CN112558922A - Four-transistor memory computing device based on separated word lines - Google Patents
Four-transistor memory computing device based on separated word lines Download PDFInfo
- Publication number
- CN112558922A CN112558922A CN202110190886.0A CN202110190886A CN112558922A CN 112558922 A CN112558922 A CN 112558922A CN 202110190886 A CN202110190886 A CN 202110190886A CN 112558922 A CN112558922 A CN 112558922A
- Authority
- CN
- China
- Prior art keywords
- transistor
- word line
- bit
- bit line
- gate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F7/00—Methods or arrangements for processing data by operating upon the order or content of the data handled
- G06F7/38—Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation
- G06F7/48—Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state device; using unspecified devices
- G06F7/544—Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state device; using unspecified devices for evaluating functions by calculation
- G06F7/5443—Sum of products
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/063—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Optimization (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Neurology (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Pure & Applied Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Semiconductor Memories (AREA)
Abstract
The invention provides a four-transistor memory computing device based on a separated word line, which comprises: m × n in-memory compute bit cells arranged in an array, the in-memory compute bit cells comprising: a transistor T1, a transistor T2, a transistor T3, and a transistor T4; the source of the transistor T1 and the source of the transistor T2 are both connected to a power supply, the drain of the transistor T1 is connected to the gate of the transistor T2, and the gate of the transistor T1 is connected to the drain of the transistor T2; the source of the transistor T3 is connected to the bit line BL, the drain of the transistor T3 is connected to the gate of the transistor T1, and the gate of the transistor T3 is connected to the word line WLL; the source of the transistor T4 is connected to the bit line BLB, the drain of the transistor T4 is connected to the gate of the transistor T2, and the gate of the transistor T4 is connected to the word line WLR. The invention realizes word line separation by using the structure of two word lines, has simple calculation logic, accelerates the calculation process and reduces the structure area.
Description
Technical Field
The invention relates to the technical field of memory computing, in particular to a four-transistor memory computing device based on a separation word line.
Background
Deep Convolutional Neural Networks (DCNNs) demonstrate that inference accuracy can be further improved, and deep learning is moving towards edge computation. The most common operation in DCNNs is Multiplication and Accumulation (MAC), which controls power and delay. The MAC operation has high regularity and parallelism, and is therefore very suitable for hardware acceleration. However, the amount of memory access severely limits the energy efficiency of conventional digital accelerators, and thus memory computing (IMC) is becoming increasingly attractive for DCNN acceleration.
The existing memory computing unit is basically based on six or more tubes, and obviously, the structure area is large, the computing process is slow, and the power consumption is also large.
Disclosure of Invention
The invention aims to provide a four-transistor memory computing device based on a separated word line, which is used for accelerating a computing process and reducing the structural area.
To achieve the above object, the present invention provides a split word line based quad-memory computing device, comprising:
the device comprises a column decoding driver, a row decoding driver, a memory computing array, n switches and n analog-digital converters; the in-memory computation array comprises m multiplied by n in-memory computation bit units arranged in an array;
the n first bit line ends of the column decoding drivers are respectively connected with the n bit lines BL; n second bit line ends of the column decoding drivers are respectively connected with n bit lines BLB;
m first word line ends of the row decoding driver are respectively connected with m word lines WLL; m second word line ends of the row decoding driver are respectively connected with m word lines WLR;
the first bit line ends corresponding to the m memory computing bit units in the j +1 th column are all connected with the j th bit line BL, the second bit line ends corresponding to the m memory computing bit units in the j +1 th column are all connected with the j th bit line BLB, the first word line ends corresponding to the n memory computing bit units in the i +1 th row are all connected with the i th word line WLL, and the second word line ends corresponding to the n memory computing bit units in the i +1 th row are all connected with the i th word line WLR; wherein i is a positive integer greater than or equal to 0 and less than m, and j is a positive integer greater than or equal to 0 and less than n;
the jth switch is connected with a jth bit line BLB, and a first input end of the jth analog-digital converter is respectively connected with the jth switch and the jth bit line BL;
the in-memory computation bit cell comprises: a transistor T1, a transistor T2, a transistor T3, and a transistor T4;
the source of the transistor T1 and the source of the transistor T2 are both connected to a power supply, the drain of the transistor T1 is connected to the gate of the transistor T2, and the gate of the transistor T1 is connected to the drain of the transistor T2;
the source of the transistor T3 is connected to the bit line BL, the drain of the transistor T3 is connected to the gate of the transistor T1, and the gate of the transistor T3 is connected to the word line WLL;
the source of the transistor T4 is connected to the bit line BLB, the drain of the transistor T4 is connected to the gate of the transistor T2, and the gate of the transistor T4 is connected to the word line WLR.
