In-memory computation device and in-memory computation method to perform multiplication operation in memory cell array according to bit orders
An in-memory computation device and computation method are provided. The in-memory computation method includes: providing a memory cell block of a memory cell array to store a plurality of weight values, and providing a plurality of memory cells on the memory cell block to store a plurality of corresponding bits of each of the weight values; respectively transmitting a plurality of input signals to the plurality of bit lines through an input buffer; providing the plurality of memory cells to perform a multiplication operation of the plurality of input signals and the plurality of weight values to generate a plurality of first operation results respectively corresponding to a plurality of bit orders; and performing an addition operation on the plurality of first operation results to generate a second operation result according to the plurality of bit orders by a sense amplifier.
1. An in-memory computation method, comprising:
providing a memory cell block of a memory cell array to store a plurality of weight values, and providing a plurality of memory cells on the memory cell block to store a plurality of corresponding bits of each of the weight values;
respectively transmitting a plurality of input signals to the plurality of bit lines through an input buffer;
providing the plurality of memory cells to perform a multiplication operation of the plurality of input signals and the plurality of weight values to generate a plurality of first operation results respectively corresponding to a plurality of bit orders; and
performing an addition operation on the plurality of first operation results to generate a second operation result according to the plurality of bit orders by a sense amplifier.
2. The in-memory computation method according to claim 1 , wherein a step of respectively transmitting the plurality of input signals to the plurality of bit lines through the input buffer comprises:
providing the input buffer to transmit a plurality of bits of each of the input signals to a plurality of corresponding bit lines in a time-sharing manner.
3. The in-memory computation method according to claim 1 , wherein a step of respectively transmitting the plurality of input signals to the plurality of bit lines through the input buffer comprises:
providing the input buffer to transmit a plurality of bits of each of the input signals to a plurality of corresponding bit lines in parallel,
wherein a number of the plurality of corresponding bit lines is equal to multiplication of a bit number of each of the weight values and a bit number of each of the input signals.
4. The in-memory computation method according to claim 3 , wherein the plurality of corresponding bit lines in a same bit order corresponding to each of the input signals are adjacently arranged.
5. The in-memory computation method according to claim 3 , wherein the plurality of corresponding bit lines are sequentially arranged according to a bit order of the plurality of bits of each of the corresponding input signal.
6. The in-memory computation method according to claim 3 , further comprising:
generating an input arrangement sequence according to a bit order of the correspondingly generated plurality of first operation results by the plurality of bits of each of the input signal; and
providing the input buffer to respectively transmit the plurality of bits of each of the input signals to the plurality of corresponding bit lines according to the input arrangement sequence.
7. The in-memory computation method according to claim 3 , wherein the memory cell block comprises a plurality of memory columns, and the at least one word line is a plurality of word lines and respectively corresponds to the memory columns, the in-memory computation method further comprises:
providing each of the memory columns to store at least one of the plurality of bits of the plurality of weight values in a field corresponding to each of the input signals.