CN114093438A - Based on Bi2O2Se multi-mode library network time sequence information processing method - Google Patents
Based on Bi2O2Se multi-mode library network time sequence information processing method Download PDFInfo
- Publication number
- CN114093438A CN114093438A CN202111263213.XA CN202111263213A CN114093438A CN 114093438 A CN114093438 A CN 114093438A CN 202111263213 A CN202111263213 A CN 202111263213A CN 114093438 A CN114093438 A CN 114093438A
- Authority
- CN
- China
- Prior art keywords
- modal
- library network
- substrate
- library
- time sequence
- 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.)
- Granted
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 12
- 238000003672 processing method Methods 0.000 title claims abstract description 10
- 238000012549 training Methods 0.000 claims abstract description 18
- 239000000463 material Substances 0.000 claims abstract description 15
- 238000000034 method Methods 0.000 claims abstract description 14
- 230000003287 optical effect Effects 0.000 claims abstract description 14
- 238000001816 cooling Methods 0.000 claims abstract description 12
- 238000010438 heat treatment Methods 0.000 claims abstract description 10
- 238000013461 design Methods 0.000 claims abstract description 8
- 238000012545 processing Methods 0.000 claims abstract description 8
- 230000005669 field effect Effects 0.000 claims abstract description 7
- 239000000758 substrate Substances 0.000 claims description 24
- 108020004705 Codon Proteins 0.000 claims description 20
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 claims description 13
- 239000010703 silicon Substances 0.000 claims description 13
- 150000001413 amino acids Chemical class 0.000 claims description 11
- 229910052710 silicon Inorganic materials 0.000 claims description 11
- 230000008859 change Effects 0.000 claims description 7
- 239000002131 composite material Substances 0.000 claims description 6
- 239000002135 nanosheet Substances 0.000 claims description 5
- 238000001208 nuclear magnetic resonance pulse sequence Methods 0.000 claims description 5
- 108010076504 Protein Sorting Signals Proteins 0.000 claims description 4
- 108091026890 Coding region Proteins 0.000 claims description 2
- 229910000449 hafnium oxide Inorganic materials 0.000 claims description 2
- TWNQGVIAIRXVLR-UHFFFAOYSA-N oxo(oxoalumanyloxy)alumane Chemical compound O=[Al]O[Al]=O TWNQGVIAIRXVLR-UHFFFAOYSA-N 0.000 claims description 2
- 229910052814 silicon oxide Inorganic materials 0.000 claims description 2
- 238000005562 fading Methods 0.000 abstract description 2
- 229940024606 amino acid Drugs 0.000 description 8
- 239000004065 semiconductor Substances 0.000 description 7
- 238000009826 distribution Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 230000001965 increasing effect Effects 0.000 description 6
- 235000020776 essential amino acid Nutrition 0.000 description 5
- 239000003797 essential amino acid Substances 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000000638 stimulation Effects 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 230000000739 chaotic effect Effects 0.000 description 3
- 230000005611 electricity Effects 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000004936 stimulating effect Effects 0.000 description 3
- 230000007423 decrease Effects 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 238000010792 warming Methods 0.000 description 2
- AGPKZVBTJJNPAG-WHFBIAKZSA-N L-isoleucine Chemical compound CC[C@H](C)[C@H](N)C(O)=O AGPKZVBTJJNPAG-WHFBIAKZSA-N 0.000 description 1
- ROHFNLRQFUQHCH-YFKPBYRVSA-N L-leucine Chemical compound CC(C)C[C@H](N)C(O)=O ROHFNLRQFUQHCH-YFKPBYRVSA-N 0.000 description 1