Optionally, the transistor T1 and the transistor T2 are both PMOS, and the transistor T3 and the transistor T4 are both NMOS.
Alternatively, when the input activation signal is +1, the word line WLL is VDD, and the word line WLR is 0; when the input activation signal is-1, the word line WLL is 0, and the word line WLR is VDD; where VDD = 1V.
Optionally, when Q =0, the weight value is-1; when Q = VDD, the weight value is +1, where VDD =1V, and Q is a common point of the gate of the transistor T1, the drain of the transistor T2, and the drain of the transistor T3.
Optionally, when the input activation signal is +1 and the weight value is +1, the bit line BL is charged; when the input activation signal is-1 and the weight value is +1, discharging the bit line BLB; when the input activation signal is +1 and the weight value is-1, discharging the bit line BL; when the input activation signal is-1 and the weight value is-1, the bit line BLB is charged.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the memory computing bit unit of the invention adopts 4 transistors to carry out memory computing, further optimizes the array structure, reduces the array area, directly saves the computing circuit, realizes word line separation by using the structure of two word lines, further reduces the area of the memory computing device, and simultaneously realizes quick reasoning time and robustness.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a diagram of an in-memory computing device according to the present invention;
FIG. 2 is a diagram of a memory compute bitcell structure of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a four-transistor memory computing device based on a separated word line, which is used for accelerating a computing process and reducing the structural area.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
As shown in fig. 1, the present invention discloses a four-transistor memory computing device based on separated word lines, the device comprising: a column decoding driver II, a row decoding driver III, a memory computing array, n switches and n analog-digital converters; the in-memory calculation array comprises m multiplied by n in-memory calculation bit units arranged in an array. The column decoding driver has a column decoding function and a read/write bit line control function; the row decoding driver has a row decoding function and a word line driving function.
N first bit line ends of the column decoding driver II are respectively connected with n bit lines BL; n second bit line ends of the column decoding driver II are respectively connected with n bit lines BLB; the m first word line ends of the row decoding driver III are respectively connected with m word lines WLL; the m second word line ends of the row decoding driver III are respectively connected with m word lines WLR; the first bit line ends corresponding to the m in-memory calculation bit units (i) of the j +1 th column are all connected with the jth bit line BL (namely, the bit line BL [ j ]), the second bit line ends corresponding to the m in-memory calculation bit units (i) of the j +1 th column are all connected with the jth bit line BLB (namely, the bit line BLB [ j ]), the first word line ends corresponding to the n in-memory calculation bit units (i) of the i +1 th row are all connected with the ith word line WLL (namely, the bit line WLL [ i ]), and the second word line ends corresponding to the n in-memory calculation bit units (i) of the i +1 th row are all connected with the ith word line WLR (namely, the bit line WLR [ i ]); wherein i is a positive integer greater than or equal to 0 and less than m, and j is a positive integer greater than or equal to 0 and less than n; the jth switch (i.e., SW [ j ]) is connected to the jth bit line BLB (i.e., BLBj), the first input terminal of the jth analog-to-digital converter (i.e., Qj) is connected to the jth switch and the jth bit line BL, respectively, the second input terminal of the jth analog-to-digital converter is connected to the given value Vref, and finally the accumulated voltage is output through the output terminal of the jth analog-to-digital converter.
The memory computing device of the present invention can operate in two modes: the first is the access mode of the weight access. The second is the calculation mode for binary multiply-accumulate bMAC operation.
