Abstract:
A method for processing information about a polygonal pattern in the context of pattern recognition in a computer environment, the method comprising the steps of: (a) taking the corners of the pattern in turn in order from a first of the corners through the last of the corners, forming a first ordered sequence M of values representing the respective scalar distances between each corner of the polygonal pattern and each other corner, the sequence including (i) the distances between said first of said corners and each other corner in turn and (ii) taking each other corner in turn, the distances between that other corner and each other corner starting with the said first corner, and (iii) including at appropriate positions in the sequence zero values to correspond to the distance between each corner and itself; (b) re-ordering the values of said first sequence M to form a second ordered sequence P1 comprising a set of values p1 which includes for each corner starting with said first corner, values in a series starting with and including the said zero value for that corner and including the other values for that corner in one direction through, the order in which they appear in said first sequence; and (c) converting said values in the second ordered sequence P1 to respective equal-length character strings representative of the values.
Abstract:
An integrated circuit having a memory which is reconfigurable as a main memory or as a cache. The integrated circuit may be a microprocessor chip with a memory that is reconfigurable to operate as an on-chip main memory or as an on-chip cache. Alternatively, the integrated circuit may be a stand-alone memory chip that is reconfigurable to operate as a main memory or as a cache.
Abstract:
A distributed shared memory multi-processor system includes a System Control Unit (SCU) which is made up of a system control unit address section (SCUA) and system control unit data sections (SCUDs). The SCU is scalable by dividing the control and data flow functions of the SCU, and then parallelizing the data path. This allows the number of processors in the system to be increased or higher performance processors to be added by increasing the number of SCUDs and reprogramming crossbar switches incorporated in the SCUA and SCUDs. This results in the overall increase of the multi-processor system performance.
Abstract:
The invention relates to the use of certain fumaric acid monoalkyl esters as salts or a free acid either alone or in combination with a dialkyl fumarate for producing pharmaceutical preparations for use in transplantation medicine, especially for treating, alleviating or suppressing host-versus-graft reactions. For this purpose, the fumaric acid monoalkyl esters may be used in combination with preparations conventionally used in transplantation medicine and immuno suppressives, especially such as cyclosporines.
Abstract:
The present invention relates to the use of one or more salts of fumaric acid monoalkyl esters of the general formula optionally in admixture with dialkyl fumarate of the formula wherein A is a bivalent cation from the series consisting of Ca, Mg, Zn or Fe or a monovalent cation from the series Li, Na or K, respectively, and n denotes the numeral 1 or 2 depending on the type of cation, and, optionally, commonly used pharmaceutical excipients and vehicles for preparing a pharmaceutical composition in the form of micro-tablets or micro-pellets for the treatment of psoriatic arthritis, neurodermatitis, psoriasis and enteritis regionalis Crohn.
Abstract:
A portable device (10) for selective message recordation and playback and scheduling includes a portable unit (11) and an optional homebase station (12). Portable unit (11) includes a housing (15), unit controller (20), tactile input mechanism (40), audio processor (60) and video processor (80). Portable unit (11) allows textual, audible and visual message inputs and their selective playback based on time or other preselected textual, audible or visual stimulus. A selectively detachable stylus (51) may be removed to engage a touch panel (50) for making operational selections and entering information. Optional homebase station (12) may include a variety of features for supporting and supplementing those of handheld unit (11), such as a floppy drive (112) message memory, input keys (114), a battery charging circuit (115) and an external communication circuit (117).
Abstract:
A cache memory which is partitioned during write operations into several regions in a programmable or adaptive manner is disclosed. Each region is used to store a different type of data. The partitioning of the cache memory has no effect during read operations. The partitioning can be achieved dynamically for improved optimization. The partitioned cache is particularly suited for an instruction cache wherein one of the regions stores sequential instructions and another region stores branch target instructions. However, the partitioned cache is also useful for a data cache.
Abstract:
A device (10) adding apparatus as a keyboard input to a handheld computer (12) lacking such input (such as pen computers and personal digital assistants) includes a keyboard (11) and a planar motion pivot mechanism (30). Planar motion pivot mechanism (30) includes a body (31) formed as a boot that encloses an end of the handheld computer (12), and a pivot pin (32). The user simply rotates keyboard (11) from a nonoperational position substantially covering the handheld computer display (14), to an operational position at substantially a right angle to the display (14).
Abstract:
A device (10) adding keyboard input to a handheld computer (12) lacking such input (such as pen computers and personal digital assistants) includes a keyboard (11) and a planar motion pivot mechanism (30). Planar motion pivot mechanism (30) includes a body (31) formed as a boot that encloses an end of the handheld computer (12), and a pivot pin (32). The user simply rotates keyboard (11) from a nonoperational position substantially covering the handheld computer display (14), to an operational position at substantially a right angle to the display (14).
Abstract:
A scalable register file including first and second micro-register files organized in a pipelined fashion to minimize the access time of the register file where there are a large number of registers or multiple functional units. Interposed between the first and second micro-register files are a first plurality of pipeline registers for storing the register contents fetched from the first micro-register file during a first pipeline cycle. A second plurality of pipeline registers are coupled to the second micro-register files for storing the register contents fetched from the second micro-register file during a second pipeline stage and those registers being stored in the first plurality of pipeline registers. The first plurality of pipeline registers are coupled to the bit lines of the second micro-register file. Enable logic is coupled to each of the first plurality pipeline registers to selectively present the contents of the first plurality of pipeline register to the second plurality of pipeline registers if there were contents stored in a first pipeline register during the first pipeline cycle. Alternatively, a multiplexer can be used to present the register contents stored in the first plurality of pipeline registers to the second plurality of pipeline registers.