- FFEARJCKVFRZRR-BYPYZUCNSA-N L-methionine Chemical compound CSCC[C@H](N)C(O)=O FFEARJCKVFRZRR-BYPYZUCNSA-N 0.000 description 1
- COLNVLDHVKWLRT-QMMMGPOBSA-N L-phenylalanine Chemical compound OC(=O)[C@@H](N)CC1=CC=CC=C1 COLNVLDHVKWLRT-QMMMGPOBSA-N 0.000 description 1
- QIVBCDIJIAJPQS-VIFPVBQESA-N L-tryptophane Chemical compound C1=CC=C2C(C[C@H](N)C(O)=O)=CNC2=C1 QIVBCDIJIAJPQS-VIFPVBQESA-N 0.000 description 1
- KZSNJWFQEVHDMF-BYPYZUCNSA-N L-valine Chemical compound CC(C)[C@H](N)C(O)=O KZSNJWFQEVHDMF-BYPYZUCNSA-N 0.000 description 1
- ROHFNLRQFUQHCH-UHFFFAOYSA-N Leucine Natural products CC(C)CC(N)C(O)=O ROHFNLRQFUQHCH-UHFFFAOYSA-N 0.000 description 1
- KDXKERNSBIXSRK-UHFFFAOYSA-N Lysine Natural products NCCCCC(N)C(O)=O KDXKERNSBIXSRK-UHFFFAOYSA-N 0.000 description 1
- 239000004472 Lysine Substances 0.000 description 1
- AYFVYJQAPQTCCC-UHFFFAOYSA-N Threonine Natural products CC(O)C(N)C(O)=O AYFVYJQAPQTCCC-UHFFFAOYSA-N 0.000 description 1
- 239000004473 Threonine Substances 0.000 description 1
- QIVBCDIJIAJPQS-UHFFFAOYSA-N Tryptophan Natural products C1=CC=C2C(CC(N)C(O)=O)=CNC2=C1 QIVBCDIJIAJPQS-UHFFFAOYSA-N 0.000 description 1
- KZSNJWFQEVHDMF-UHFFFAOYSA-N Valine Natural products CC(C)C(N)C(O)=O KZSNJWFQEVHDMF-UHFFFAOYSA-N 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- RKTYLMNFRDHKIL-UHFFFAOYSA-N copper;5,10,15,20-tetraphenylporphyrin-22,24-diide Chemical compound [Cu+2].C1=CC(C(=C2C=CC([N-]2)=C(C=2C=CC=CC=2)C=2C=CC(N=2)=C(C=2C=CC=CC=2)C2=CC=C3[N-]2)C=2C=CC=CC=2)=NC1=C3C1=CC=CC=C1 RKTYLMNFRDHKIL-UHFFFAOYSA-N 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000010894 electron beam technology Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 229960000310 isoleucine Drugs 0.000 description 1
- AGPKZVBTJJNPAG-UHFFFAOYSA-N isoleucine Natural products CCC(C)C(N)C(O)=O AGPKZVBTJJNPAG-UHFFFAOYSA-N 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 229930182817 methionine Natural products 0.000 description 1
- 239000010445 mica Substances 0.000 description 1
- 229910052618 mica group Inorganic materials 0.000 description 1
- 239000002055 nanoplate Substances 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- COLNVLDHVKWLRT-UHFFFAOYSA-N phenylalanine Natural products OC(=O)C(N)CC1=CC=CC=C1 COLNVLDHVKWLRT-UHFFFAOYSA-N 0.000 description 1
- 108090000623 proteins and genes Proteins 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 239000004474 valine Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C60/00—Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
-
- 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
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Testing Or Measuring Of Semiconductors Or The Like (AREA)
Abstract
The invention discloses a Bi-based material2O2A Se multi-mode library network timing information processing method. The method utilizes a layered two-dimensional material Bi with high electron mobility and stable properties2O2Se is used as an effective layer channel to prepare the multi-mode photo-thermal sensor with the back gate field effect transistor structure, the multi-mode library network design is realized according to the high-dimensional and nonlinear memory fading characteristics of the device to electric pulses, optical pulses, heating pulses and cooling pulses, and the multi-mode library network is used for processing time sequence information, so that the training cost is low, the efficiency is high, and the precision is high.
Description
Technical Field
The invention relates to a Bi based on a novel semiconductor2O2Multimodal library network sequential processing (multimodal Reservoir temporal and sequential processing based on new semiconductor Bi) of Se2O2Se) method, belonging to the field of novel materials and multimodal computing.
