Showing papers by "NEC published in 1990"

Distributed multiparty desktop conferencing system: MERMAID

Abstract: This describes a distributed multiparty desktop conferencing system (MERMAID) and presents its preliminary brief evaluation, obtained as a result of daily use. MERMAID, which is designed based on group collaboration system architecture, provides an environment for widely distributed participants, seated at their desks, to hold real-time conferences by interchanging information through video, voice, and multimedia documents. This system is implemented by using narrow-band ISDN, high-speed data network, and UNIX-based EWSs with electronic writing pads, image scanners, video cameras, microphone-installed loudspeakers, etc. The system provides participants with the means for sharing information in such multimedia forms as video images, voice, text, graphics, still images, and hand drawn figures.

Call control with transmission priority in a packet communication network of an ATM type

Abstract: In a high-speed packet multiplex communication network including a transmission line with a predetermined bandwidth and accommodating a plurality of information sources, the sources having various packet delivery rates over a range of between a peak rate and a lower rate than an average rate and demanding various transport performances, the sources are preliminarily classified into a plurality of types according to transport performances required and different transmission priorities are assigned to the different types, respectively. Bandwidths of sources of first priority and second priority are determined ones corresponding to the peak rate and the average rate, respectively. A virtual bandwidth may be calculated for the second priority source as a value between the peak and average rates. In response to connection requests from the sources, each of the connection requests is admitted when a bandwidth defined by the priority of each source is accepted in a residual bandwidth of the predetermined bandwidth, and the packets from the source of the first priority are preferentially transmitted to the transmission line, packets of the second priority source are transmitted when packets of the first priority source are absent. Thus, high bandwidth efficiency is insured while the high transport performance of the first priority source is maintained.

A self-healing network with an economical spare-channel assignment

Abstract: With the advent of networking technologies intelligent network elements, such as the digital cross-connect system (DCS), will make it possible to dynamically reconfigure a network for restoration purposes. Both restoration control of DCSs and spare-channel design issues are presented, and how they work together so that a fast and economical SONET self-healing network is obtained. In order to achieve fast restoration, a distributed control mechanism that is applicable to both line and path restoration is proposed. The proposed method allows the shared use of spare channels for various failure scenarios, including multiple failure cases, so that the efficient use of spare channels can be achieved. A linear-programming-based scheme is proposed to obtain spare-channel assignment, where a network-flow technique is used. Through a simulation study, a fast and economical self-healing network is verified. >

Enhanced spontaneous emission from GaAs quantum wells in monolithic microcavities

Abstract: Enhanced spontaneous emission has been observed with wavelength‐sized monolithic Fabry–Perot cavities containing GaAs quantum wells. With an on‐resonance cavity structure, the photoluminescence intensity increases in the cavity axis direction, and the spontaneous emission lifetime is experimentally found to decrease.

Speech coding/decoding method having an excitation signal

Abstract: A speech coding method in which spectrum parameter representing a spectrum envelope and a pitch parameter representing a pitch are obtained from an input discrete speech signal. A frame interval is divided into subintervals in accordance with the pitch parameter. A sound source signal in one of the subintervals is obtained by obtaining a multipulse with respect to a difference signal obtained by performing prediction on the basis of a past sound source signal. Correction information for correcting at least one of the amplitude and the phase of the sound source signal are obtained and output in other pitch intervals in the frame.

On the computational complexity of approximating distributions by probabilistic automata

Abstract: We introduce a rigorous performance criterion for training algorithms for probabilistic automata (PAs) and hidden Markov models (HMMs), used extensively for speech recognition, and analyze the complexity of the training problem as a computational problem. The PA training problem is the problem of approximating an arbitrary, unknown source distribution by distributions generated by a PA. We investigate the following question about this important, well-studied problem: Does there exist an efficient training algorithm such that the trained PAs provably converge to a model close to an optimum one with high confidence, after only a feasibly small set of training data? We model this problem in the framework of computational learning theory and analyze the sample as well as computational complexity. We show that the number of examples required for training PAs is moderate—except for some log factors the number of examples is linear in the number of transition probabilities to be trained and a low-degree polynomial in the example length and parameters quantifying the accuracy and confidence. Computationally, however, training PAs is quite demanding: Fixed state size PAs are trainable in time polynomial in the accuracy and confidence parameters and example length, but not in the alphabet size unless RP e NP. The latter result is shown via a strong non-approximability result for the single string maximum likelihood model probem for 2-state PAs, which is of independent interest.

