We examine a concatenated code which consists of a rate 1/2, 4-state turbo code (an inner code) and a single-parity-check product code (an outer code), and discuss the decoding structure called a double concatenated decoding scheme. From our Monte Carlo simulation trials, we show the advantage of the concatenated codes over turbo codes alone. Specifically, when we use an interleaver of 4096 bits, the Eb/No to obtain a BER of 10-6 is about 1.45 dB for the concatenated code. On the other hand, it is more than 2.5 dB for the turbo code alone. So, the Eb/No improvement can be achieved by about 1 dB. This improvement in Eb/No was also obtained for the interleavers of 8192 and 2048 bits. Therefore, the concatenated codes using a double concatenated decoding scheme can solve the problem of the BER flattening in decoding of turbo codes.

Common-mode (CM) radiation from a cable attached to a conducting enclosure has a typical dipole-type antenna structure, in which an equivalent noise voltage source located at the connector excites the attached cable against the enclosure to produce radiated emissions. Based on this mechanism, a simple method for predicting the CM radiation from the cable/enclosure structure was proposed. The method combines an equivalent dipole approximation with sinusoidal current distribution and CM current measurement at a specified location on the cable. Its validity was examined in comparison with the far-field measurement and finite-difference time-domain (FDTD) modeling. The predicted resonance frequencies and CM radiation levels were validated with engineering accuracy, i.e., within 30 MHz and 6 dB, respectively, from the measured and FDTD-modeled results in the frequencies above 150 MHz.

Dual-Banyan is a buffered banyan asynchronous transfer mode (ATM) switch encompassing bifurcated queueing as its buffering strategy. This paper describes an efficient analytical model, based on iterative calculations, for performance evaluation of the Dual-Banyan switch under uniform and non-uniform traffic patterns with much less time than the simulation. The efficiency of the given model is verified through a comparison with simulation results. Another benefit of our model is the possibility to adopt it in any non-uniform incoming traffic. At last, we compare performance of Dual-Banyan switch and Input Buffer Banyan, and show that Dual-Banyan switch has significant better performance levels.

We have measured the non line-of-sight (NLOS) propagation characteristics of microwave frequencies in an urban environment with a base station antenna situated well above the surrounding buildings. When these characteristics are compared with the results of measurements made in the same environment with a low base station antenna height, it can be seen that with a low base station antenna height the attenuation coefficient varies greatly between line-of-sight (LOS) and NLOS environments, whereas with a high base station antenna height there is no variation of this sort. This is because the waves arriving NLOS environments from a high base station antenna do so primarily as a result of rooftop diffraction, and the path loss does not vary much over regions of equal distance between the base station and mobile station. We have confirmed that the frequency characteristics of relative loss in NLOS environments with a high antenna height follow a relationship of 22.8 log f, which is more or less the same as the characteristic for the UHF band. By modifying the frequency terms of the Sakagami model (used for UHF band) based on this trend to allow it to handle microwave frequencies, a close correspondence is seen between the results of actual measurements and the values predicted by the extended model.

Wireless network systems introducing both of the cellular concept and the ad-hoc concept have been proposed. Communication between two nodes in a cell is guaranteed by relaying capability of the base station in these systems. Additionally, two nodes can directly communicate with each other while they are close to each other. We call this type of network a two-hop wireless network. The teletraffic performance of this network depends on various parameters such as the size of a cell, location of nodes, the communication range of nodes, channel assignment schemes, teletraffic behavior and so on. The purpose of this paper is to theoretically analyze the teletraffic performance of the network, which has been evaluated by computer simulation, by introducing a simple model. This paper shows a technique to analyze the performance in this model. Also, this paper considers the range in which the two-hop wireless network works well for the efficient use of channels.

This paper investigates the cell search time performance of our previously proposed three-step cell search method in a two-cell site environment by laboratory and field experiments supporting asynchronous cell site operation, which is one of the most notable features of wideband direct sequence code division multiple access (W-CDMA) mobile communications. The cell search methods used in the paper are based on the ongoing third generation partnership project (3GPP), in which our original scheme was refined with respect to several points in order to reduce the complexity of the receiver. The experimental results demonstrate that the method achieves the fast cell search time of less than one second in real multipath-fading channels. The cell search is accomplished in less than approximately 700 msec at 90% of the detection probability when 4.7% and 0.5% of the total transmit power of a cell site is assigned to the common pilot channel (CPICH) and synchronization channels (SCHs), respectively, in a two-cell site environment. We also elucidate that the cell search time at the detection probability of 90% using time switched transmit diversity (TSTD) is decreased by approximately 100 msec compared to that without TSTD in low-mobility environments such as the average vehicular speed of 5 km/h with a transmit power assignment of the CPICH of 4.7%.

