensional complex structures. An excitation signal of FDTD made use of plane wave. An absorbing boundary condition used the most reflection less perfectly matched layer in the outset plane. An empty room surrounded a wall composed of brick, concrete and case that there are indoor furniture in the room were simulated. As simulation outcome, we could identify frequently rising reflection, refraction, scattering of objects and a fading effect of multipath at indoor propagation environment, calculate mean excess delay and rms delay spread for receiver design.

This paper describes a design of a Carry Propagation Free Adder/Subtracter (CPFA/S) VLSI using the Adiabatic Dynamic CMOS Logic (ADCL) circuit technology. Using a PSPICE simulator, energy dissipation of the ADCL 1 bit CPFA/S is compared with that of the CMOS 1 bit CPFA/S. As a result, energy dissipation of the proposed ADCL circuits is about 1/3 as high as that of the CMOS circuits. The transistors count, propagation-delay time and energy dissipation of the ADCL 4 bit CPFA/S are compared with those of the ADCL 4 bit Ripple Carry Adder/Subtracter (RCA/S). The transistors count and propagation-delay time are found to be reduced by 7.02% and 57.1%, respectively. Also, energy dissipation is found to be reduced by 78.4%. Circuit operation and performance are evaluated using a chain of the ADCL 1 bit CPFA/S fabricated in a 1.2 µm CMOS process. The experimental results show that addition and subtraction are operated with clock frequencies up to about 1 MHz. In addition, the total power dissipation of the ADCL 1 bit CPFA/S is 28.7 µW including the power supply.

It is found that the supercontinuum spectrum is generated from cross-phase modulated soliton pulses which are propagated through a dispersion-flattened/decreasing fiber with low birefringence. The cross-phase modulation is achieved by exciting two orthogonally polarized modes in a birefringent fiber and the effect of input azimuth of linearly polarized pulses is discussed theoretically and numerically.

A bent-waveguide-based multimode interference (MMI) demultiplexer is designed for the operation at 0.85- and 1.55-µm wavelengths using the three-dimensional semi-vectorial beam-propagation method. First, it is shown that the use of a straight MMI waveguide results in a long coupler length of more than 1000µm for wavelength demulitplexing. To reduce the coupler length, we next introduce a bent MMI waveguide. Bending with a radius of 1500µm leads to a coupler length of less than 200µm. After designing two output waveguides connected to the MMI section, we finally choose a coupler length to be 175µm for efficient demultiplexing properties. Consequently, an output power of more than 90% can be obtained, leading to a low insertion loss of 0.34dB at both 0.85- and 1.55-µm wavelengths. The demultiplexer achieves small polarization dependence, i.e., less than 2dB difference in contrast and 0.02dB difference in insertion loss.

We have classified parenchymal echo patterns of cirrhotic liver into four types, according to the size of hypo echoic nodular lesions. The NN (neural network) technique has been applied to the characterization of hepatic parenchymal diseases in ultrasonic B-scan texture. We employed a multilayer feedforward NN utilizing the back-propagation algorithm. We extracted 1616 pixels in the two-dimensional regions. However, when a large area is used, input data becomes large and much time is needed for diagnosis. In this report, we used DCT (discrete cosine transform) for the feature extraction of input data, and compression. As a result, DCT was found to be suitable for compressing ultrasonographic images.

One of the biggest limitations of BP algorithm is its low rate of convergence. The Variable Learning Rate (VLR) algorithm represents one of the well-known techniques that enhance the performance of the BP. Because the VLR parameters have important influence on its performance, we use learning automata (LA) to adjust them. The proposed algorithm named Adaptive Variable Learning Rate (AVLR) algorithm dynamically tunes the VLR parameters by learning automata according to the error changes. Simulation results on some practical problems such as sinusoidal function approximation, nonlinear system identification, phoneme recognition, Persian printed letter recognition helped us better to judge the merit of the proposed AVLR method.

Ray tracing is an efficient method for analyzing transmission characteristic of indoor wireless systems. However for simulating the transmission characteristic, using a path profile obtained by ray tracing, calculation times become enormous and there is no good theoretical model which can link a path analysis result with digital transmission characteristics evaluation. To overcome such problems, in this paper, a simple calculation method on spatial distribution of error occurrence due to intersymbol interference (ISI) based on "the equivalent transmission-path (ETP) model" is proposed. The ETP model is a technique that can simply estimate statistics of errors due to ISI that arise in Rayleigh and Nakagami-Rice fading environments. If a simple calculation method proposed in this paper is used, calculation time of digital transmission characteristics evaluation become tremendously shorter and results of this method agree with those of exact simulations with sufficient accuracy.