Background
The recognition and processing of information with time sequence characteristics is important for machine learning, such as speech recognition, gene sequencing, etc., and the recognition and processing of photoelectric information and temperature information with time sequence characteristics is also widely applied to robot vision and touch perception systems. The traditional calculation method for information identification and processing by using the recurrent neural network needs a large amount of training cost. Therefore, in the face of a large amount of sensing information with time sequence characteristics, a multi-modal information processing method with lower training cost and higher efficiency is needed.
Disclosure of Invention
In order to realize information processing of multi-mode time sequence information and reduce training cost, the invention provides a novel Bi-based method2O2Provided is a multimode library network timing information processing method of a Se semiconductor.
The invention provides a method for processing multi-modal library network time sequence information, which is established on the basis of a novel layered two-dimensional material Bi2O2Based on a Se multi-mode photo-thermal sensor. Layered two-dimensional material Bi2O2Se has ultrahigh electron mobility, good environmental stability and is very sensitive to optical signals and thermal signals. In the invention, Bi2O2Se is transferred from its growth substrate (typically a mica substrate) onto a silicon/high-k (Si/high-k) substrate to make back gate MOSFET structure devices by CMOS compatible processes. The device can realize high-dimensional and nonlinear memory characteristics of electric pulse, optical pulse, heating pulse and cooling pulse, thereby realizing multi-mode (electric, optical, heating and cooling) library network design, wherein the electric pulse signal and the heating pulse signal are usedThe signal exhibits a current suppressing effect, and exhibits a current enhancing effect for the optical pulse signal and the cooling pulse signal. In one embodiment of the invention, 21 base sequences of 8 essential amino acids in a human body are identified by a multi-modal library network design mode, and compared with a traditional electric pulse coding mode, the multi-modal photothermal signal coding efficiency is higher and the multi-modal photothermal signal coding accuracy is higher.
In particular, the invention provides a Bi-based material2O2The multi-mode photo-thermal sensor of Se is a layered two-dimensional material Bi2O2The back gate field effect transistor with Se as an effective layer channel comprises a substrate and a channel positioned on the substrate, wherein the two ends of the channel are respectively a source electrode and a drain electrode, the substrate is a silicon/high-k (Si/high-k) substrate, and the channel is Bi2O2Se nanosheet.
In the above multimode photothermal sensor, the silicon/high-k substrate is a composite substrate composed of a silicon substrate and a high-k dielectric layer thereon, and is generally obtained by growing a dielectric layer having a high k value on the silicon substrate, such as a silicon/aluminum oxide composite substrate or a silicon/hafnium oxide composite substrate; the Bi2O2The thickness of the Se nanosheet is preferably 10-20 nm.
Adding Bi2O2Se nanoplates are transferred from their growth substrate onto a silicon/high-k substrate by electron beam Exposure (EBL) on Bi2O2Defining a source drain region at two ends of the Se nano sheet, depositing metal in the source drain region and stripping to form a source electrode and a drain electrode, thus obtaining the Bi-based semiconductor device2O2Multimodal photothermal sensor of Se.
Bi2O2Se has a nonlinear suppression effect on the current when stimulating electric pulses, because electrons generated by the autodoping effect are excited to Se vacancy and bound by the Se vacancy under the large Vd, so that the number of channel electrons is reduced, the current is reduced, unbalanced electrons are generated when stimulating the optical pulses due to the grating effect and the radiant heat effect, the electron concentration is increased, the current is increased, the electron mobility is mainly influenced when stimulating the temperature pulses, the lattice scattering is increased by the warming pulse, the electron mobility is reduced, and the current is reducedThe cooling pulse reduces lattice scattering, increases carrier mobility, and increases current.
Based on the multi-modal photo-thermal sensor, the multi-modal library network time sequence information processing method provided by the invention comprises the following steps:
1) by using Bi based on layered two-dimensional materials2O2The Se multi-mode photo-thermal sensor selects two, three or four pulse signals to design a multi-mode library network according to the memory characteristics of the electric pulse signals, the optical pulse signals, the heating pulse signals and the cooling pulse signals;
2) coding the time sequence information, and then applying two, three or four corresponding pulse signal sequences to the multi-modal photothermal sensor according to the coding sequence to obtain a corresponding electrical conduction state;
3) assigning the electrical state obtained in the step 2) to the library node of the multi-modal library network designed in the step 1), and training the multi-modal library network;
4) and 3) identifying the time sequence information to be detected by utilizing the multi-modal library network trained in the step 3).