A noise suppression algorithm with high speech quality based on weighted noise estimation and MMSE STSA is proposed. The proposed algorithm continuously updates the estimated noise by weighted noisy speech in accordance with an estimated SNR. The spectral gain is modified with the estimated SNR so that it can better utilize the improvement in noise estimation. With a better noise estimate, a more correct SNR is obtained resulting in the enhanced speech with low distortion. Subjective evaluation results show that five-grade mean opinion scores of the new algorithm with and without a speech codec are improved by as much as 0.35 and 0.40 respectively, compared with either the original MMSE STSA or the EVRC noise suppression algorithm.

An L-band low quiescent current and low distortion SiGe heterojunction bipolar transistor (HBT) driver amplifier having a self base bias control circuit is described. Since the size of this bias circuit is small and it does not need an external control circuit, it is easy to be integrated with the driver amplifier on a single chip. According to the output power level, the self base bias control circuit, which is the combination of a constant base voltage circuit and p-metal oxide semiconductor (MOS) FET current mirror with a constant current source, automatically controls the base voltage, and allows low quiescent current at low output power level and low distortion at high output power level. The simulated results show that the driver amplifier having the self base bias control circuit achieves 1 dB power compression point (P1 dB) improvement of 2.4 dB compared with the driver amplifier having a conventional constant base voltage under the same quiescent current condition. The fabricated driver amplifier with the proposed bias circuit shows high P1 dB of 15.0 dBm with low quiescent current of 15.3 mA.

This paper proposes a multipath interference canceller (MPIC) for orthogonal code multiplexed channels in the W-CDMA forward link and evaluates by computer simulation the improvement in BER performance owing to the multipath interference (MPI) suppression effect obtained by the MPIC. The simulation results show that a one-stage MPIC, which removes the MPI from the common pilot channel (PICH), common control channel (CCH), and synchronization channel (SCH), achieves a sufficient MPI suppression effect, and that the required received Eb/N0 of the traffic channel (TCH) at the average BER of 10-3 using the MPIC for the common channels is decreased by approximately 6.5 dB compared to that with a matched filter (MF)-based Rake receiver (the transmit power ratio of each common channel to TCH: ΔPICH/TCH=0 dB, ΔCCH/TCH=5 dB, ΔSCH/TCH=3 dB, without fast transmit power control (TPC) and antenna diversity reception). We also show that by using MPIC, the required transmit Eb/N0 at the average BER of 10-3, when the ratio of the target Eb/I0 of the 9-interfering users to desired user is ΔInt/Des=6 dB with fast TPC, is increased by only approximately 0.6 dB compared to that when ΔInt/Des=0 dB. This implies that the preferential MPI suppression from high-rate TCHs that abates the increase in complexity in a mobile terminal is effective in increasing the link capacity in the forward link.

Methods that precisely compensate for propagation distortion using pilot symbols are widely used in mobile communications. We describe such a pilot-symbol-assisted technique for precise compensation of flat fading and frequency offset. This technique provides a wide range of offset compensation. Conventional methods using fast Fourier transform (FFT) compensate for both slow and fast fading, but their tolerable range of frequency offset is very limited. By composing a system with an approximate frequency estimator, we can estimate and compensate for fading and a large frequency offset at the same time. The estimation and compensation are carried out by periodic pilot symbols and no other index sequences are needed. This method enables high-data-rate transmission. We describe the method and provide a theoretical analysis for the compensable range of fading and frequency offset for a transmission frame structure with pilot symbols. Then, we evaluate the method by computer simulation.

Studies on artificial neural network have been conducted for a long time, and its contribution has been shown in many fields. However, the application of neural networks in the real world domain is still a challenge, since nature does not always provide the required satisfactory conditions. One example is the class size imbalanced condition in which one class is heavily under-represented compared to another class. This condition is often found in the real world domain and presents several difficulties for algorithms that assume the balanced condition of the classes. In this paper, we propose a method for solving problems posed by imbalanced training sets by applying the modified large-scale neural network "CombNET-II. " CombNET-II consists of two types of neural networks. The first type is a one-layer vector quantization neural network to turn the problem into a more balanced condition. The second type consists of several modules of three-layered multilayer perceptron trained by backpropagation for finer classification. CombNET-II combines the two types of neural networks to solve the problem effectively within a reasonable time. The performance is then evaluated by turning the model into a practical application for a fog forecasting problem. Fog forecasting is an imbalanced training sets problem, since the probability of fog appearance in the observation location is very low. Fog events should be predicted every 30 minutes based on the observation of meteorological conditions. Our experiments showed that CombNET-II could achieve a high prediction rate compared to the k-nearest neighbor classifier and the three-layered multilayer perceptron trained with BP. Part of this research was presented in the 1999 Fog Forecasting Contest sponsored by Neurocomputing Technical Group of IEICE, Japan, and CombNET-II achieved the highest accuracy among the participants.

