Understanding traffic characteristics in mobile communications is invaluable for planning, designing, and operating cellular networks, and various mobility models have therefore come to be developed to predict traffic characteristics. In this paper, cell-dwell-time distribution and transition probability in a virtual cellular system are first estimated from the results of measuring taxi motion using the Global Positioning System (GPS) for large-city and small-city ranges of motion. Then, on the basis of simulations using these estimations, traffic characteristics like handoff rate and channel blocking probability in a cellular system are evaluated. It was found that a difference between large and small cities could be observed in speed distribution and direction-of-travel probability, but only a slight difference in cell-dwell-time distribution.

We propose a new conditional linearization based on left-right separated conditional term rewriting systems, in which the left-hand side and the right-hand side of a rewrite rule have separate variables. By developing a concept of weight decreasing joinability we first present a sufficient condition for the Church-Rosser property of left-right separated conditional term rewriting systems. Applying this result to conditional linearization, we next show sufficient conditions for the unique normal form property and the Church-Rosser property of non-duplicating (unconditional) term rewriting systems even if they are non-left-linear or overlapping.

The IETF Mobile IPv6 enables any IPv6 node to both cache the Care-of Address associated with a mobile node's home address, and to directly send packets addressed to a mobile node at the Care-of Address using the IPv6 routing header. Support for optimizing the route is built in as a fundamental part of the protocol. Several hierarchical schemes have been presented recently on top of the Mobile IPv6. These schemes separate micro-mobility from macro-mobility and exploit a mobile node's locality. They can reduce the number of signaling messages sent to a home network and improve hand-off performance. However, existing hierarchical schemes do not achieve route optimization. When external correspondent nodes send packets to a mobile node, these packets are intercepted by an intermediate mobility agent encapsulated and routed to the mobile node. In this paper, we propose a new hierarchical scheme that enables any correspondent node to cache two Care-of Addresses; the mobile node's temporary address and the intermediate mobility agent's address. Also, we introduce two lifetimes managing the two Care-of Addresses. Until the lifetime associated with the mobile node's temporary address expires, a correspondent node can send packets directly to the mobile node. If the lifetime expires but the lifetime associated with the intermediate mobility agent's address has not expired, the correspondent node sends packets to the intermediate mobility agent. This proposal can reduce delay in packet delivery and optimize routing. Furthermore, based on the mobility of a mobile node, we introduce more reduced frequency of binding update and longer period of the lifetimes than the existing hierarchical schemes. Therefore, our proposal can reduce the binding update bandwidth as well as the packet delivery bandwidth lower than those of the IETF IPv6 and the existing hierarchical schemes.

A simple and an efficient algorithm for polygon morphing is proposed in this paper. We adopt the parametric curve representation based on Fourier parameter estimation to transfer the traditional morphing process in spatial domain to a process in the parametric space instead. The principles are to express the polygon as the union of matching segments that are described by the estimated Fourier parameters. We have also designed a data resampling method that effectively controls the shape morphing according to the corresponding curvature values. Intermediate objects in-between the source and target polygons are then constructed based on the interpolation of Fourier parameters of the two polygons. Fourier parameters of the resampled polygons can be obtained efficiently by using the fast Fourier transform (FFT) algorithm. The experimental results show that the appearances of the morphed objects are superior to the ones obtained by the methods available.

In quantum communication theory, a realization of the optimum quantum receiver that minimizes the error probability is one of fundamental problems. A quantum receiver is described by detection operators. Therefore, it is very important to derive the optimum detection operators for a realization of the optimum quantum receiver. In general, it is difficult to derive the optimum detection operators, except for some simple cases. In addition, even if we could derive the optimum detection operators, it is not trivial what device corresponds to the operators. In this paper, we show a realization method of a quantum receiver which is described by a projection-valued measure (PVM) and apply the method to 3-ary phase-shift-keyed (3PSK) coherent-state signals.