And 2) training the multi-mode library network through the known time sequence information, calculating the weight change and the accuracy rate, continuously updating the weight until the accuracy rate reaches a maximum value and does not change any more, and finishing the training.
The multi-modal library network time sequence information processing method can realize the coding and identification of the amino acid codon base sequence, one base is coded by two-bit binary in the step 2), namely, the base is represented by four bases respectively by '00', '01', '10' and '11', and two, three or four pulse signal sequences corresponding to the amino acid codons are applied to the multi-modal photothermal sensor to obtain corresponding electric conduction states; assigning the obtained electrical state to a library node of the multi-modal library network for training in the step 3); and 4) applying a pulse sequence corresponding to the base sequence to be detected to the multi-modal photothermal sensor, and inputting the obtained corresponding electrical state into the trained multi-modal library network, so that the amino acid coded by the base sequence to be detected can be identified.
In one embodiment of the invention, one base is encoded in two binary digits, respectively "00" for base G, "01" for base U, "10" for base A, and "11" for base C. Then, the base sequences of 21 codons of 8 essential amino acids of the human body are coded in four modes of electricity, light, temperature rise and temperature drop, and the sequence information is identified by combining a library network.
The technical advantages of the invention are mainly reflected in that:
1) novel layered Bi based on high mobility and stable properties2O2The Se semiconductor is used as an effective layer of a device to prepare the multi-mode photo-thermal sensor with the back gate field effect transistor structure.
2) The device can present high-dimensionality and nonlinear memory attenuation characteristics to four stimuli of electricity, light, temperature rise and temperature drop, and can realize multi-modal library network design.
3) The coding operation of the base sequence of the 21 codons of the 8 essential amino acids of the human body can be realized by a multi-modal library network design mode.
4) The manufacturing process of the invention is completely compatible with the existing CMOS process, and can realize large-scale integration.
Drawings
FIG. 1 shows a novel layered material Bi used in the present invention2O2Schematic diagram of Se three-dimensional stereo lattice structure.
FIGS. 2 to 5 are based on Bi, respectively2O2And the current characteristics of the back gate field effect transistor device made of the Se material under the stimulation of electric, optical, temperature-raising and temperature-lowering pulses.
FIGS. 6 to 9 show the results of encoding the base sequences of 21 codons for 8 essential amino acids in human body by electric, optical, temperature-raising and temperature-lowering pulse methods, respectively.
FIG. 10 is a schematic diagram of a library network designed by three ways of light, temperature rise and temperature drop for the base sequences of 21 codons of 8 amino acids and a schematic diagram of a library network calculation process in the example.
FIG. 11 is a weight distribution diagram in the example after training of a library network for single electric coding and multi-modal light, temperature-raising and temperature-lowering coding for the base sequences of 21 codons of 8 amino acids, wherein (a) is the library network for single electric coding and (b) is the library network for multi-modal light, temperature-raising and temperature-lowering coding.
Fig. 12 is a chaotic matrix corresponding to the weight distribution shown in fig. 11, in which (a) is a single electrically encoded library network, and (b) is a multi-modal light, temperature-increasing, and temperature-decreasing encoded library network.
FIG. 13 shows the recognition accuracy of library network training of base sequences of 21 codons for 8 amino acids in examples, in which (a) is a single electrically encoded library network and (b) is a multi-modal light, temperature-raising, and temperature-lowering encoded library network.
Detailed Description
The invention provides a Bi-based novel layered semiconductor material2O2Method for realizing multi-mode library network timing information processing by Se back gate field effect transistor, which transfers Bi with high mobility and stable property2O2Se acts as an active layer channel. According to Bi2O2The Se device realizes the multi-mode library network design for the high-dimensional and nonlinear memory fading characteristics of electric pulses, optical pulses, heating pulses and cooling pulses, and encodes and identifies the base sequences of 21 codons of 8 essential amino acids of a human body. The present invention will be described in detail with reference to the accompanying drawings.