This paper elucidates the optimum bandwidth per sub-carrier in the reverse link for multicarrier (MC)/DS-CDMA using a 10 to 80-MHz bandwidth in a multipath fading channel with numerous resolved multipaths, taking into account all major effects, i.e., the improvement in the Rake time diversity effect and the degradation in the path search and the channel estimation due to multipath interference (MPI). In the paper, we assume a broadband channel model with the maximum delay time of up to approximately 1 µsec simulating a microcell with the radius of less than 1 km in an urban area. The simulation results clarify that the improvement in the radio link performance is almost saturated at a bandwidth greater than approximately 40 MHz when the spreading factor of the channel is SF=32, and the best performance is achieved at the bandwidth of approximately 20-40 MHz when SF=4, employing two-branch antenna diversity reception (an average equal power delay profile and an exponential decay power delay profile are assumed, where the number of multipaths is changed from 12 to 48 for both profiles). This is generated by the tradeoff between the improvement in the Rake time diversity effect and the increased MPI in addition to the degradation in accuracy of the path search and channel estimation associated with a lower average received signal-to-interference plus background noise power ratio. Therefore, we conclude that MC/DS-CDMA, where each sub-carrier has the bandwidth of approximately 20-40 MHz, is one of the most promising candidates for broadband packet wireless access in the reverse link.

The surface film of a slip ring is important for the sliding contact phenomenon. The surface film is affected by atmospheric temperature, humidity and air pressure. The main objective of our study is to examine the effect of oxygen gas on the sliding contact phenomenon. In the present experiment, we examined the contact voltage drop for continuous sliding when the atmosphere is changed from low pressure to atmospheric pressure by introducing oxygen (O2 20%+N2 80%) or nitrogen gas. As a result, the contact voltage drop increases rapidly with increasing gas pressure, and its fluctuation also becomes large. These phenomena are observed in both cases of oxygen (O2 20%+N2 80%) and nitrogen introduction. The results clearly show that the sudden increase of contact voltage drop is affected by factors other than the oxide film. Actually, the oxide film is not formed in the nitrogen atmosphere. Furthermore, the frictional coefficient of carbon and copper ring is changed at ambient atmosphere. It is inferred from these data that the contact voltage drop may be affected by the frictional coefficient. When the gas pressure decreases again, the contact voltage drop does not suffer from the effect of ambient gas. Therefore, only the resistance of the oxide film appears to affect contact voltage drop. In this paper, the effect of sliding contact phenomenon on the contact voltage drop by gas adsorption and film generation was examined.

This paper proposes a signal-to-interference power ratio (SIR) measurement method that employs the transmission gap of a data channel (TGDC) interval for precise link adaptation, in order to eliminate the influence of severe multipath interference (MPI) from a shared packet channel and to decrease further the instantaneous variations in interference components for high-speed packet transmission in the forward link using adaptive data modulation associated with the multipath interference canceller (MPIC). Computer simulation results elucidate that the required received signal energy per chip-to-background noise power spectrum density ratio (Ec/N0) based on the SIR measurement employing TGDC at the throughput of 4.2 Mbps is decreased by approximately 2.0 dB compared to the conventional method without TGDC using chip-based interference power measurement for the number of paths L = 1, and by approximately 1.5 dB compared to the conventional method using symbol-based interference power measurement for L = 2, respectively. Therefore, we show that the adaptive data modulation with the SIR measurement exploiting the TGDC interval achieves almost the maximum (i.e., almost ideal selection) throughput, without changing the SIR measurement method according to the propagation conditions such as the number of multipaths.

This paper attempts to establish a theory for a general auto-associative memory model. We start by defining a new concept called supporting function to replace the concept of energy function. As known, the energy function relies on the assumption of symmetric interconnection weights, which is used in the conventional Hopfield auto-associative memory, but not evidenced in any biological memories. We then formulate the information retrieving process as a dynamic system by making use of the supporting function and derive the attraction or asymptotic stability condition and the condition for convergence of an arbitrary state to a desired state. The latter represents a key condition for associative memory to have a capability of learning from variant samples. Finally, we develop an algorithm to learn the asymptotic stability condition and an algorithm to train the system to recover desired states from their variant samples. The latter called sample learning algorithm is the first of its kind ever been discovered for associative memories. Both recalling and learning processes are of finite convergence, a must-have feature for associative memories by analogy to normal human memory. The effectiveness of the recalling and learning algorithms is experimentally demonstrated.