System and method for transmitting entertainment information to authorized ones of plural receivers

Abstract: An audiovisual subscription system includes means for aperiodically inverting the lines of a transmitted video signal on a frame-by-frame basis and for decrypting encrypted PCM audio information which is transmitted along with the aperiodically inverted video information.

Thickness mode vibration piezoelectric transformer

Abstract: Low and high impedance thickness mode vibration portions are integrally laminated. Each vibrator portion comprises electrode layers between which a piezoelectric layer polarized in the direction of thickness is interposed. In each vibrator portion, every other electrode layer is connected to respective common external terminals. These common external terminals are of either four-terminal construction consisting two pairs, one pair for the low impedance mode vibrator portion and the other pair for the high impedance thickness mode vibrator portion, or three-terminal construction in which one of each pair are united together into a common terminal. Difference between the impedances is set on difference in the number of electrode layers and/or on difference in dielectric constant between the piezoelectric layers.

Thin film capacitor and manufacturing method thereof

Abstract: This film capacitors in accordance with the present invention include a silicon electrode (22), a first electrode layer (24) consisting of either one of titanium, titanium silicide, titanium nitride, tantalum, molybdenum, tungsten, tantalum silicide, molybdenum silicide, tungsten silicide, alloys thereof and compounds thereby, formed on the silicon electrode, a second electrode layer (25) formed on it consisting of platinum, palladium or rhodium, a dielectric layer (26) formed on it consisting of an oxide ferroelectric substance such as BaTiO₃ and a third electrode layer (27) formed on top of it. As the first electrode layer, use may also be made of rhenium oxide, osmium oxide, rhodium oxide or iridium oxide.

Self-healing network with distributed failure restoration capabilities

Abstract: In the event of a failure on a transmission line, a sender, or a terminal node of the failed line, broadcasts CONTROL packets to each of all adjacent tandem nodes, containing a hop count and node trace data. In response to a CONTROL packet, each tandem node increments the hop count of the packet, updates its node trace data, makes a record of all arriving CONTROL packets, and broadcasts copies of each arriving CONTROL packet to adjacent nodes if the hop count is smaller than a prescribed value. A chooser, which is the other terminal node of the failed line, responds to CONTROL packets from a given node for sending back to the given node as many RETURN packets as there are failed channels. Each tandem node makes a record of all arriving RETURN packets, checks the node trace data of each RETURN packet against that of the CONTROL packets in the record to detect if there is a spare channel to an adjacent node on a route to the sender with a least hop count, transmits each RETURN packet to it if a spare channel is detected, or causes retransmission of a RETURN packet from the chooser to another tandem node if no spare channel is detected. The sender responds to each RETURN packet by switching a link from a failed channel to a spare channel indicated by the packet. An END packet is sent back from the sender to each tandem node along which the RETURN packet travelled between the chooser to the sender. On reaching the chooser, the END packet is used to switch a link from a failed channel to a spare channel indicated by the received END packet.

Method of stacking semiconductor substrates for fabrication of three-dimensional integrated circuit

Abstract: A semiconductor substrate stacking method comprises the steps of preparing first and second thin film devices each in the form of a thin film having a connection electrode formed on an upper surface thereof and a connection electrode formed on an undersurface thereof, each of the thin film devices being bonded at its upper surface thereof to a support plate by adhesive. The first thin film device is stacked and bonded onto a base substrate having a device formed thereon and a connection electrode formed on the device, in such a manner that the device formed on the base substrate faces the undersurface of the first thin film device and the connection electrode formed on the device formed on the base substrate is in alignment with and in contact with the undersurface connection electrode formed on the first thin film device. The support plate and the adhesive of the first thin film device are removed so that the upper surface of the first thin film device and the upper surface connection electrode formed on the upper surface of the first thin film device are exposed. Similarly, the second thin film device is stacked and bonded onto the first thin film device stacked on the base substrate and the support plate and the adhesive of the second thin film device is removed.

Decision feedback equalizer including forward part whose signal reference point is shiftable depending on channel response

Abstract: A center tap of a transversal filter, which forms part of a forward equalizer, is shifted by at least one symbol interval. An error signal is generated using a decision output and tap outputs of the forward and backward equalizers. An impulse response estimation circuit is provided to receive the decision output of the predetermined tap signals of the forward equalizer. The forward equalizer is controlled in accordance with the channel response to effectively achieve adaptive equalization.