Novel layered semiconductor material Bi2O2Se belongs to a tetragonal lattice structure, and Se layers and BiO layers are alternately arranged with each other as shown in FIG. 1.
With Bi2O2Se material is used for preparing back gate field effect transistor for channel, and a large Vd (is applied to the device)>3V), the current of the test device exhibits a non-linear decreasing trend at large Vd voltage pulses, as shown in fig. 2. This is because electrons are excited to Se vacancies, and the vacancies trap electrons, resulting in a decrease in channel electron concentration.
Because of Bi2O2Se is very sensitive to optical signals, so that when a periodically switched optical pulse signal is applied, the current of the test device shows a non-linear increasing trend under the stimulation of the optical pulse, as shown in fig. 3. This is becauseBi2O2The Se has a grating effect and a radiation heat effect, and generated photogenerated holes and hot holes are bound by a main state, so that the electron concentration of a channel is increased.
Bi2O2Se is also very sensitive to temperature change, a thermoelectric sheet is used for realizing a heating pulse signal, and the current of a test device shows a nonlinear reduction trend under the stimulation of the heating pulse, as shown in figure 4. This is because increasing the temperature increases lattice scattering, which decreases electron mobility.
Similarly, we use the thermoelectric chip to implement the cooling pulse signal, and the current of the test device shows a non-linear decreasing trend under the stimulation of the cooling pulse, as shown in fig. 5. This is because lowering the temperature reduces lattice scattering and increases carrier mobility.
Furthermore, we encode the base sequence of 21 codons of 8 amino acids in human body in four ways (electricity, light, temperature rise and temperature drop), each codon is composed of 3 bases arranged in a certain sequence, we encode one base by two-bit binary, namely "00" for base G, "01" for base U, "10" for base A, and "11" for base C. FIG. 6 is the conductance values encoded by electric pulses, FIG. 7 is the conductance values encoded by light pulses, FIG. 8 is the conductance values encoded by warming pulses, and FIG. 9 is the conductance values encoded by cooling pulses, in each of which (a) shows the encoding of valine codons, (b) shows the encoding of isoleucine codons, (c) shows the encoding of leucine codons, (d) shows the encoding of lysine and tryptophan codons, (e) shows the encoding of phenylalanine and methionine codons, and (f) shows the encoding of threonine codons.
Library network calculation is performed through two modes of single electrical coding and multi-mode light, temperature rise and temperature fall, and the whole calculation flow is shown in fig. 10.
1) Applying pulse sequences corresponding to different amino acid codons to the device, namely applying a single electric pulse sequence, applying multi-mode light, heating and cooling pulse sequences, and then respectively obtaining corresponding electric conduction states, wherein the electric conduction state obtained by the multi-mode pulse signal is 3 times that obtained by the single electric pulse signal;
2) the electrical conducting states are assigned to the library nodes of the library network by pychar software by using python language, so that the number of the library nodes contained in the multi-modal library network is three times of the number of the library nodes in the single electrical operation library network;
3) training is carried out through the software, weight change is calculated, accuracy is further calculated, weight updating is continuously carried out until the accuracy reaches a maximum value and does not change any more, training is completed, and final weight distribution, chaotic matrix data and accuracy data are output.
The weight distribution after training is shown in fig. 11, where (a) is the weight distribution of a single electrical code library network, and (b) is the weight distribution of a multi-modal light, temperature-raising, and temperature-lowering code library network. Fig. 12 shows the corresponding chaotic matrix, in which (a) is single electrical coding and (b) is multi-modal coding. FIG. 13 is a comparison of efficiency and accuracy between the two modes, wherein the multi-modal library network recognition accuracy can reach 100% after 30 times of training, and the single-modal electrical code library network recognition accuracy is still maintained at 94.3% after 500 times of training.
Claims (6)
1. A multi-modal library network timing information processing method comprises the following steps:
1) by using Bi based on layered two-dimensional materials2O2The Se multi-mode photo-thermal sensor selects two, three or four pulse signals to design a multi-mode library network according to the memory characteristics of the electric pulse signals, the optical pulse signals, the heating pulse signals and the cooling pulse signals;
2) coding the time sequence information, and then applying two, three or four corresponding pulse signal sequences to the multi-modal photothermal sensor according to the coding sequence to obtain a corresponding electrical conduction state;
3) assigning the electrical state obtained in the step 2) to the library node of the multi-modal library network designed in the step 1), and training the multi-modal library network;
4) and 3) identifying the time sequence information to be detected by utilizing the multi-modal library network trained in the step 3).