In the access mode, the read/write operations of the memory computing device are the same as the read/write operations of a conventional 6T SRAM cell. Namely: the address signal is decoded by the row decoding driver and then input to the word line WLL [ i ] and the word line WLR [ i ], and the (i + 1) th row n memory computing bit units are selected, the bit line BL [ j ] and the bit line BLB [ j ] output by the column decoding driver select the (j + 1) th row m memory computing bit units, the bit line BL [ j ] and the bit line BLB [ j ] close the switch SW [ j ], so that the voltage calculated by the (i + 1) th row j +1 th row memory computing bit units is input to the bit line BL [ j ] and combined with the voltage on the bit line BLB [ j ], and then output through the output end of the analog-digital converter Q [ j ].
In the calculation mode, m rows of memory bit units (phi) are activated simultaneously, each row of Input activation signals (Input) are pre-coded into a word line (WLL) and a Word Line (WLR), and weight values are stored in each memory bit unit (phi). When the Input activation signal Input [ i ] is '+ 1', the corresponding word line WLL [ i ] is set to '1', and the corresponding word line WLR [ i ] is set to '0'. When an activation signal Input [ i ] = '-1' is Input, the word line WLL [ i ] is set to 0, and the word line WLR [ i ] is set to 1. Bit line BL [ j ] and bit line BLB [ j ] are precharged to 0.5V. When the result of multiplication between the Input activation signal Input [ i ] and the weight value is "+ 1", the bit line BL [ j ] or the bit line BLB [ j ] is charged, and when the result is '-1', the bit line BL [ j ] or the bit line BLB [ j ] is discharged, and the result of multiplication in the j +1 th column is accumulated in the form of voltage on the bit line BL [ j ] and the bit line BLB [ j ]. Then opening switch SW [ j ], merging the voltages on bit line BL [ j ] and bit line BLB [ j ] to generate total voltage value VBL [ j ] to be outputted in digital mode through the output end of analog-digital converter Q [ j ]. According to the invention, a plurality of bit lines BL are output in parallel, and complete parallel calculation and high throughput are realized.
As shown in fig. 2, the in-memory compute bit cell (r) includes: a transistor T1, a transistor T2, a transistor T3, and a transistor T4; the source of the transistor T1 and the source of the transistor T2 are both connected to the power supply VDD, the drain of the transistor T1 is connected to the gate of the transistor T2, and the gate of the transistor T1 is connected to the drain of the transistor T2; the source of the transistor T3 is connected to the bit line BL, the drain of the transistor T3 is connected to the gate of the transistor T1, and the gate of the transistor T3 is connected to the word line WLL; the source of the transistor T4 is connected to the bit line BLB, the drain of the transistor T4 is connected to the gate of the transistor T2, and the gate of the transistor T4 is connected to the word line WLR. Q is a common point of the gate of the transistor T1, the drain of the transistor T2, and the drain of the transistor T3, and QB is a common point of the gate of the transistor T2, the drain of the transistor T1, and the drain of the transistor T4.
The in-memory calculation bit unit (i) comprises the following steps: the first step of pre-charging makes the bit line BL [ j ] and bit line BLB [ j ] charged to 0.5V, the second step of charging is closed, the input activation signal is transmitted to the word line WLL [ i ]/WLR [ i ] through the row decoding driver, the multiplication result of the input activation signal and the weighted value generates charging and discharging on the bit line BL [ j ]/BLB [ j ]; and thirdly, performing analog-to-digital conversion by an analog-to-digital converter Q [ j ] and outputting a result, wherein a specific multiplication and accumulation operand table is shown in table 1.
