Digital filters with memory
View Patent ↗A memory device that, in certain embodiments, includes a memory element coupled to a bit-line and a quantizing circuit coupled to the memory element via the bit-line. In some embodiments, the quantizing circuit includes an analog-to-digital converter having an input and output and a digital filter that includes memory. The input of the analog-to-digital converter may be coupled to the bit-line, and the output of the analog-to-digital converter may be coupled to the digital filter.
1. A method of writing to an internal data storage location, the method comprising:
receiving, in a quantizing circuit, a value that corresponds to data to be written to an internal data storage location;
locally storing the value in memory in the quantizing circuit;
adjusting a memory-element property that is used to store data;
reading data stored by the internal data storage location with the quantizing circuit; and
determining whether the data read from the internal data storage location is generally equal to the data to be written to the internal data storage location by comparing the locally stored value in memory to the data read from the internal data storage location;
wherein the value is a preset value that is approximately equal to the data being written times 2 m plus one-half of 2 m , where m is a number of least-significant digits of a counter that are truncated.
2. The method of claim 1 , comprising selecting the internal data storage location from among a plurality of internal data storage locations on an electrical conductor.
3. The method of claim 1 , wherein reading comprises converting an analog signal from the internal data storage location to a pulse-density modulated bit-stream.
4. The method of claim 1 , wherein the value is generally equal to the data being written.
5. The method of claim 1 , wherein locally storing the value comprises storing the value in memory in a counter.
6. The method of claim 5 , wherein the memory comprises SRAM.
7. The method of claim 5 , wherein reading comprises counting with an array of cascaded flip-flops comprised by the counter.
8. A method of writing to an internal data storage location, the method comprising:
calculating a preset value based on data to be written to an internal data storage location and a signal value that indicates whether the internal data storage location stores the data to be written, wherein the signal value is independent of the data to be written;
presetting a quantizing circuit according to the preset value;
adjusting a property of the internal data storage location that is used to store data;
reading a value from the internal data storage location with the quantizing circuit;
determining whether the internal data storage location stores the data to be written by determining whether the quantizing circuit outputs the signal value; and
adjusting the property of the internal data storage location a second time if the quantizing circuit does not output the signal value;
wherein calculating a preset value comprises multiplying the data to be written to the internal data storage location by 2 m , where m is a number of least-significant digits of an output of the quantizing circuit that are truncated and adding plus or minus one-half of 2 m to the product of the multiplication.
9. The method of claim 8 , comprising presetting with the quantizing circuit the second time if the quantizing circuit does not output the signal value.
10. The method of claim 8 , comprising locally storing the preset value in memory in the quantizing circuit.
11. The method of claim 8 , wherein:
calculating a preset value comprises inverting each digit of a number; and
each digit of the signal value is approximately equal to one.
12. The method of claim 8 , wherein:
calculating a preset value comprises inverting each digit of a number and adding one; and
each digit of the signal value is approximately equal to zero.
13. The method of claim 8 , wherein:
calculating a preset value comprises subtracting a number from one-half of one plus the largest value of a counter comprised by the quantizing circuit; and
the signal value is an output of the counter in which the most-significant digit is substantially equal to one.
14. The method of claim 8 , wherein:
calculating a preset value comprises adding a number to one-half of the largest value of a counter comprised by the quantizing circuit; and
the signal value is an output of the counter in which the most-significant digit is substantially equal to zero.