We have proposed the packet SS-CDMA scheme for downlink of SS-CDMA flexible wireless cellular network. Transmission packet is framed with synchronization block with 11 chip Barker code and information block with orthogonal spreading code. The chip synchronization is carried out using short code surface acoustic wave (SAW) matched filter. The code de-spreading is carried out using in-line de-spreader. Multi-channel downlink of 63 channels can be designed using orthogonal m-sequence. Simulation results show more than 15 channels without degradation from theoretical value can be used under multi-path environment. The packet SS-CDMA modem has been implemented using a 2.4 GHz front-end SAW matched filter. Degradation of Eb/N0 of less than 0.5 dB is experimentally achieved with four-channel multiplex.

The third generation cellular system will provide high bandwidth data service and multimedia applications are proposed on the high-speed data link. On the other hand, cellular handsets are getting equipped with a short-range radio communication device that is mainly targeted to connect cellular handsets to computers. In this paper, we propose a percolating data delivery mechanism for a cellular-ad hoc integrated network utilizing multicast/broadcast communication, which is endorsed by data synchronization, and single-hop ad hoc networks for information shower/exchange services. The mechanism lessens data traffic in both the cellular and the ad hoc network for data delivery and tolerates unexpected disconnection in the ad hoc network. The mechanism also compensates data delivery in areas out of the cellular service. Our implementation of the data delivery mechanism utilizes Bluetooth for an ad hoc network configuration, and delivers data in the cellular network to the Bluetooth ad hoc network via the data synchronization mechanism. We evaluate communication traffic and delivery time on the prototype system and discuss about the efficiency of the mechanism.

In any ill-conditioned information-transfer system, as in long-distance communication, we often must construct feedback confirmation channels, in order to confirm that informations received at destinations are correct. Unfortunately, for such systems, undesirable uncertain fluctuations may be induced not only into forward communication channels but also into feedback confirmation channels, and it is such difficult that transmitters always confirm correct communications. In this paper, two fuzzy-set-valued mappings are introduced into both the forward communication channel and the feedback confirmation channel, separately, and overall system-behaviors are discussed from the standpoint of functional analysis, by means of fixed point theorem for a system of generalized equations on fuzzy-set-valued mappings. As a result, a good mathematical condition is successfully obtained, for such information-transfer systems, and fine-textured estimations of solutions are obtained, at arbitrary levels of values of membership functions.

Off-line state identification methods for a sequential machine using a homing sequence or an adaptive homing sequence (AHS) are well-known in the automata theory. There are, however, so far few studies on the subject of the on-line state estimator such as a state observer (SO) which is used in the linear system theory. In this paper, we shall construct such an SO for a Moore machine based on the state identification process by means of AHSs, and discuss the convergence property of the SO.

This paper proposes an adaptive zone selection (AZS) scheme with adaptive modulation for wireless packet transmission systems to achieve high throughput and low delay performances even under non-uniform traffic conditions. In the proposed system, based on the measurement of the propagation path characteristics to each access point (AP) as well as broadcasted blocking probability information from each AP, a terminal autonomously selects an AP and modulation parameters that give the minimum transmission failure probability determined by both the call blocking rate due to lack of radio resource and packet error rate due to severe channel conditions. Computer simulation confirms that the proposed scheme greatly improves throughput and delay performances especially under non-uniform traffic conditions.

This paper describes experimental results on whole word HMM-based speech recognition of connected digits in Japanese with special focus on the training data size and the "sheep and goats" problem. The training data comprises 757000 digits uttered by 2000 speakers, while the testing data comprises 399000 digits uttered by 1700 speakers. The best word error rate for unknown length strings was 1.64% obtained using context dependent HMMs. The word error rate was measured for various subsets of the training data reduced both in the number of speakers (s) and the number of utterances per speakers (u). As a result, an empirical formula of s[{min(0.62s0.75, u)}0.74 + {max(0, u-0.62s0.75)}0.27] = D(Ew) was developed, where Ew and D(Ew) designate word error rate and effective data size, respectively. Analyses were conducted on several aspects of the low performance speakers accounting for the major part of recognition errors. Attempts were also made to improve their recognition performance. It was found that 33% of the low performance speakers are improved to the normal level by speaker clustering centered around each low performance speaker.