2. The method of claim 1, wherein the multi-modal library network timing information processing method is characterized in that the multi-modal photothermal sensor is a layered two-dimensional material Bi2O2The back gate field effect transistor with Se as an effective layer channel comprises a substrate and a channel positioned on the substrate, wherein the two ends of the channel are respectively a source electrode and a drain electrode, the substrate is a silicon/high-k substrate, and the channel is Bi2O2Se nanosheet.
3. The method of claim 2, wherein the silicon/high-k substrate is a composite substrate comprising a silicon substrate and a high-k dielectric layer thereon, and is selected from a silicon/aluminum oxide composite substrate and a silicon/hafnium oxide composite substrate.
4. The multi-modal library network timing information processing method of claim 2, wherein the Bi is2O2The thickness of the Se nanosheet is 10-20 nm.
5. The method for processing the time sequence information of the multi-modal library network as claimed in claim 1, wherein the steps 2) to 3) train the multi-modal library network through the known time sequence information, calculate the weight change and the accuracy rate, and continuously update the weight until the accuracy rate reaches a maximum value and does not change any more, and the training is completed.
6. The method of claim 1, wherein the base sequence of the amino acid codons is encoded and identified by the method, one base is binary-encoded in two bits in step 2), i.e., "00", "01", "10" and "11" represent four bases, respectively, and two, three or four pulse signal sequences corresponding to the amino acid codons are applied to the multi-modal photothermal sensor to obtain corresponding electrical conduction states; assigning the obtained electrical state to a library node of the multi-modal library network for training in the step 3); and 4) applying a pulse sequence corresponding to the base sequence to be detected to the multi-modal photothermal sensor, and inputting the obtained corresponding electrical state into the trained multi-modal library network, so that the amino acid coded by the base sequence to be detected can be identified.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111263213.XA CN114093438B (en) | 2021-10-28 | 2021-10-28 | Bi-based2O2Se multi-mode library network time sequence information processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111263213.XA CN114093438B (en) | 2021-10-28 | 2021-10-28 | Bi-based2O2Se multi-mode library network time sequence information processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114093438A true CN114093438A (en) | 2022-02-25 |
| CN114093438B CN114093438B (en) | 2024-09-24 |
Family
ID=80298030
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111263213.XA Active CN114093438B (en) | 2021-10-28 | 2021-10-28 | Bi-based2O2Se multi-mode library network time sequence information processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114093438B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117743927A (en) * | 2023-12-20 | 2024-03-22 | 北京大学 | Time sequence signal analysis method based on multi-mode sensing input library network |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106011783A (en) * | 2016-07-07 | 2016-10-12 | 北京大学 | High-mobility-ratio lamellar Bi2O2Se semiconductor film and preparation method thereof |
| CN109402739A (en) * | 2018-12-19 | 2019-03-01 | 清华-伯克利深圳学院筹备办公室 | A kind of two dimension bismuth oxygen selenium atom crystalline material, and its preparation method and application |
| WO2019048965A1 (en) * | 2017-09-06 | 2019-03-14 | 株式会社半導体エネルギー研究所 | Physical property prediction method and physical property prediction system |
| CN110438567A (en) * | 2019-07-22 | 2019-11-12 | 中国科学院上海微系统与信息技术研究所 | A kind of preparation method of semiconductor selenium bismuth oxide single crystal thin-film material |
| WO2021038362A1 (en) * | 2019-08-29 | 2021-03-04 | 株式会社半導体エネルギー研究所 | Property prediction system |
| CN112588303A (en) * | 2020-11-23 | 2021-04-02 | 安徽大学 | Preparation method of selenium-bismuth oxide nanosheet and heterojunction type photoelectrode based on preparation method |
-
2021
