Home | Links | Contact Us | More About Intellectual Property | Bookmark
Search patents:
Home Databases Charge-domain-vector-matrix-product-processing-system

 Systematic skew reduction through buffer resizing
The present invention addresses the foregoing problems by systematically aligning delay ranges at ...


 Systematic approach for regularity extraction
1. Overview The invention includes a general approach to extract functional regularity for ...


 Method for designing semiconductor integrated circuit
An object of the present invention is eliminating redundant registers as many as possible while ...


 Method of extracting layout parasitics for nets of an integrated circuit using a connectivity-based approach
A layout parasitic extraction system is disclosed. The system may be coupled with layout network ...


 Method for channel routing, and apparatus
OF THE PREFERRED EMBODIMENT It is an advantage of the present invention, that it can be used to ...


 Method for assessing the reliability of interconnects
What is claimed is: 1. A method for testing interconnect structures, comprising: forming a ...


 Partitioning and reordering methods for static test sequence compaction of sequential circuits
The present invention included methods of compacting a sequential circuit test vector set by ...


 Worst case design parameter extraction for logic technologies
The present invention provides for more realistic worst case extreme determinations for an ...


 Fusible resistor
It is the object of the present invention to provide a fusible resistor which meets all of these ...


 Resistor insertion fuse
What I claim is: 1. A fuse for use in a relatively high voltage circuit comprising: a first ...


 Charge domain vector-matrix product processing system

Details
Inventors: Chiang, Alice M.;
Assignee: Massachusetts Institute of Technology (Cambridge, MA)
Primary Examiner: Smith; Jerry
Assistant Examiner: Trammell; Jim
Attorney, Agent or Firm: Engellenner; Thomas J., Lappin; Mark G.

A charge domain vector-matrix product processing system. The system includes a charge coupled device tapped delay line, an array of digital parallel shift register memory devices, and a signal processor. A sampled analog signal is stored within the tapped delay line, and multiple vectors of m-bit words are stored within the digital memory device. The signal processor sucessively applies vectors from the digital memory device and charge packets from the tapped delay line to an array of digital-analog multipliers. The signal processor then sums the outputs of the digital-analog multipliers and produces an output charge packet corresponding to a respective element of the vector-matrix product.

DETAILED DESCRIPTION The present invention is an improved charge domain vector-matrix product network for generating the signals representative of the product of an N-element vector and an N.
times.
K element matrix.
The network includes a charge coupled device (CCD) N-stage tapped delay line, an N.
times.
K M-bit digital parallel shift register memory device, N M-bit charge domain digital-analog multipliers, a charge summing device and a controller.
In the preferred form of the invention, the shift register memory device is a charge domain device, but in other forms of the invention, different types may be used, such as CMOS.
The tapped delay line is adapted to establish a succession of N charge packets therein in response to a succession of N applied input signals, where each packet has a magnitude corresponding to one of the element of the vector.
In the delay line, the charge packets are shifted from stage-to-stage along said delay line as the vector data is applied to the product network.
The delay line includes N floating gate sensing electrodes, each overlying one of the stages of the line and being adapted to provide a potential representative of the magnitude of a charge package currently within the underlying stage.
The digital parallel shift register memory device is adapted to storing N.
times.
K M-bit words, where each of the words is representative of the value of a corresponding element of the matrix.
In the preferred form, the shift register memory device includes K stages in a stack configuration, each stage being adapted to store N M-bit words.
The shift register memory device is responsive to an applied shift signal to selectively shifting the stored words from stage-to-stage from the input stage to the output stage of the stack.
The digital-analog multipliers each an analog input port and a digital input port, and provide a charge packet having a magnitude proportional to the product of a potential applied to the analog input port and an M-bit digital signal applied to the digital input port



Related patents
  Bipolar transistor with particular base structure
The present invention was made in order to solve the foregoing problems, and its object is to reduce the parasitic capacitance between the base and the collector in the ...
  Method and apparatus for extracting parameters for an electrical structure
OF THE PREFERRED EMBODIMENT The present invention provides for parameter extraction for an electrical structure. This parameter extraction is based on a definition of ...
  Automatic performing system capable of detection and correction of errors in performance information
Wherefore, an object of this invention is to provide an automatic performing system in which an error in the transmission of performance information is corrected by ...
  System for determining the operations of an integrated circuit and processor for use therein
A first embodiment of the present invention is shown in FIG. 2. The system comprises a translator 2 which receives information supplied by the manufacturer concerning ...
  Logic equation fault analyzer
What is claimed is: 1. A computer system including: A) a plurality of system modules, each of said system modules including means for sensing the occurrence of a ...
  Methods and test platforms for developing an application-specific integrated circuit
What is claimed is: 1. A method for developing a programmed ASIC integrated circuit which includes a signal processor core, a RAM memory and a ROM memory for management ...
  Method for testing reflection LCD projector and display panel pixel area thereof
Accordingly, it is the object of the present invention to provide a highly reliable and efficient testing method for a CMOS silicon wafer LCD in a reflection-type LCD ...
  Audio sample tracker
Briefly, and in general terms, the present invention provides a system for minimizing the error between a future state of the system after a delay and the predicted ...
  Optical proximity correction system
OF THE INVENTION Referring now to FIG. 14, there is illustrated a simplified procedure for mask OPC according to a first embodiment of the present invention. When ...
  Modifying a design layer of an integrated circuit using overlying and underlying design layers
The problems outlined above are in large part solved by the proximity correction technique hereof in which a design layer of an integrated circuit is altered by spatial ...

0.054

Archive: All patents - Links

Copyright (c)2006 Eipa-patents.org - All rights reserved