TABLE 1 multiply-accumulate operand table
As shown in table 1, Input is an Input activation signal, Weight is a Weight value, value is a voltage value obtained by combining two parts, and Q is a common point of the gate of the transistor T1, the drain of the transistor T2, and the drain of the transistor T3. In table 1, the value of the Input activation signal Input is represented by a combination of high and low levels of the word line WLL and the word line WLR, and when the Input activation signal is +1, the terminal voltage of the word line WLL is VDD (1V), and the word line WLR is 0V; when the input activation signal is-1, the terminal voltage of a word line WLL is 0V, and the word line WLR is VDD; when the input activation signal is 0, word lines WLL and WLR are both VRST, i.e., 0.5V. When Q =0, the weight value is-1; when Q = VDD, the weight value is + 1. When the input activation signal is +1 and the weight value is +1, charging the bit line BL, wherein the value voltage value is 1; when the input activation signal is-1 and the weight value is +1, discharging the bit line BLB, wherein the value voltage value is-1; when the input activation signal is +1 and the weight value is-1, discharging the bit line BL, wherein the value voltage value is-1; when the input activation signal is-1 and the weight value is-1, the bit line BLB is charged, and the value voltage is 1.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to assist in understanding the core concepts of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (5)
1. A split word line based quad-memory computing device, the device comprising:
the device comprises a column decoding driver, a row decoding driver, a memory computing array, n switches and n analog-digital converters; the in-memory computation array comprises m multiplied by n in-memory computation bit units arranged in an array;
the n first bit line ends of the column decoding drivers are respectively connected with the n bit lines BL; n second bit line ends of the column decoding drivers are respectively connected with n bit lines BLB;
m first word line ends of the row decoding driver are respectively connected with m word lines WLL; m second word line ends of the row decoding driver are respectively connected with m word lines WLR;
the first bit line ends corresponding to the m memory computing bit units in the j +1 th column are all connected with the j th bit line BL, the second bit line ends corresponding to the m memory computing bit units in the j +1 th column are all connected with the j th bit line BLB, the first word line ends corresponding to the n memory computing bit units in the i +1 th row are all connected with the i th word line WLL, and the second word line ends corresponding to the n memory computing bit units in the i +1 th row are all connected with the i th word line WLR; wherein i is a positive integer greater than or equal to 0 and less than m, and j is a positive integer greater than or equal to 0 and less than n;
the jth switch is connected with a jth bit line BLB, and a first input end of the jth analog-digital converter is respectively connected with the jth switch and the jth bit line BL;
the in-memory computation bit cell comprises: a transistor T1, a transistor T2, a transistor T3, and a transistor T4;
the source of the transistor T1 and the source of the transistor T2 are both connected to a power supply, the drain of the transistor T1 is connected to the gate of the transistor T2, and the gate of the transistor T1 is connected to the drain of the transistor T2;
the source of the transistor T3 is connected to the bit line BL, the drain of the transistor T3 is connected to the gate of the transistor T1, and the gate of the transistor T3 is connected to the word line WLL;
the source of the transistor T4 is connected to the bit line BLB, the drain of the transistor T4 is connected to the gate of the transistor T2, and the gate of the transistor T4 is connected to the word line WLR.
2. The split word line-based quad-transistor memory computing device of claim 1, wherein the transistors T1 and T2 are both PMOS and the transistors T3 and T4 are both NMOS.
3. The split-word-line-based quad-memory computing device of claim 1, wherein when the input activation signal is +1, word line WLL is VDD and word line WLR is 0; when the input activation signal is-1, the word line WLL is 0, and the word line WLR is VDD; where VDD = 1V.
4. The discrete word line-based quad-memory computing device of claim 1, wherein when Q =0, the weight value is-1; when Q = VDD, the weight value is +1, where VDD =1V, and Q is a common point of the gate of the transistor T1, the drain of the transistor T2, and the drain of the transistor T3.
