Abstract: Cost/performance of VLIW architecture is improved by reducing the number of slots in the instruction issue register.

Abstract: A measuring unit which is not easily accessible often has a power source which supplies power to the elements for the measuring operation, more particularly, the conversion of an analog measured signal into digital measured data, and for storage thereof, as required. These measured data of a measuring unit are transmitted to a base station without any contact being made the moment this base station is brought into the neighborhood of the measuring unit and transmits a signal. For a minimum load on the power source power from the power source is not supplied to the transmitter for the transmission of measured data from the measuring unit to the base station. Instead, a signal transmitted by the base station provides transmission power. In the measuring unit this signal provides a voltage for feeding the transmitter.

Abstract: A VLIW processor uses a compressed instruction format that allows greater efficiency in use of cache and memory. Instructions are byte aligned and variable length. Branch targets are uncompressed. Format bits specify how many issue slots are used in a following instruction. NOPS are not stored in memory. Individual operations are compressed according to features such as whether they are resultless, guarded, short, zeroary, unary, or binary. Instructions are stored in compressed form in memory and in cache. Instructions are decompressed on the fly after being read out from cache.

Abstract: A travel direction speaking system that is based on a portable laptop computer. The computer determines a route between an origin and a destination using an electronic map, and prepares driving instructions based on the route. Each driving instruction includes the spatial position (longitude and latitude) of a change in direction at which the driving instruction applies. The system includes a Global Positioning System (GPS) unit that provides the GPS determined position of the laptop computer. The computer compares the GPS position to the spatial position and outputs the corresponding driving instruction when the two positions are inferred to be within a specified range of each other. The range can be determined based distance or on a time period required to travel from the current position to a position at which the instructions should be spoken.

Abstract: A 64 QAM signal constellation reduces phase noise as compared with a rectangular constellation, but requires a fairly simple decoder. The constellation has decision regions which are approximately rectangular; allows for quadrant decoding; and has constellation points representable by a small number of bits.

Abstract: In a method for measuring a reference position of a tool relative to a workpiece, the tool is fastened to a first holder of a machine tool. The workpiece is to be manufactured using the tool and is fastened to a second holder of the machine tool. The tool provides a distinguishing mark on the workpiece in a predetermined reference position of the first holder relative to the second holder. Subsequently, the distinguishing mark is detected by a sensor. Therefore, the reference position of the tool relative to the workpiece is determined from a position of the first holder relative to the second holder. The tool and the sensor are fastened in fixed, permanent positions relative to one another and relative to the first holder. The accuracy of measurement of the position of the tool depends exclusively on the accuracy of positioning of the first holder relative to the second holder.

Abstract: For TXRF (Total Reflection XRF) or GEXRF (Grazing Emission XRF) a very flat sample surface and a suitably defined region in which the sample material is present are often required. Both requirements can be satisfied by means of a (preferably synthetic) sample carrier on which a first region which has a small liquid contact angle is surrounded by a second region having a large contact angle. When applied in the first region, a droplet of solvent containing the sample material will completely wet said region, whereas the second region will not be wetted. As the solvent evaporates, the boundary of the droplet does not retreat but the first region remains fully wetted, resulting in uniform coverage of this region by sample material and also in a suitably defined sample spot.

Abstract: A processing device performs operations in response to program instructions. In particular, values are written to a data memory of the system, which alters a defined visible state of the system. In the event of an exception (e.g. a pagefault or TLB miss in a virtual memory system), control returns to a recent checkpoint instruction after handling the exception, and instructions are re-executed. A record/replay circuit is provided in the form of an event memory, which remembers only those values read from the data memory since the last checkpoint. During the re-execution, the recorded values are reproduced instead of performing actual memory reads, and all memory accesses are suppressed. When the re-execution reaches the point where execution was originally interrupted, recording begins again to prepare for any further exception which may arise before the next checkpoint.

Abstract: An RTA or similar rule-based agent (102) is enabled to construct a virtual simulation of its environment by the provision of an internal prediction module (100). The prediction module comprises a rule population run in tandem with the population (106) defining the agent behaviors. The prediction module rules represent predictions of agent behavior state changes and their accuracy is periodically checked (110), with more accurate rules being assigned a higher fitness rating. A genetic algorithm (108) defines mutation operators by which further rules are evolved from those having the highest fitness. When a predetermined level of fitness (prediction accuracy) is achieved, the prediction module rules provide (116) a virtual simulation of the environment to the agent behaviors by direct control in place of external (real world) stimuli.

Abstract: In an apparatus for handling objects such as documents and tools a menu of options (30) is displayed, which options (31a . . . 31h) are placed in a curved band (31) around the position of the cursor at the moment of activating the menu. The shape of the curved band is optimized for the movement of a hand-held pointer device, such as a stylus, and allows easy movement of the pointer device along the options. The band (31) may follow part of the circumference of a circle of which circle inner portions (32, 33) may be used for the display of several sets of sub-options.