We developed a distributed control algorithm to solve the problem of a trade-off between transient response and stability. We applied it to a congestion control algorithm for transmitting best-effort packets such as transmission control protocol (TCP) packets over the Internet. A new transmission power control algorithm suitable for transmitting best-effort packets over the wireless Internet was also developed using the distributed control algorithm. We showed that in a steady state, TCP connections can use the bandwidth efficiently over both wired and wireless Internet when the proposed control algorithms are used. The transient response was also evaluated and it was found that the packet transmission rate and the transmission power adjusted by the proposed control algorithms converge to a steady state faster than when adjusted by conventional control algorithms while maintaining the stability of network systems.

For checkpointing to be practical, it has to introduce low overhead for the targeted application. As a means of reducing the overhead of checkpointing, this paper proposes a probabilistic checkpointing method, which uses block encoding to detect the modified memory area between two consecutive checkpoints. Since the proposed technique uses block encoding to detect the modified area, the possibility of aliasing exists in encoded words. However, this paper shows that the aliasing probability is near zero when an 8-byte encoded word is used. The performance of the proposed technique is analyzed and measured by using experiments. An analytic model which predicts the checkpointing overhead is first constructed. By using this model, the block size that produces the best performance for a given target program is estimated. In most cases, medium block sizes, i.e., 128 or 256 bytes, show the best performance. The proposed technique has also been implemented on Unix based systems, and its performance has been measured in real environments. According to the experimental results, the proposed technique reduces the overhead by 11.7% in the best case and increases the overhead by 0.5% in the worst case in comparison with page-based incremental checkpointing.

This paper proposes adaptive antenna array transmit diversity (AAA-TD) in the W-CDMA forward link with frequency division duplexing (FDD), based on adaptively-generated receiver antenna weights in the reverse link, which only track the changes in the average signal-to-interference power ratio (SIR) and direction of arrival (DOA) but with the calibration of the phase/amplitude variations of the parallel RF receiver/transmitter circuits corresponding to the number of array antennas. The laboratory and field experimental results exploiting AAA-TD are presented to show the strong multipath interference (MPI) suppression effect especially from high-rate users with large transmission power. Laboratory experiments elucidate that by using AAA-TD with four antennas, the required transmitted SIR before multiplying the transmitter antenna weights at the average BER of 10-3 is decreased by approximately 13 dB compared to that with one omni-directional antenna transmitter. Field experiments also show that although an error floor above 10-2 is observed with one omni-directional antenna transmitter when the transmitted SIR is -12 dB due to severe MPI, no error floor is observed when employing 4-antenna AAA-TD and the loss of the required received signal power at the average BER of 10-3 from the single-user case is suppressed to below approximately 5 dB. Therefore, we show that AAA-TD is very effective in suppressing severe MPI especially from high rate users with large transmission power due to its adaptive main lobe and null steering.

Call demand suddenly and greatly increases in the aftermath of a major disaster, because people want to check on their families and friends in the stricken area. Many call attempts in mobile cellular systems are blocked due to the limited radio frequency resources. In this paper, as a solution to this problem, limiting the holding time of calls is investigated and a dynamic holding time limit (DHTL) method, which varies the holding time limit dynamically based on the number of call attempts, is proposed. The effect of limiting the holding time is investigated first using a computer simulation with a constant and heavy traffic load model. This simulation shows that the average holding time of calls is decreased as the holding time limit is reduced. But it also shows limiting the holding time decreases the number of calls blocked and forced call terminations at handover considerably. Next, a simple estimation method for the holding time limit, which reduces the blocking rate to the normal rate for increasing call demand, is described. Finally, results are given of a simulation, which show that the DHTL method keeps good performance for a sudden and great traffic load fluctuation condition.

We derived the design requirements that wireless systems and congestion control algorithms must satisfy to transmit best-effort Internet protocol (IP) packets over wireless systems. We proved that, if these requirements are satisfied, congestion control algorithms are robust against unfairness in the systems and can provide near-maximum throughputs in various environments. From the viewpoint of the design requirements, we investigated the effect of automatic repeat request (ARQ) on the throughputs of best-effort IP connections, and showed why ARQ can improve the throughputs while too large a number of retransmissions degrade them. We also investigated the effect of variance in packet transmission rates and clarified what kind of congestion control algorithm degrades the throughputs.