CN111130557B - Data reconstruction method based on distributed quasi-Newton projection tracking - Google Patents
Data reconstruction method based on distributed quasi-Newton projection tracking Download PDFInfo
- Publication number
- CN111130557B CN111130557B CN201911409497.1A CN201911409497A CN111130557B CN 111130557 B CN111130557 B CN 111130557B CN 201911409497 A CN201911409497 A CN 201911409497A CN 111130557 B CN111130557 B CN 111130557B
- Authority
- CN
- China
- Prior art keywords
- data
- support set
- reconstructed
- representing
- computing node
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 239000011159 matrix material Substances 0.000 claims abstract description 27
- 238000005259 measurement Methods 0.000 claims abstract description 22
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000011084 recovery Methods 0.000 abstract description 2
- 238000004422 calculation algorithm Methods 0.000 description 7
- 238000010276 construction Methods 0.000 description 2
- 239000002689 soil Substances 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
- H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
The application discloses a data reconstruction method based on distributed quasi-Newton projection tracking, which comprises the following steps: dividing data to be reconstructed into a public part and an individual part; each computing node sends a support set of data to be reconstructed to a neighbor computing node; each computing node obtains a common part support set according to the support set calculated by the computing node and the support set obtained from the neighbor computing node; each computing node iteratively reconstructs compressed data according to the common part support set, the collected compressed data, the measurement matrix and sparsity of the common part and the individual part. According to the application, in the data reconstruction process, each computing node can obtain the data in the whole network, acquire global information, and can be applied to the recovery of the data in the distributed network with public components, so that the requirements of more scenes are met; and the data reconstruction speed is higher, and the accuracy is higher.
Description
Technical Field
The application relates to a data reconstruction method based on distributed quasi-Newton projection tracking, and belongs to the technical field of data reconstruction.
Background
With the continuous development of information technology, the scale of people to acquire data is getting larger and larger. In many distributed applications, it is important how the reconstruction of data can be achieved quickly and efficiently. The data to be processed in the distributed system are distributed in a plurality of nodes, and the data are required to be transmitted to a server for joint reconstruction of the data. Therefore, a large amount of information needs to be transferred in the distributed network upon data reconstruction, thereby causing consumption of bandwidth and delay of data reconstruction.
At present, the data reconstruction in the distributed system mainly adopts a centralized processing method, all the node data are transmitted to a server for centralized processing, a large amount of bandwidth is required to be consumed, a long time delay is generated, and the requirements of the increasingly large-scale distributed system on the rapid and accurate reconstruction of the compressed data cannot be met. The existing distributed bayesian algorithm performs iterative reconstruction by decomposing data into a public part and an individual part and utilizing variational Bayesian inference, but because only the data information of the public part is interacted among all the computing nodes, all the computing nodes cannot acquire global information, certain application scenes (for example, in intelligent traffic, all the computing nodes are required to know global information for adjustment) cannot be satisfied, and the speed and the accuracy of data reconstruction still need to be improved.
Disclosure of Invention
The application aims to provide a data reconstruction method based on distributed quasi-Newton projection tracking, which can effectively solve the problems existing in the prior art, realize faster and more accurate data reconstruction, acquire global information and meet the requirements of more scenes.
In order to solve the technical problems, the application adopts the following technical scheme: the data reconstruction method based on the distributed quasi-Newton projection tracking comprises the following steps: dividing data to be reconstructed into a public part and an individual part; each computing node sends a support set of data to be reconstructed to a neighbor computing node; each computing node obtains a common part support set according to the support set calculated by the computing node and the support set obtained from the neighbor computing node; each computing node iteratively reconstructs compressed data according to the common part support set, the collected compressed data, the measurement matrix and sparsity of the common part and the individual part.
Preferably, the method specifically comprises the following steps:
s1, dividing data to be reconstructed into a public part and an individual part; initializing a support set and a data residual error of data to be reconstructed of each computing node;
s2, each computing node sends the obtained latest support set of the data to be reconstructed to the neighbor computing nodes;
s3, each computing node obtains a public part support set according to the latest support set calculated by the computing node and the latest support set obtained from the neighbor computing node;
s4, each computing node obtains an updated support set of data to be reconstructed, updated reconstructed sparse data and updated data residual errors according to the common part support set, the collected compressed data, the measurement matrix and the sparsity of the common part and the individual part by using a MODENPP function;
s5, determining whether the square of the second norm of the updated data residual is smaller than the square of the second norm of the last obtained data residual? If yes, go to S2; otherwise, the reconstructed sparse data obtained last time is output as final reconstructed data.