- 2021-10-28 CN CN202111263213.XA patent/CN114093438B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106011783A (en) * | 2016-07-07 | 2016-10-12 | 北京大学 | High-mobility-ratio lamellar Bi2O2Se semiconductor film and preparation method thereof |
| WO2019048965A1 (en) * | 2017-09-06 | 2019-03-14 | 株式会社半導体エネルギー研究所 | Physical property prediction method and physical property prediction system |
| CN109402739A (en) * | 2018-12-19 | 2019-03-01 | 清华-伯克利深圳学院筹备办公室 | A kind of two dimension bismuth oxygen selenium atom crystalline material, and its preparation method and application |
| CN110438567A (en) * | 2019-07-22 | 2019-11-12 | 中国科学院上海微系统与信息技术研究所 | A kind of preparation method of semiconductor selenium bismuth oxide single crystal thin-film material |
| WO2021038362A1 (en) * | 2019-08-29 | 2021-03-04 | 株式会社半導体エネルギー研究所 | Property prediction system |
| CN112588303A (en) * | 2020-11-23 | 2021-04-02 | 安徽大学 | Preparation method of selenium-bismuth oxide nanosheet and heterojunction type photoelectrode based on preparation method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117743927A (en) * | 2023-12-20 | 2024-03-22 | 北京大学 | Time sequence signal analysis method based on multi-mode sensing input library network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114093438B (en) | 2024-09-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lei et al. | High‐performance memristor based on 2d layered bioi nanosheet for low‐power artificial optoelectronic synapses | |
| Zhu et al. | Optogenetics-inspired tunable synaptic functions in memristors | |
| Larkin et al. | Phase transition in uniaxial ferroelectrics | |
| CN114093438A (en) | Based on Bi2O2Se multi-mode library network time sequence information processing method | |
| US20060148254A1 (en) | Activated iridium oxide electrodes and methods for their fabrication | |
| Huang et al. | Multifunctional optoelectronic synapses based on arrayed MoS2 monolayers emulating human association memory | |
| TW202042117A (en) | Neural network classifier using array of stacked gate non-volatile memory cells | |
| Zhu et al. | Hierarchies in visual pathway: functions and inspired artificial vision | |
| Wang et al. | Enhanced multiwavelength response of flexible synaptic transistors for human sunburned skin simulation and neuromorphic computation | |
| Li et al. | Organic optoelectronic synaptic devices for energy-efficient neuromorphic computing | |
| CN112380768A (en) | BP neural network-based LED chip life prediction method | |
| Fang et al. | Multimodal In‐Sensor Computing Implemented by Easily‐Fabricated Oxide‐Heterojunction Optoelectronic Synapses | |
| Li et al. | Photonic synapses with ultralow energy consumption for artificial visual perception and brain storage | |
| CN109545960A (en) | A kind of memristor and its preparation method and application of conductance continuous variable | |
| Sun et al. | Bio-inspired vision and neuromorphic image processing using printable metal oxide photonic synapses | |
| CN111950720A (en) | Novel brain-like vision system | |
| Yuan et al. | Silicon carbide nanowire-based multifunctional and efficient visual synaptic devices for wireless transmission and neural network computing | |
| Saleem et al. | An opto-electronic HfO x-based transparent memristive synapse for neuromorphic computing system | |
| Jiang et al. | Bioinspired In‐Sensor Reservoir Computing for Self‐Adaptive Visual Recognition with Two‐Dimensional Dual‐Mode Phototransistors | |
| Hu et al. | Transmembrane voltage analyses in spheroidal cells in response to an intense ultrashort electrical pulse | |
| Fang et al. | Two-terminal photoelectric dual modulation synaptic devices for face recognition | |
| CN111081875A (en) | Ferroelectric polarization regulated artificial synapse device and preparation method thereof | |
| Wang et al. | Ultrathin flexible inorganic device for long-term monitoring of light and temperature | |
| Kim et al. | Multi‐Level Long‐Term Memory Resembling Human Memory Based on Photosensitive Field‐Effect Transistors with Stable Interfacial Deep Traps | |
| CN114141813A (en) | Based on Bi2O2Se multi-modal feeling storage and calculation integrated system and implementation method thereof |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |