公开(公告)号：US20230140784A1

公开(公告)日：2023-05-04

申请号：US17995859

申请日：2020-09-29

Applicant: INSTITUTE OF MICROELECTRONICS OF THE CHINESE ACADEMY OF SCIENCES

Inventor： Feng ZHANG ,  Zhisheng CHEN

IPC: G06K7/10

Abstract: A signal processing method is applied to an RFID electronic tag, and includes: coding a digital baseband signal to obtain a coded signal; performing phase-shift keying modulation on the coded signal to obtain a first modulated signal; performing OFDM modulation on the first modulated signal to obtain a second modulated signal; and sending the second modulated signal to an RFID reader, by means of which the OFDM demodulation, phase-shift keying demodulation, and decoding are performed sequentially on the second modulated signal. According to the signal processing method and device, and the RFID system of one or more embodiments of present disclosure, the RFID system can be caused to effectively utilize bandwidth, thereby achieving high-speed transmission of signals and significantly reducing a bit error ratio of signal transmission.

公开(公告)号：US20240153554A1

公开(公告)日：2024-05-09

申请号：US17756315

申请日：2021-07-02

Applicant: INSTITUTE OF MICROELECTRONICS OF THE CHINESE ACADEMY OF SCIENCES

Inventor： Feng ZHANG ,  Qirui REN

IPC: G11C13/00

CPC classification number: G11C13/004 ,  G11C13/0059 ,  G11C2013/0045

Abstract: The present disclosure discloses a data readout circuit of a RRAM and a RRAM circuit. The data readout circuit of the RRAM comprises: an adaptive current sense amplifier (CSA) and a reference current generator, the adaptive CSA is configured to electrically connect to the RRAM, and the adaptive CSA is electrically connected to the reference current generator; the reference current generator is configured to generate a basic reference current; the adaptive CSA is configured to obtain a reference current according to the basic reference current and a bit-line current of the RRAM; and the adaptive CSA is configured to compare the size of the reference current and that of the bit-line current so as to read out stored data. The present disclosure can improve the problem of data readout error due to the degradation of high resistance state of the RRAM.

公开(公告)号：US20240305479A1

公开(公告)日：2024-09-12

申请号：US17756314

申请日：2021-04-19

Applicant: INSTITUTE OF MICROELECTRONICS OF THE CHINESE ACADEMY OF SCIENCES

Inventor： Feng ZHANG ,  Yiming WANG ,  Qirui REN

IPC: H04L9/32 ,  H04L9/08

CPC classification number: H04L9/3278 ,  H04L9/0869 ,  H04L2209/12

Abstract: An encryption method includes: receiving cipher data which is binary data; determining target components in a resistive memory array according to values of respective bits in the cipher data; determining current values generated by respective columns of components according to the target components; and generating key data according to the current values generated by the respective columns of components. The present disclosure can effectively reduce computing power and power consumption of an encryption process in an edge device.

公开(公告)号：US20230162007A1

公开(公告)日：2023-05-25

申请号：US17753140

申请日：2021-02-22

Applicant: INSTITUTE OF MICROELECTRONICS OF THE CHINESE ACADEMY OF SCIENCES

Inventor： Feng ZHANG ,  Qiang HUO

IPC: G06N3/0464 ,  G06N3/065

CPC classification number: G06N3/0464 ,  G06N3/065

Abstract: The present disclosure discloses a method and apparatus for convolution operation of a convolutional neural network. The method comprises acquiring input voltages used for characterizing pixel values; when the input voltages are scanned through convolutional sliding windows, obtaining times of reusing of the input voltages in the convolutional sliding windows; grouping the input voltages based on a difference in the times of reusing of the input voltages; extracting the input voltages in same groups once and performing convolution calculation with convolution kernels respectively, to obtain a result corresponding to each group; obtaining a result of convolution operation based on the result corresponding to each group, to implement convolution operation in the convolutional neural network. The present disclosure reduces energy consumption during convolution operations effectively.

公开(公告)号：US20220244872A1

公开(公告)日：2022-08-04

申请号：US17595847

申请日：2021-04-22

Applicant: INSTITUTE OF MICROELECTRONICS OF THE CHINESE ACADEMY OF SCIENCES

Inventor： Feng ZHANG ,  Qiang HUO ,  Zhisheng CHEN ,  Qirui REN

IPC: G06F3/06 ,  G11C13/00

Abstract: The present disclosure discloses a storage method, a data processing method, a device and an apparatus based on a non-volatile memory, the method comprising: acquiring a weight value that needs to be stored in the non-volatile memory; determining a conductivity value corresponding to the weight value according to a first conversion method if the non-volatile memory is a high-resistance storage device; determining a conductivity value corresponding to the weight value according to a second conversion method which is different from the first conversion method if the non-volatile memory is a low-resistance memory device; and setting the non-volatile memory according to the conductivity value to store the weight value. The non-volatile memory and the data processing method, the device, and the apparatus provided by the present disclosure solve the problem of insufficient accuracy in weight values in existing non-volatile memories, realizing the technical effect of improving storage accuracy and data processing accuracy.