The data reconstruction is carried out by the method, particularly by combining with the MODENPP function, so that the number of data iterations in the reconstruction process is smaller, the reconstruction speed is faster, the reconstruction precision is higher, and each computing node can acquire the global information of the network.
More preferably, in step S3, the computing node selects the previous K having the highest occurrence frequency according to the latest support set calculated by itself and the latest support set obtained from the neighboring computing nodes (c) Each being a common partial support set, where K (c) Is the sparsity of the common portion. The method acquires the public part support set, so that the calculation is simpler under the condition of being accurate as much as possible, and the speed of data reconstruction is improved.
In the aforementioned data reconstruction method based on distributed quasi-newton projection tracking, step S4 specifically includes the following steps: s41, initializing data:if->R is then 0 ＝resid(y,Ax 0 ) Otherwise->k＝0，x 0 ＝0，/>Wherein (1)>Sparsity K representing common part (c) And sparsity of individual parts->And (3) summing; t (T) 0 Representing the support set->Support set representing common parts in the whole network, < ->Representing the measurement matrix A based on the support set T 0 Y represents the collected compressed data, x 0 Sparse signals representing the initialization reconstruction, resid represents the computation of y and +.>Is a difference in (2);
s42, let k=k+1, calculateWherein d k Indicating Newton direction, x k-1 Representing the reconstructed sparse signal obtained in the k-1 th iteration, I representing the identity matrix, Λ k Representing the front +.>Index value of maximum value and taking corresponding calculation result based on lambda k Is a projection of (2);
s43, calculatingWherein mu k Represents step length, k represents iteration number, T k-1 Representing the support set calculated in the last iteration, < +.>Representing the measurement matrix A based on T k-1 Is a projection of (2);
s44, calculating a support set T for obtaining updated data to be reconstructed k ：Wherein T is k Representing the support set obtained in the kth iteration, x k-1 Representing the k-1 st iteration acquisitionIs represented by max_indices, which represents x k-1 +μ k d k Before->Index values of the maximum values;
s45, calculating to obtain updated data x to be reconstructed k ：Wherein T is k Representing the support set calculated for the kth iteration, < >>Representing the measurement matrix A based on T k Is a projection of (2);
s46, calculating to obtain updated data residual error r k ：
S47, judgingIs it true? If yes, go to S42, otherwise directly output T obtained in this iteration k 、x k And r k 。
T is obtained by the above method k 、x k And r k The method has the advantages of simple calculation, higher reconstruction accuracy and higher reconstruction speed, and particularly the method for reconstructing d in the application k The updating method of the application ensures that the data reconstruction accuracy is higher and the reconstruction speed is faster.
In the foregoing data reconstruction method based on distributed quasi-newton projection tracking, in step S1, a support set and a data residual of data to be reconstructed of each computing node are initialized by the following methods: initializing sparsity of a public part and an individual part, and obtaining a support set and a data residual error of data to be reconstructed according to the collected compressed data, a measurement matrix and the sparsity of the public part and the individual part by using an MODENPP function; wherein the common part support set is set to be empty.
In the above data reconstruction method based on distributed quasi-newton projection tracking, in step S2, the iterative reconstruction of the compressed data, i.e. the construction of the objective function, makes it converge:
wherein y is the collected compressed data, A is the measurement matrix, and x is the sparse signal to be reconstructed.
Compared with the prior art, the method and the device have the advantages that the supporting set calculated by each calculation node in each iteration is exchanged with other calculation nodes, so that the common sparse part supporting set can be obtained, and the common sparse part supporting set is used as data of the next iteration to be subjected to iterative calculation until the data reach convergence. According to the application, in the data reconstruction process, each computing node can obtain the data in the whole network, acquire global information, and can be applied to the recovery of the data in the distributed network with public components, so that the requirements of more scenes are met; and the data reconstruction speed is higher, and the accuracy is higher.