5. The split-word-line-based quad-transistor memory computing device of claim 1, wherein when an input activation signal is +1 and a weight value is +1, then charging a Bit Line (BL); when the input activation signal is-1 and the weight value is +1, discharging the bit line BLB; when the input activation signal is +1 and the weight value is-1, discharging the bit line BL; when the input activation signal is-1 and the weight value is-1, the bit line BLB is charged.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110190886.0A CN112558922A (en) | 2021-02-20 | 2021-02-20 | Four-transistor memory computing device based on separated word lines |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110190886.0A CN112558922A (en) | 2021-02-20 | 2021-02-20 | Four-transistor memory computing device based on separated word lines |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112558922A true CN112558922A (en) | 2021-03-26 |
Family
ID=75034379
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110190886.0A Pending CN112558922A (en) | 2021-02-20 | 2021-02-20 | Four-transistor memory computing device based on separated word lines |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112558922A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114530176A (en) * | 2022-04-25 | 2022-05-24 | 中科南京智能技术研究院 | Distributed bit line compensation digital-analog mixed memory computing array |
CN114882921A (en) * | 2022-07-08 | 2022-08-09 | 中科南京智能技术研究院 | Multi-bit computing device |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10510386B1 (en) * | 2018-08-29 | 2019-12-17 | National Tsing Hua University | Dynamic bit-line clamping circuit for computing-in-memory applications and clamping method thereof |
CN111883191A (en) * | 2020-07-14 | 2020-11-03 | 安徽大学 | 10T SRAM cell, and memory logic operation and BCAM circuit based on 10T SRAM cell |
CN111899775A (en) * | 2020-07-24 | 2020-11-06 | 安徽大学 | SRAM memory cell circuit capable of realizing multiple logic functions and BCAM operation |
CN112036562A (en) * | 2020-11-05 | 2020-12-04 | 中科院微电子研究所南京智能技术研究院 | Bit cell applied to memory computation and memory computation array device |
CN112071344A (en) * | 2020-09-02 | 2020-12-11 | 安徽大学 | Circuit for improving linearity and consistency of calculation in memory |
-
2021
- 2021-02-20 CN CN202110190886.0A patent/CN112558922A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10510386B1 (en) * | 2018-08-29 | 2019-12-17 | National Tsing Hua University | Dynamic bit-line clamping circuit for computing-in-memory applications and clamping method thereof |
CN111883191A (en) * | 2020-07-14 | 2020-11-03 | 安徽大学 | 10T SRAM cell, and memory logic operation and BCAM circuit based on 10T SRAM cell |
CN111899775A (en) * | 2020-07-24 | 2020-11-06 | 安徽大学 | SRAM memory cell circuit capable of realizing multiple logic functions and BCAM operation |
CN112071344A (en) * | 2020-09-02 | 2020-12-11 | 安徽大学 | Circuit for improving linearity and consistency of calculation in memory |
CN112036562A (en) * | 2020-11-05 | 2020-12-04 | 中科院微电子研究所南京智能技术研究院 | Bit cell applied to memory computation and memory computation array device |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114530176A (en) * | 2022-04-25 | 2022-05-24 | 中科南京智能技术研究院 | Distributed bit line compensation digital-analog mixed memory computing array |
CN114882921A (en) * | 2022-07-08 | 2022-08-09 | 中科南京智能技术研究院 | Multi-bit computing device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112151091B (en) | 8T SRAM unit and memory computing device | |
CN112151092B (en) | Storage unit, storage array and in-memory computing device based on 4-pipe storage | |
CN112992223B (en) | Memory computing unit, memory computing array and memory computing device | |
CN111816232B (en) | In-memory computing array device based on 4-pipe storage structure | |
CN111816231B (en) | Memory computing device with double-6T SRAM structure | |
CN113255904B (en) | Voltage margin enhanced capacitive coupling storage integrated unit, subarray and device | |
CN113035251B (en) | Digital memory computing array device | |
CN112558919B (en) | Memory computing bit unit and memory computing device | |
CN112036562B (en) | Bit cell applied to memory computation and memory computation array device | |
CN114089950B (en) | Multi-bit multiply-accumulate operation unit and in-memory calculation device | |
CN113257306B (en) | Storage and calculation integrated array and accelerating device based on static random access memory | |
CN114546335B (en) | Memory computing device for multi-bit input and multi-bit weight multiplication accumulation | |
CN112884140B (en) | Multi-bit memory internal computing unit, array and device | |
CN113782072B (en) | Multi-bit memory computing circuit | |
CN112558922A (en) | Four-transistor memory computing device based on separated word lines | |
CN114300012B (en) | Decoupling SRAM memory computing device | |
CN113823343B (en) | Separated computing device based on 6T-SRAM | |
CN113838504B (en) | Single-bit memory computing circuit based on ReRAM | |
CN112232502B (en) | Same or memory unit and memory array device | |
CN113077050B (en) | Digital domain computing circuit device for neural network processing | |
CN114895869B (en) | Multi-bit memory computing device with symbols | |
CN113986195B (en) | Delay type single-bit memory computing unit and device | |
CN116204490A (en) | 7T memory circuit and multiply-accumulate operation circuit based on low-voltage technology | |
CN113391786B (en) | Computing device for multi-bit positive and negative weights | |
CN113258910B (en) | Computing device based on pulse width modulation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210326 |