Signal reproducing apparatus for optical recording and reproducing equipment with compensation of crosstalk from nearby tracks and method for the same

Abstract: An apparatus and a method for reproducing a signal, free of cross-talk and intersymbol interference, from an optical recording medium. The signal reproducing apparatus and method is for use in combination with an optical source for emitting a plurality of light beams directed to a center track, which contains the information to be reproduced, and the tracks adjacent to the center track on a recording medium. The signals received from the reflections of the light beams from the adjacent tracks are processed by a frequency characteristic converter and then added to the signal from the center track to eliminate the cross-talk in the signal from the center track created by the signals from the adjacent tracks.

3-D data acquisition by Rainbow Range Finder

Abstract: The Rainbow Range Finder (RRF) has the ability to obtain range information for all image pixels with only one frame TV camera imaging during 1/30 s. The authors propose a novel practical calibration scheme which makes range measurement for all visible pixels possible. An experimental system was developed on that basis, and measurement accuracy was evaluated. A curved surface and a plain surface could be measured with sufficient accuracy. The feasibility of a real-time system with an accuracy of 1% in regard to the measurement region size was demonstrated. >

A capacitor-over-bit-line (COB) cell with a hemispherical-grain storage node for 64 Mb DRAMs

Abstract: A novel capacitor-over-bit-line (COB) cell with a hemispherical-grain (HSG) poly-Si storage node has been developed. This memory cell provides large storage capacitance by increasing the effective surface area of a simple storage node and is manufacturable by optical delineation. The feasibility of the COB cell for 64-Mb DRAMs has been verified by a 64-kb test memory with 1.8- mu m/sup 2/ cells using a 0.4- mu m design rule, storage capacitance of 30 fF, 7-nm-SiO/sub 2/-equivalent dielectric film, and a storage node height of 0.5 mu m. >

Thin-film capacitors and process for manufacturing the same

Abstract: A thin-film capacitor comprises a substrate, a first electrode, of polycrystalline silicon, a second electrode, a dielectric, and a third electrode such as aluminum in the structure stacked in sequence from bottom to top. The second electrode directly on which the dielectric layer is formed is made from a member selected from the group consisting of ruthenium, ruthenium oxide, ruthenium silicide, rhenium, rhenium oxide, rhemium silicide, osmium, osmium oxide, osmium silicide, rhodium, rhodium oxide, rhodium silicide, iridium, iridium oxide and iridium silicide.

Process of forming bump on electrode of semiconductor chip and apparatus used therefor

Abstract: A process of forming bumps on respective electrodes (118a to 118v) of a semiconductor chip (SC21) comprises the steps of a) preparing a bonding apparatus equipped with a bonding tool (114) three-dimensionally movable, b) forming a small ball at the leading end of a bonding wire passing through the bonding tool, c) causing the bonding tool to press the small ball against the upper surface of one of the electrodes for bonding thereto, d) moving the bonding tool in a direction leaving from the upper surface of the electrode by a distance, e) moving the bonding tool in a horizontal direction (X21 or Y21) substantially parallel to the upper surface of the electrode so that the bonding wire is cut from the small ball with the leading end of the bonding tool, and f) repeating the steps b) to e) for producing bumps on the respective upper surfaces of the other electrodes, in which the horizontal direction is different depending upon the location of the electrode.

Neutron-diffraction study of Tl 2 Ba 2 CuO 6+δ with various T c 's from 0 to 73 K

Abstract: We refined the crystal structures of four pseudotetragonal samples of ${\mathrm{Tl}}_{2}$${\mathrm{Ba}}_{2}$${\mathrm{CuO}}_{6+\mathrm{\ensuremath{\delta}}}$ with ${\mathit{T}}_{\mathit{c}}$'s of 0 (metallic), 48, 58, and 73 K by Rietveld analysis of time-of-flight neutron-powder-diffraction data. The presence of excess oxygen atoms located at an interstitial site between double TlO layers was confirmed. The change in oxygen content and the corresponding one in hole carrier concentration in this compound are caused by incorporation and release of the excess oxygen, whose amount is about 0.1 per formula unit for the metallic nonsuperconducting sample. The highest ${\mathit{T}}_{\mathit{c}}$ value of about 85 K in ${\mathrm{Tl}}_{2}$${\mathrm{Ba}}_{2}$${\mathrm{CuO}}_{6+\mathrm{\ensuremath{\delta}}}$ is achieved in the absence of excess oxygen. As the oxygen content is decreased, the c value increases, and apical ${\mathrm{O}}^{2\mathrm{\ensuremath{-}}}$ ions go away from the ${\mathrm{CuO}}_{2}$ layer, while ${\mathrm{Ba}}^{2+}$ ions approach it. It was also suggested that about 5% of a Tl site is substituted by Cu. A significant portion of hole carriers in this compound is attributed to substitution of Cu for Tl and incorporation of excess oxygen.

