Magnetic random access memory (MRAM) possesses the attractive properties of non-volatility, radiation hardness, nondestructive readout, low voltage, high access speed, unlimited read and write endurance and high density. MRAM technology is described for the devices using giant magnetoresistance (GMR) and tunneling magnetoresistance (TMR) materials in this paper. The TMR type MRAM architectures using ferromagnetic tunneling junctions (MTJ) are more attractive for mainstream RAM applications than the GMR type, because the signal of the TMR type is larger than that of the GMR type. A MRAM device with an MTJ plus MOS transistor switch architecture, which can provide large signal-to noise ratio, is detailed. A design of the MTJ element is discussed and the requirements for the junction resistance and the TMR needed for the memory device are demonstrated based on the simple signal voltage calculations. The TMR significantly decreases with increasing bias voltage, which leads to the reduction of the signal voltage for the actual MRAM. A ferromagnetic double tunneling junction is proposed for the high density MRAM application demanding large signal voltage, because of the smaller degradation of the TMR for the bias voltage, compared with that of the conventional single junctions. Recent trials of MRAM fabrication are introduced, which demonstrates high-speed access time. Finally, challenges for the higher bit density MRAM above Gb are discussed, and it is noticed that higher signal voltage, lower power consumption for writing and novel cell designs are needed for the achievement.

In distributed network systems, it is one of the most important problems how to assign the files to servers in view of cost and delay. It is obvious that there is a trading-off relationship between costs and delays in these systems. In order to evaluate the optimization that the total cost is minimized subject to the total delay, we have presented the Optimal File Allocation Model as 0-1 integer programming, and have investigated the general characteristics in distributed systems. In this model, we have introduced many cost and delay parameters to evaluate the total cost and delay in the system more exactly. In constructing practical systems, it is necessary to investigate the weight and the contribution of each parameter to the total cost. It is very useful to show how to estimate cost and delay parameters on the basis of this analysis. In this paper, we analyze the sensitivity of these parameters and make clear the influence between principal parameters.

A chain ferroelectric random-access memory (chain FeRAM) is a solution for future high-density and high-speed nonvolatile memory. One memory cell consists of one transistor and one ferroelectric capacitor connected in parallel, and one memory cell block consists of plural cells and a block selecting transistor in series. This configuration realizes small memory cell of 4F2 size and fast random access time. This paper shows an overview and trend of chain FeRAM architecture. First, the concept of chain FeRAM is presented, and basic operations including two cell-plate driving schemes are discussed. Second, assuming multi-megabit generation, ideal features and performances are discussed in terms of die size, speed and other aspects. Third, the prototype of chain FeRAM and the practical cell structure for megabit-scale memories using 0.5 µ m 2-metal CMOS process are demonstrated. By introducing fast and compact cell-plate drive technique, this prototype achieves random access time of 37-ns and read/write cycle time of 80-ns, which are the fastest speeds reported for FeRAMs. Fourth, after discussing future memory cell trend and problems respecting scaled FeRAMs, a gain cell block approach for future gigabit-scale chain FeRAMs is introduced. This realizes both a small average cell size and a large cell signal even at small cell polarization.

Effects of imprint on signal margin in FeRAM with Pt/SrBi2Ta2O9/Pt capacitors have been investigated. Imprint, induced during high temperature storage, significantly reduced the signal margin and hence determines lifetime of FeRAM. Initial signal margin of 470 mV is reduced to 290 mV after storage at 175C for 96 hours. From the reduction rate of the signal margin, it is estimated that imprint lifetime of the FeRAM is more than 10 years even though the storage temperature is 175C.

This paper proposes a λ-ring system that is a wavelength-based self-healing-ring application unlike ordinary fiber-based ones. To design survivable networks of interconnected such self-healing ring systems, a virtual mesh network scheme is used, in which wavelength assignment in virtual links can be considered according to the λ-ring-system or fiber-ring-system applications of the bidirectional wavelength-path switched architecture. Integer-programming-based design problems are then formulated that minimize the total fiber length in these self-healing-ring applications. Numerical examples show that the λ-ring-system application is always superior to 4-fiber and 2-fiber-ring-system applications and 1+1 end-to-end path protection.

Sequence estimation (SE) of narrow-band digital FM signals, such as CPFSK and GMSK, with non-coherent limiter/discriminator (L/D) and integrate and dump (I&D) detection is investigated in detail using both analysis and simulation. The BER is studied from approximate upper and lower bounds obtained through Chernoff bounding techniques and minimum error event path probability along with a Gaussian noise assumption for high SNR. Various IF filters and the dependence of the error probability upon modulation index are considered. The results show an optimum modulation index around h 0.55, and clearly demonstrate the effectiveness and limitations of sequence estimation.