In addition, the inventor randomly selects K index values not greater than n in the simulation data (i.e. the measurement matrix A is a randomly generated m×n matrix, and randomly generates data at the corresponding index values as the sparse signal x to be reconstructed i By randomly generated signals x i Calculating y i ＝A i x i +e i Wherein e is i Is gaussian noise). At a calculated node number of 5, sparsenessm=120, n=300, support set +.>Support set of common parts in the whole network +.>k＝0，x 0 =0, data residual r 0 ＝y i Under the condition of (1), the data reconstruction error by using the method is 5.5965e-4, compared with the reconstruction error of 6.7254e-4 of the existing distributed compressed data reconstruction algorithm (namely the distributed Bayesian algorithm in the background art), the reconstruction error is improved by 16.8%, and the reconstruction speed is improved by 21.3%; under the condition that real data (a data set of the soil temperature of the upstream ecological hydrological wireless sensor network in the 2012 black river basin (a black river ecological hydrological remote sensing test: the water ecological hydrological wireless sensor network BNET soil temperature and humidity observation data set http:// westdc.westgis.ac. cn/heihe/view/uuid/0a2e1ce6-f322-4d0f-82ee-70446123dba 1)) are adopted, when the data set is applied to the calculation node number 35, m=60, n=128, the data reconstruction error is 0.0083 by using the method disclosed by the application, and compared with the reconstruction error of the existing distributed compressed data reconstruction algorithm (namely, the distributed Bayesian algorithm mentioned in the background art), the reconstruction error of the distributed compressed data reconstruction algorithm is 0.0095, which is designed, the reconstruction speed is improved by 12.6%, and the reconstruction speed is improved by 15.4%.
Drawings
FIG. 1 is a method flow diagram of one embodiment of the present application;
FIG. 2 is a flow chart of a method of obtaining a support set of updated data to be reconstructed, updated reconstructed sparse data, and updated data residuals.
In order that the above-recited objects, features and advantages of the present application will be more clearly understood, a more particular description of the application will be rendered by reference to the appended drawings and appended detailed description. It should be noted that, without conflict, the embodiments of the present application and features in the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, however, the present application may be practiced in other ways than those described herein, and therefore the scope of the present application is not limited to the specific embodiments disclosed below.
Detailed Description
Embodiments of the application: the data reconstruction method based on the distributed quasi-Newton projection tracking, as shown in fig. 1, comprises the following steps: dividing data to be reconstructed into a public part and an individual part; each computing node sends a support set of data to be reconstructed to a neighbor computing node; each computing node obtains a common part support set according to the support set calculated by the computing node and the support set obtained from the neighbor computing node; each computing node iteratively reconstructs compressed data according to the common part support set, the collected compressed data, the measurement matrix and sparsity of the common part and the individual part.
In order to further improve the accuracy and speed of data reconstruction, the method comprises the following steps:
s1, dividing data to be reconstructed into a public part and an individual part; initializing a support set and a data residual error of data to be reconstructed of each computing node;
s2, each computing node sends the obtained latest support set of the data to be reconstructed to the neighbor computing nodes;
s3, each computing node obtains a public part support set according to the latest support set calculated by the computing node and the latest support set obtained from the neighbor computing node;
s4, each computing node obtains an updated support set of data to be reconstructed, updated reconstructed sparse data and updated data residual errors according to the common part support set, the collected compressed data, the measurement matrix and the sparsity of the common part and the individual part by using a MODENPP function;
s5, determining whether the square of the second norm of the updated data residual is smaller than the square of the second norm of the last obtained data residual? If yes, go to S2; otherwise, the reconstructed sparse data obtained last time is output as final reconstructed data.
In step S4, a subspace tracking (SP), a compressed sampling matching pursuit (CoSaMP) algorithm may be further utilized to obtain a support set of updated data to be reconstructed, updated reconstructed sparse data, and updated data residuals.
Optionally, in the present application, in step S3, the computing node calculates the latest according to itselfSupport set and latest support set obtained from neighbor computing node, and the top K with highest occurrence frequency is selected (c) Each as a common partial support set (i.eWherein (1)>Representing a supporting set of common parts in the whole network,representing a set of other computing nodes communicable with the current computing node, +.>Representing a set of support sets of all data calculated by the current computing node and other computing nodes communicable with the current computing node), wherein K (c) Is the sparsity of the common portion.