Abstract: A compressed instruction format for a VLIW processor allows greater efficiency in use of cache and memory. Instructions are byte aligned and variable length. Branch targets are uncompressed. Format bits specify how many issue slots are used in a following instruction. NOPS are not stored in memory. Individual operations are compressed according to features such as whether they are resultless, guarded, short, zeroary, unary, or binary. Instructions are stored in compressed form in memory and in cache. Instructions are decompressed on the fly after being read out from cache.

Abstract: A cold finger for an energy dispersive x-ray analyzer is surrounded by an insulating material which includes two layers of aluminum coated polyester and a layer of plastic netting sandwiched between the two layers of aluminum coated polyester. The material is flexible and hugs the cold finger and reduces thermal loss.

Abstract: Registers are divided into a global pool and a local pool. Code to be used in the processor must allocate registers from the global pool for values that live across decision trees and from the local pool for local values. The processor only accepts interrupts and exceptions during successful inter decision tree jumps. The code signals such jumps with special interruptable jump operations. To put a ceiling on clock cycles between interruptable jumps, each loop should have at least one successful interruptable jump and/or backwards jumps should be interruptable. The interrupt and exception handlers use only registers from the local pool. During a context switch, only contents of the global pool are saved in the task record.

Abstract: A radiation detector having an evacuated envelope, a radiation detector on a cold finger support in the evacuated space, a closed cycle gas cooling system to cool the cold finger to provide cryogenic operation of the radiation detector, and a getter in the evacuated space to maintain an evacuated condition. The evacuated envelope includes a radiation window. The radiation detector is preferably an X-ray detector employed in an energy dispersive spectrometry system. The evacuated space is preferably held at a pressure of less than about 1 mTorr to achieve molecular flow of remaining gas molecules, minimizing parasitic heat input. The closed cycle gas cooling system employs compressed refrigerant, which is precooled in a counterflow heat exchanger and allowed to expand in proximity to the cold finger, thus absorbing heat and maintaining cryogenic temperatures. A getter material, preferably activated carbon, is provided to absorb gasses and maintain the low pressure during operation.

Abstract: The invention relates to an examination method whereby a respective spectrum with a number of spectral values is measured for a number of locations. Collective evaluation of the old spectra is enabled by the following steps:a) formation of a data matrix from the spectral vectors formed by the series of spectral values of a respective spectrum, the spectral vectors being arranged in the columns (or in the rows) of the data matrix in a location-dependent manner,b) singular value decomposition of the data matrix in order to obtain three matrices whose product corresponds to the data matrix, the first (third) matrix consisting of spectrally dependent vectors, whereas the second matrix is a diagonal matrix and the third (first) matrix consists of location-dependent vectors,c) evaluation of at least one of the three matrices.

Abstract: A method and apparatus for path planning are presented. Path planning involves propagating cost waves in a configuration space representation of a task space. A space variant metric and budding are used for cost wave propagation. The disclosed method and apparatus are readily adaptable to robots with n degrees of freedom.

Abstract: A telecommunications system includes a network of coupled telecommunication exchanges (PABX1, PABX2, PABX3). At least one of the exchanges is coupled to a processor (COMP) via a so-called Computer Telephony Interface (CTI). The exchanges monitor state information for extensions (TLPH1, TLPH2, TLPH3) that are connected to the exchanges. A router routes the state information to the exchange which is coupled to the processor. That exchange transfers the state information to the processor.

Abstract: A barrier is used to synchronize parallel processors. The barrier is "fuzzy", i.e. it includes several instructions in each instruction stream. None of the processors performing related tasks can execute an instruction after its respective fuzzy barrier until the others have finished the instruction immediately preceding their respective fuzzy barriers. Processors therefore spend less time waiting for each other. A state machine is used to keep track of synchronization states during the synchronization process.

Abstract: A neural digital processor (10) that includes circuitry (14) for applying a function ANLF to neural potentials. ANLF approximates a non-linear activation function NLF. The circuitry includes another neural processor which operates with another non-linear activation function CNLF. CNLF is a simple function, for example a ramp. The circuitry (14) may comprise elements (36.sub.1, 36.sub.2, 64) in common with apparatus (75) for calculating a derivative of the approximation function ANLF. The precision of approximation of the non-linear activation function NLF can be predetermined.

Abstract: A barrier is used to synchronize parallel processors. The barrier is "fuzzy", i.e. it includes several instructions in each instruction stream. None of the processors performing related tasks can execute an instruction after its respective fuzzy barrier until the others have finished the instruction immediately preceding their respective fuzzy barriers. Processors therefore spend less time waiting for each other. A state machine is used to keep track of synchronization states during the synchronization process.