An electrophysiological experiment showed that spike timing was precise to less than one millisecond. This result indicates the possibility in the precise time codings. For a high accurate time coding, reconsideration of a neural mechanism which decides firing time is required. From such viewpoint, we quantitatively examined change in firing time with interference between two synaptic inputs through Hodgkin-Huxley (HH) and integrate-and-fire (IF) model neurons. The precise firing times in the HH model neuron were extremely different from those in the IF model neuron. In this paper, the relations of input intensity to firing time are investigated in the other more two pulse generation models: Morris-Lecar (ML) and FitzHugh-Nagumo (FN) model. The result of the ML model in a certain parameter set (type-I) exhibited monotone decreasing like that of the IF model while the result of the ML model in the otter parameter set (type-II) exhibited non-monotone decreasing like that of the HH model. The result of the FN model exhibited non-monotone decreasing like the HH model despite its qualitativeness. Next the firing patterns in the four model neurons on a model of V1 (primary visual area) and LGN (lateral geniculate nucleus) with circular and mutual excitatory connections are investigated to show how dependent on model neurons the firing patterns are. The spikes in the HH, the ML type-II, and the FN model neurons elicited synchronous oscillations while the spikes in the IF and the ML type-I model neurons did not; the firing patterns dramatically changed with the dependence on the model neurons.

We have demonstrated a dynamically stable polarization operation of GaInAs/GaAs vertical-cavity surface-emitting laser (VCSEL) array grown on a GaAs (311)B substrate. A fabricated 3 3 VCSEL array consists of devices with an oxide aperture of 3.4 µm 3.4 µm. The threshold current was 0.61 0.05 mA and the threshold voltage was 1.79 0.03 V. All of the devices exhibited single-transverse mode operation with an injection current up to three times the threshold. The side-mode suppression ratio (SMSR) is larger than 30 dB. The array also exhibited stable-polarization operation with an orthogonal polarization suppression ratio (OPSR) of over 25 dB. We measured the time-resolved OPSR under square-wave direct modulation. It was found that the orthogonal non-lasing mode was suppressed even at the first peak of relaxation oscillation with OPSR > 17 dB. In an experiment of 5 Gb/s non-return-to-zero (NRZ) pseudo-random bit sequence (PRBS) modulation, the OPSR was maintained being greater than 27 dB. These observed stable and single polarization characteristics were originated from the anisotropic optical gain of strained GaInAs/GaAs quantum wells formed on (311)B substrate. We also carried out an intensity noise measurement under single-transverse and stable polarization operation for the first time. The relative intensity noise (RIN) of -140 dB/Hz was obtained. In addition, we achieved 2.5 Gb/s data transmission through a 100 m multi-mode fiber. No error floor was observed at -20 dBm of received power level. The minimum received power is determined by the thermal noise.

This paper proposes a media synchronization scheme with causality control for distributed multimedia applications in which the temporal and causal relationships exist among media streams such as computer data, voice, and video. In the scheme, the Δ-causality control is performed for causality, and the Virtual-Time Rendering (VTR) algorithm, which the authors previously proposed, is used for media synchronization. The paper deals with a networked shooting game as an example of such applications and demonstrates the effectiveness of the scheme by experiment.