Speaker-independent word recognition using a neural prediction model

Abstract: A speech recognition model called the neural prediction model (NPM) is proposed. The model uses a sequence of multilayer perceptrons (MLPs) as a separate nonlinear predictor for each class. It is designed to represent temporal structures of speech patterns as recognition cues. In particular, temporal correlation in successive feature vectors of a speech pattern is represented in the mappings formed as MLP input-output relations. Temporal distortion of speech is efficiently normalized by a dynamic-programming technique. Recognition and training algorithms are presented based on the combination of dynamic-programming and back-propagation techniques. Evaluation experiments were conducted using ten-digit vocabulary samples uttered by 107 speakers. A 99.8% recognition accuracy was obtained. This suggests that the model is effective for speaker-independent speech recognition. >

A learning criterion for stochastic rules

Abstract: This paper proposes a learning criterion for stochastic rules. This criterion is developed by extending Valiant's PAC (Probably Approximately Correct) learning model, which is a learning criterion for deterministic rules. Stochastic rules here refer to those which probabilistically asign a number of classes, lYr, to each attribute vector X. The proposed criterion is based on the idea that learning stochastic rules may be regarded as probably approximately correct identification of conditional probability distributions over classes for given input attribute vectors. An algorithm (an MDL algorithm) based on the MDL (Minimum Description Length) principle is used for learning stochastic rules. Specifically, for stochastic rules with finite partitioning (each of which is specified by a finite number of disjoint cells of the domain and a probability parameter vector associated with them), this paper derives target-dependent upper bounds and worst-case upper bounds on the sample size required by the MDL algorithm to learn stochastic rules with given accuracy and confidence. Based on these sample size bounds, this paper proves polynomial-sample-size learnability of stochastic decision lists (which are newly proposed in this paper as a stochastic analogue of Rivest's decision lists) with at most k literals (k is fixed) in each decision, and polynomial-sample-size learnability of stochastic decision trees (a stochastic analogue of decision trees) with at most k depth. Sufficient conditions for polynomial-sample-size learnability and polynomial-time learnability of any classes of stochastic rules with finite partitioning are also derived.

Emitter size effect on current gain in fully self-aligned AlGaAs/GaAs HBT's with AlGaAs surface passivation layer

Abstract: The emitter size effect for fully self-aligned AlGaAs-GaAs heterojunction bipolar transistors (HBTs) with depleted AlGaAs passivation layers, in which the partially thinned AlGaAs emitter is self-aligned by using the dual sidewall process, is investigated. It is demonstrated that drastic improvement in the emitter size effect can be achieved with an AlGaAs passivation layer as small as 0.2 mu m in width, due to the surface recombination current reduction by a factor of 1/40 in the extrinsic base region. It has also been found that the base current is dominated by excess leakage current in the proton-implanted isolation region. >

Dual port memory buffers and a time slot scheduler for an atm space division switching system

Abstract: An asynchronous transfer mode switching network system relays packets stored in packet buffer units to output ports designated by the packets, and a time slot scheduling unit assigns time slots to the packets stored in the packet buffer units upon arrival at the packet buffer units for preventing the packets from collision in a space division switching unit, wherein each of the packet buffer units sequentially writes new packets into respective memory locations but randomly reads out the new packets in the time slots assigned by the time slot scheduling unit so that the throughput of the space division switching unit is improved.