The inter-cell interference between uplinks and downlinks in CDMA packet communication systems, employing a shared-TDD scheme, is evaluated under cellular environments. It is found that interference between base stations rarely degrades uplink throughput, but interference between mobile stations substantially degrades downlink throughput. A transmission power control scheme is proposed to improve the downlink throughput. The proposed scheme increases the transmission power of downlink packets when they are re-transmitted, and thus, improves the signal-to-interference ratio of the downlink re-transmission packets. Computer simulation shows that this scheme increases downlink throughput without sacrificing the uplink throughput until the uplink throughput reaches a maximum value.

We describe a simplified receiver structure having several receiving antennas (i.e., an adaptive array antenna system) and using time-division-multiplexing (TDM) signal processing. Three simplified receiver structures were investigated for use in the antenna system. To confirm the feasibility of using a TDM receiver, both a TDM receiver and a conventional adaptive array receiver were constructed for testing. In our proposed system, several repetitions of the constant modulus algorithm (CMA) are used to reduce co-channel interference (CCI). The frame format used for both receivers was the same as that of the personal handy phone system in Japan. The laboratory testing was done using a fading simulator to enable measurement of the bit error rate. The results are very promising and show the feasibility of the TDM receiver.

In QoS networks, routing algorithms for QoS traffic have to provide the transmission path satisfying its QoS requirement while achieving high utilization of network resources. Therefore, server-based QoS routing algorithms would be more effective than distributed routing ones which are very common on the Internet. Furthermore, we believe that rerouting function enhances the advantage of their algorithms in which an already accepted flow with established path is replaced on some other path in order to accept newly arriving transmission request if it can not be accepted without doing so. Thus in this paper, we will propose a rerouting algorithm with the server-based QoS routing and evaluate its performance in terms of the blocking probability by computer simulation. In addition, we will investigate the impact of the amount of traffic with high-priority on the performance in some network topologies. Through some simulation results, we also discuss some issues arising in improving the effectiveness of rerouting.

This paper proposes an automatic system which can perform the entire diagnostic process from the extraction of the liver to the recognition of a tumor. In particular, the proposed technique uses shape information to identify and recognize a lesion adjacent to the border of the liver, which can otherwise be missed. Because such an area is concave like a bay, morphological operations can be used to find the bay. In addition, since the intensity of a lesion can vary greatly according to the patient and the slice taken, a decision on the threshold for extraction is not easy. Accordingly, the proposed method extracts the lesion by means of a Fuzzy c-Means clustering technique, which can determine the threshold regardless of a changing intensity. Furthermore, in order to decrease any erroneous diagnoses, the proposed system performs a 3-D consistency check based on three-dimensional information that a lesion mass cannot appear in a single slice independently. Based on experimental results, these processes produced a high recognition rate above 91%.

Network dimensioning is an important issue to provide stable and QoS-rich communication services. A reliable estimation of bandwidths of links between the end-to-end path is a first step towards the network dimensioning. Pathchar is one of such tools for the bandwidth estimation for every link between two end hosts. However, pathchar still has several problems. If unexpectedly large errors are included or if route alternation is present during the measurement, the obtained estimation is much far from the correct one. We investigate the method to eliminate those errors in estimating the bandwidth. To increase the reliability on the estimation, the confidence interval for the estimated bandwidth is important. For this purpose, two approaches, parametric and nonparametric approaches, are investigated to add the confidence intervals. Another important issue is the method for controlling the measurement period to eliminate the measurement overheads. In this paper, we propose a measurement method to adaptively control the number of measurement data sets. Through experimental results, we show that our statistical approaches can provide the robust estimation regardless of the network conditions.

We have manufactured large-scaled highly reliable MNOS EEPROMs over the last twenty years. In particular, at the present time, the smart-card microcontroller incorporating an embedded 32-kB MNOS EEPROM is rapidly expanding the markets for mobile applications. It might be said that we have established the conventional MNOS nonvolatile semiconductor memory technology. This paper describes the device design concepts of the MNOS memory, which include the optimization and control of the tunnel oxide film thickness (1.8 nm), and the scaling guideline that considers the charge distribution in the trapping nitride film. We have developed a high-performance MONOS structure and have not found any failure due to the MONOS devices in high-density EEPROM products during 10-year data retention tests after 105 erase/write cycles. The future development of this highly reliable MNOS-type memory will be focussed on the high-density cell structure and high-speed programming method. Recently, some promising ideas for utilizing an MNOS-type memory device, such as 1-Tr/bit cell for byte-erasable full-featured EEPROMs and 2-bit/Tr cell for flash EEPROMs have been proposed. We are convinced that MNOS technology will advance into the area of nonvolatile semiconductor memories because of its high reliability and high yield of products.

