Abstract:
Die Erfindung betrifft einen Verfahren zum Betreiben eines Analog Dividierers, wobei ein Sägezahn- oder Dreieckssignal aus einer erste Eingangsspannung (U1) als Divisor gebildet wird, und wobei dieses Sägezahn- oder Dreieckssignal mittels eines ersten Komparators (KO1) mit einer zweiten Eingangsspannung (U2) als Dividend in der Weise verglichen wird, dass als ein erstes Vergleichssignal (SIG1 OUT ) ein pulsweitenmoduliertes Signal erzeugt wird, dessen Mittelwert als Quotienten der Division ausgegeben wird. Dabei wird das Sägezahn- oder Dreieckssignal mittels eines ersten und eines zweiten Reglers (REG1, REG2) gebildet und dem ersten Regler (REG1) die erste Eingangsspannung (U1) oder eine dazu proportionale Spannung und das Sägezahn- oder Dreieckssignal oder ein dazu proportionales Signal in der Weise zugeführt, dass der obere Spitzenwert des Sägezahn- oder Dreieckssignals der ersten Eingangsspannung nachgeregelt wird. Des Weiteren wird dem zweiten Regler (REG2) das Bezugspotenzial und das Sägezahn- oder Dreieckssignal in der Weise zugeführt, dass der untere Spitzenwert des Sägezahn- oder Dreieckssignals dem Wert des Bezugspotenzials nachgeregelt wird.
Abstract:
A current mode multiplier circuit is provided based on the square root voltage-current relationship of an MOS transistor. The circuit includes first, second and third MOS transistors with a common aspect ratio, and first and second current sources that respectively provide first and second input currents that represent first and second factors to be multiplied. The first and second MOS transistors produce first and second voltages as a function of the first and second input currents, and the third MOS transistor produces a third current as a function of the first and second voltages. In response to the third current, the circuit produces a product signal that represents a product of the first and second factors.
Abstract:
Schaltungsanordnung zur Parametereinstellung mit mindestens einer ersten analogen Multipliziereinrichtung (1, 2, 3), der ein Eingangssignal (17, 18, 19) sowie ein einem Parameter entsprechendes Steuersignal zugeführt wird und die ein Ausgangssignal (20, 21, 22) abgibt, mit einer zur ersten Multiplikationseinrichtung (1, 2, 3) identischen zweiten Multiplikationseinrichtung (25), der ein erstes Referenzsignal sowie ein dem ersten Steuersignal entsprechendes zweites Steuersignal zugeführt werden und die ein Ausgangssignal (Ui) abgibt, und mit einer Regeleinrichtung (38 bis 45), die das Ausgangssignal (Ui) der zweiten Multipliziereinrichtung (25) mit einem zweiten Referenzsignal (Us) vergleicht und daraus die Steuersignale ableitet.
Abstract:
In one example in accordance with the present disclosure a device is described. The device includes at least two memristive cells. Each memristive cell includes a memristive element to store one component of a complex weight value. The device also includes a real input multiplier coupled to the memristive element to multiply an output signal of the memristive element with a real component of an input signal. An imaginary input multiplier of the device is coupled to the memristive element to multiply the output signal of the memristive element with an imaginary component of the input signal.
Abstract:
In one example in accordance with the present disclosure a device is described. The device includes at least two memristive cells. Each memristive cell includes a memristive element to store one component of a complex weight value. The device also includes a real input multiplier coupled to the memristive element to multiply an output signal of the memristive element with a real component of an input signal. An imaginary input multiplier of the device is coupled to the memristive element to multiply the output signal of the memristive element with an imaginary component of the input signal.
Abstract:
Examples herein relate to circuits for computing vector-matrix multiplications involving negative values. A first memory crossbar array may be mapped to a first matrix which includes the positive values of an input matrix. A second memory crossbar array may be mapped to a second matrix which includes the negative values of the input matrix. An analog-to-digital converter may generate digital intermediate multiplication results based on analog results computed by the memory crossbar arrays. The digital intermediate multiplication results may include an intermediate result corresponding to a multiplication of each of the first vector and second vector with each of the first matrix and the second matrix. A controller may aggregate the digital intermediate results to generate a combined multiple result that represents the vector-matrix multiplication of the input vector and the input matrix.
Abstract:
Provided in one example is a nonvolatile memory cross-bar array. The array includes: a number of junctions formed by a number of row lines intersecting a number of column lines; a first set of controls at a first set of the junctions coupling between a first set of the row lines and a first set of the column lines; a second set of controls at a second set of the junctions coupling between a second set of the row lines and a second set of the column lines; and a current collection line to collect currents from the controls of the first set and the second set through their respective column lines and output a result current corresponding to a sum of a first dot product and a second dot product.
Abstract:
A method of obtaining a dot product using a memristive dot product engine with a nulling amplifier includes applying a number of programming voltages to a number of row lines within a memristive crossbar array to change the resistance values of a corresponding number of memristors located at intersections between the row lines and a number of column lines. The method also includes applying a number of reference voltages to the number of the row lines and applying a number of operating voltages to the number of the row lines. The operating voltages represent a corresponding number of vector values. The method also includes determining an array output based on a reference output and an operating output collected from the number of column lines.
Abstract:
An analog multiplier includes a bias circuit, a level shifter, a multiplying circuit, and a current mirror. The analog multiplying circuit is used for inputting a first voltage and a second voltage, and outputting a product current. The product current is proportional to a product of the first voltage and the second voltage. The analog multiplier is implemented by a few devices, thereby having a simple architecture and being capable of being driven by a small amount of power.
Abstract:
A high-speed analog multiplier circuit (Figs. 3-6) for multiplying two analog inputs (X, Y) comprises a signum generator (21) having an X input multiplier (at 26, 27) and having an output connected to a high-speed electronic switch (22). A Y analog input multiplicand (at 28) is also connected to the high-speed electronic switch (22) and the output of the electronic switch (32, 33) having a positive input (33), a negative input (32) and a resultant output (24) for producing the signum function (X) times Y. The electronic switch (22) is operated by the signum generator (21) so that the presence of a high Q output from the signum generator (21) is effective to connect the Y multiplicand input (28) to the positive input (33) of the differential amplifier (23) and that the presence of a high Q input (27) from the signum generator (21) is effective to connect the Y multiplicand input (28) to the negative input (32) of the differential amplifier (23).