Wavelength division switching system with reduced optical components using optical switches

Abstract: In a wavelength division switching system having m 1 inlet highways and m 2 outlet highways, m 1 splitters are coupled respectively to the inlet highways to split each multiplex signal into m 2 ×n output signals for coupling to respective inputs of m 2 ×n optical switches. Each optical switch is responsive to a switching control signal for coupling one of its inputs to one of m 2 ×n tunable wavelength filters capable of passing one of n different wavelengths, in response to a wavelength selecting signal, to one of m 2 ×n fixed wavelength converters which are organized into m 2 groups. These wavelength converters provide conversion of the input signals so that the converted wavelength is unique in each of the m 2 groups. An array of m 2 combiners are respectively associated with the groups of the fixed wavelength converters and with the outlet highways. Each combiner has n inputs coupled respectively to the outputs of the fixed wavelength converters of the associated group to multiplex signals appearing at the n inputs thereof for coupling to the associated outlet highway.

Watch type paging receiver

Abstract: A watch type paging receiver which may be put on person's wrist by a belt includes a body on which a display device such as a liquid crystal display device is provided. An inverted F-shaped main antenna is implemented by a part of the display device. A loop antenna is provided in addition to the main antenna and implemented by the belt. The main and loop antennas are selectively rendered active depending upon the level of a received signal. Even when the length of the belt is changed to suit person's wrist, accurate matching is established on the basis of the resulting length of the belt. A primary battery in the form of a solar battery and a secondary battery are associated with the receiver body and the belt, respectively. A vibration source which is adapted for tactual alerting forms a part of the belt.

Direct conversion receiver with dithering local carrier frequency for detecting transmitted carrier frequency

Abstract: An FSK receiver includes first and second channel paths to which the received FSK signals are applied. Each channel path includes a mixer circuit, a low-pass filter connected to the mixer circuit, and a limiting amplifier connected to the low-pass filter. A voltage-controlled oscillator generates a local carrier and a phase shifter applies its output to the mixer circuits with a 90° phase difference therebetween. A frequency detector is coupled to one of the limiting amplifiers for generating a signal representative of the deviation frequency minus the offset frequency of the received FSK signal. An average circuit detects an average value of the signal from the frequency detector. A comparator circuit detects whether the frequency representative signal lies in or outside of the neighborhood of the average value. A ramp voltage of constant amplitude is periodically generated if the signal is detected as lying outside of the neighborhood and maintained at an instantaneous value if the signal is detected as lying in the neighborhood. The voltage-controlled oscillator is responsive to the ramp generator to control the frequency of the local carrier in search of the transmitted carrier frequency.

Radio communications system adaptively assigning channels using power levels of idle channels

Abstract: In a mobile radio communications system comprising a control station and base stations, each base station is capable of sensing power levels of signals received on idle channels. In response to a polling signal periodically sent from the control station, each base station returns a signal indicating the detected power levels, the identifications of the idle channels, the identification of a busy channel or no-busy channel indication, and the identification of each of the base stations. The control station derives an interference table from the contents of the signals received from all base stations. The table defines interference coefficients between all pairs of base stations in matrix form according to the identifications of the idle and busy channels and the identifications of the base stations. From the interference coefficients a plurality of interference estimations are derived for all possible combinations of channels and base stations. A minimum value of the interference estimations is detected and the channels associated with the minimum value are assigned to the base stations. The assigned channel identifiers are communicated from the control station to the base stations.

CT simulator: a new 3-D planning and simulating system for radiotherapy: Part 1. Description of system.

Abstract: A real time CT-linked 3-D treatment planning system, called a CT simulator, has been developed. The basic system consists of a CT scanner, a multi-image display component, a treatment planning device with real time visual optimization, and a laser beam projecting component. All the components are connected on line. The system can be conveniently used for 3-D planning and simulation for radiation therapy within a reasonably short period of time.

Adaptive transform coding with an adaptive block size (ATC-ABS)

Abstract: A coding technique is presented for high-quality audio signals based on adaptive transform coding (ATC). Adaptive block size selection by the proposed algorithm ensures an appropriate block size resulting in improved SNR (signal-to-noise ratio) for a wide variety of source signals. A feedback approach, based on SNR, and a feedforward approach, based on interblock differences in input time-domain samples, to adaptive block size assignment are proposed and evaluated. Computer simulation results show that average segmental SNR by the feedback approach is improved by as much as 4.8 dB over the conventional fixed-block-size ATC. The feedforward approach is realized with much-simplified hardware; nevertheless, its SNR degradation from that by the feedback approach is 1.6 dB, even in the worst case. Both approaches are successful in pre-echo suppression to a satisfactory level. Time-domain aliasing cancellation has the potential to increase the superiority of the new algorithm. >