Selective attention mechanism, plays an important role in human visual perception, can be investigated by developing an approach to perceiving the multi-meaningful-dotted-pattern in a color blindness plate (CBP). In this Letter, a perception model driven by a simple active vision mechanism is presented for the image segmentation and understanding of a CBP. Experiments show that to understand one meaningful pattern in an image containing multi-meaningful patterns, the active visual search (i.e., pattern attention) is a very useful function.

Cache memories are small fast memories used to temporarily hold the contents of main memory that are likely to be referenced by processors so as to reduce instruction and data access time. In study of cache performance, most of previous works have employed simulation-based methods. However, that kind of researches cannot precisely explain the obtained results. Moreover, when a new processor is designed, huge simulations must be performed again with several different parameters. This research classifies cache structures for superscalar processors into four types, and then represents analytical model of instruction fetch process for each cache type considering various kinds of architectural parameters such as the frequency of branch instructions in program, cache miss rate, cache miss penalty, branch misprediction frequency, and branch misprediction penalty, and etc. To prove the correctness of the proposed models, we performed extensive simulations and compared the results with the analytical models. Simulation results showed that the proposed model can estimate the expected instruction fetch rate accurately within 10% error in most cases. This paper shows that the increase of cache misses reduces the instruction fetch rate more severely than that of branch misprediction does. The model is also able to provide exact relationship between cache miss and branch misprediction for the instruction fetch analysis. The proposed model can explain the causes of performance degradation that cannot be uncovered by the simulation method only.

Linear discriminant analysis (LDA) is a basic tool of pattern recognition, and it is used in extensive fields, e.g. face identification. However, LDA is poor at adaptability since it is a batch type algorithm. To overcome this, new algorithms of online LDA are proposed in the present paper. In face identification task, it is experimentally shown that the new algorithms are about two times faster than the previously proposed algorithm in terms of the number of required examples, while the previous algorithm attains better final performance than the new algorithms after sufficient steps of learning. The meaning of new algorithms are also discussed theoretically, and they are suggested to be corresponding to combination of PCA and Mahalanobis distance.

Coherence resonance in propagating spikes generated by noise in spatially distributed excitable media is studied with computer simulation and circuit experiment on the FitzHugh-Nagumo model. White noise is added to the one end of the media to generate spikes, which propagate to the other end. The mean and standard deviation of the interspike intervals of the spikes after propagation take minimum values at the intermediate strength of the added noise. This shows stronger coherence than obtained in the previous studies.

We have developed a new scheme to provide Diffserv-based QoS over ATM access networks. Well-known Diffserv over ATM scheme requires some extension for conventional routers with ATM interfaces. The routers must map their Diffserv classes of services into ATM QoS classes and forward IP packets into prioritized VCs based on DSCP (DiffServ Code Point). The purpose of this work is to provide Diffserv-based QoS over ATM network using conventional IP over ATM interfaces on routers. We propose DSCP snooping at ATM edge nodes, which differentiates services over a single VC between two IP domains. A prototype circuit was used to evaluate this scheme.

The differentiated services (diffserv) architecture has been proposed for implementing scalable service differentiation in the Internet. Expedited forwarding and assured forwarding have been standardized as Per-Hop Behaviors (PHB) in diffserv. Assured forwarding can be utilized to realize the service, which provides each user with a minimum guaranteed rate and a fair share of the residual bandwidth. We call it guaranteed rate (GR) service. With GR service, each packet for flow i is marked in or out based on comparison between the sending rate and the minimum guaranteed rate. When congestion occurs in networks, out packets are dropped more aggressively than in packets. Recently, several fair queuing schemes have been proposed for core stateless networks. They can achieve fairer bandwidth allocation than random early detection (RED). However, there have not been any studies that consider in/out bit usage to support GR service. This paper proposes how to extend the schemes that have been proposed for core stateless networks to allow the support of in/out bit usage. We present the performance of one of the extended schemes and compare the scheme to RED with in/out bit (RIO) in terms of fair bandwidth allocation.

In this paper, we survey the fairness issues in the congestion control mechanisms of TCP, which is one of most important service aspects in the current and future Internet. We first summarize the problems from a perspective of the fair service among connections. Several solution methods are next surveyed. Those are modifications of TCP congestion control mechanism and router support for achieving the fair service among TCP connections. We finally investigate the fair share of resources at endhosts.

In server load balancing where replicated servers are dispersed geographically and accesses from clients are distributed to replicated servers, a way of distributing the accesses from clients to an adequate server plays an important role from the viewpoint of load balancing. In the paper, we propose a new network paradigm for server load balancing using active anycast. In active anycast, an end user only sends its request to group of servers using an anycast address. When this request arrives at an active router, it selects an adequate server from the viewpoint of load balancing and changes the anycast address of a packet to the unicast address of the selected server. The decision which server is the best one from the viewpoint of server load balancing is made by an active router rather than an end user, so active anycast is a network-initiated method. Simulation results show that active anycast can accomplish efficient server load balancing, even when a small part of routers are equipped with active network technology.