Research on the propagation characteristics in the microwave band aiming at broadband mobile services is attracting much attention. Typical examples are the Unlicensed-NII (U-NII) band in the U.S. and HIPER-LAN band in Europe, i.e. 5.2 GHz. An efficient approach to revealing the propagation characteristics in the 5-GHz band is to utilize the existing propagation data accumulated by many researchers on the 2-GHz band. This paper presents the differences in path loss between the 5.2-GHz and 2.2-GHz bands in a residential area by using a 5.2-GHz/2.2-GHz dual band antenna. This antenna enables a direct comparison of 5.2 GHz and 2.2 GHz in terms of the propagation characteristics. We found that the difference in path loss between the 2.2-GHz and 5.2-GHz bands depends on only the base/mobile station antenna height. Based on this, we formulate the relationship between the heights of the base/mobile station antennas and the difference in path loss between the 2.2-GHz and 5.2-GHz bands.

This paper presents a new and robust technique for a coarse frequency offset estimation in OFDM system. As an evaluation of the proposed algorithm, we apply it to Eureka 147 DAB system. The proposed coarse frequency offset estimation algorithm is based on the differential detection technique between adjacent subcarriers to eliminate the phase shift effects of symbol timing offset and fractional frequency offset. Coarse frequency offset is determined from the correlation outputs between a received intercarrier differential phase reference symbol and several locally generated but frequency shifted intercarrier differential phase reference symbols. The performance of our estimation algorithm is evaluated by means of computer simulation and compared with that of previous proposed algorithms for DAB transmission modes I, II, III and IV. Simulation results show that the proposed algorithm generates extremely accurate estimates with a low complexity irrespective of the symbol timing offset.

A simplified analysis is presented for the reverse link maximum capacity trade-offs between each layer, spectrum efficiency and its multi-rate features of TDMA/W-CDMA and N-CDMA/W-CDMA overlaid systems with the perfect power control based on the measurement of signal-to-interference ratio (CIR). In order to suppress the multi-cross interference, the other important techniques used in the analysis are the ideal notch filtering and the signal level clipper for W-CDMA system transmitters and receivers. We firstly propose the concepts of the notch filtering depth and signal level clipping depth in the paper. The numerical results can be adopted as a guideline in designing the overlaid systems in the various cases as well as a means to investigate the flexibility of sharing of the spread spectrum and their feasibility in the future mobile communication system.

A feasibility study has been made of the detection possibility of radio wave noises, i.e., Martian atmospherics, emitted from discharges in the Martian atmosphere during large dust storms. The spacecraft NOZOMI, which was launched in 1998, is to be placed on an elliptic orbit around Mars with perigee of 150-200 km. An onboard-equipment LFA (Low Frequency Plasma wave Analyzer) has capability to measure the low frequency plasma waves in the frequency range from 10 Hz to 32 kHz. In order to know if the LFA can detect the atmospheric radio noises, the propagation characteristics of electromagnetic waves through the Martian ionosphere are studied theoretically by using a full-wave method. The ionosphere is modeled as a magneto-ionic medium based on the recent observations of magnetic anomaly by Mars Global Surveyor spacecraft, and the atmospheric constituent and electron density by Viking observations. Our calculation shows that the waves at frequencies less than a hundred hertz can propagate with low attenuation and reach to altitudes above 200 km in the whistler-mode in the regions of magnetic anomalies in the dayside ionosphere. It is shown that the radio noises emitted from electric discharge in an intense dust storm, with the intensity over -30 dBV/m/Hz at the ionospheric entry point, can be sensed by the LFA. The observational identification of Martian atmospherics will contribute to the physical study of charge/discharge process in the Martian atmosphere.

Conventional spatial transform based motion estimation algorithms are not practical because of their heavy computational loads. In this paper, we proposed motion estimation method with variable grid size, which is more efficient than conventional spatial transform based methods and gives better PSNR performance than conventional BMA.

VCSEL output light polarization was controlled by fabricating devices on (311) substrate. Stability was improved by introducing compressive strain to the quantum wells in the active layer. In experiments, the power penalty due to polarization-dependent loss in the transmission line was negligible for both VCSELs with unstrained and strained quantum well active layers on (311)B substrate. The sensitivity at 2.5 Gbps was improved in a device with a strained active layer because the intensity noise due to the polarization instability was reduced. These characteristics are discussed and compared to calculated results.