Optionally, as shown in fig. 2, step S4 specifically includes the following steps:
s41, initializing data:if->R is then 0 ＝resid(y,Ax 0 ) Otherwise->k＝0，x 0 ＝0，/>Wherein (1)>Sparsity K representing common part (c) And sparsity of individual parts->And (3) summing; t (T) 0 Representing the support set->Support set representing common parts in the whole network, < ->Representing the measurement matrix A based on the support set T 0 Y represents the collected compressed data, x 0 Sparse signals representing the initialization reconstruction, resid represents the computation of y and +.>Is a difference in (2);
s42, let k=k+1, calculateWherein d k Indicating Newton direction, x k-1 Representing the reconstructed sparse signal obtained in the k-1 th iteration, I representing the identity matrix, Λ k Representing the front +.>Index value of maximum value and taking corresponding calculation result based on lambda k Is a projection of (2);
s43, calculatingWherein mu k Represents step length, k represents iteration number, T k-1 Representing the support set calculated in the last iteration, < +.>Representing the measurement matrix A based on T k-1 Is a projection of (2);
s44, calculating a support set T for obtaining updated data to be reconstructed k ：Wherein T is k Representing the support set obtained in the kth iteration, x k-1 Representing the reconstructed sparse signal obtained in the k-1 th iteration, max_indices representing x k-1 +μ k d k Before->Index values of the maximum values;
s45, calculating to obtain updated data x to be reconstructed k ：Wherein T is k Representing the support set calculated for the kth iteration, < >>Representing the measurement matrix A based on T k Is a projection of (2);
s46, calculating to obtain updated data residual error r k ：
S47, judgingIs it true? If yes, go to S42, otherwise directly output T obtained in this iteration k 、x k And r k 。
Optionally, in step S1, the support set of data to be reconstructed and the data residuals of each computing node are initialized by: initializing sparsity of a public part and an individual part, and obtaining a support set and a data residual error of data to be reconstructed according to the collected compressed data, a measurement matrix and the sparsity of the public part and the individual part by using an MODENPP function; wherein the common part support set is set to be empty.
Optionally, in step S2, the iterative reconstruction of the compressed data, i.e. the construction of the objective function, makes it converge:
where y is the collected compressed data, A is the measurement matrix, which is an mxn matrix, and x is the sparse signal to be reconstructed.
The distributed system in the application is composed of N computing nodes, wherein each computing node manages p member nodes, each computing node contains data in each member node, and q is the sum of all data contained in the node.
Claims (1)
1. The data reconstruction method based on the distributed quasi-Newton projection tracking is characterized by comprising the following steps of:
s1, dividing data to be reconstructed into a public part and an individual part; initializing a support set and a data residual error of data to be reconstructed of each computing node;
s2, each computing node sends the obtained latest support set of the data to be reconstructed to the neighbor computing nodes;
s3, each computing node obtains a public part support set according to the latest support set calculated by the computing node and the latest support set obtained from the neighbor computing node;
s4, each computing node obtains an updated support set of data to be reconstructed, updated reconstructed sparse data and updated data residual errors according to the common part support set, the collected compressed data, the measurement matrix and the sparsity of the common part and the individual part by using a MODENPP function; the implementation of the MODENPP function, namely the specific steps of S4, are as follows:
s41, initializing data:if->R is then 0 ＝resid(y,Ax 0 ) Otherwise->k＝0，x 0 ＝0，/>Wherein (1)>Sparsity K representing common part (c) And sparsity of individual parts->And (3) summing; t (T) 0 Representing the support set->Support set representing common parts in the whole network, < ->Representing the measurement matrix A based on the support set T 0 Y represents the collected compressed data, x 0 Sparse signals representing the initialization reconstruction, resid represents the computation of y and +.>Is a difference in (2);
s42, let k=k+1, calculateWherein d k Indicating Newton direction, x k-1 Representing the reconstructed sparse signal obtained in the k-1 th iteration, I representing the identity matrix, Λ k Representing the front +.>Index value of maximum value and taking corresponding calculation result based on lambda k Is a projection of (2);
s43, calculatingWherein mu k Represents step length, k represents iteration number, T k-1 Representing the support set calculated in the last iteration, < +.>Representing the measurement matrix A based on T k-1 Is a projection of (2);
s44, calculating a support set T for obtaining updated data to be reconstructed k ：Wherein T is k Representing the support set obtained in the kth iteration, x k-1 Representing the reconstructed sparse signal obtained in the k-1 th iteration, max_indices representing x k-1 +μ k d k Before->Index values of the maximum values;
s45, calculating to obtain updated data x to be reconstructed k ：Wherein T is k Representing the support set calculated for the kth iteration, < >>Representing the measurement matrix A based on T k Is a projection of (2);
s46, calculating to obtain updated data residual error r k ：
S47, judgingIs it true? If yes, go to S42, otherwise directly output the iteration to obtainT of (2) k 、x k And r k ；
S5, judging whether the square of the two norms of the updated data residual is smaller than the square of the two norms of the data residual obtained last time; if yes, go to S2; otherwise, the reconstructed sparse data obtained last time is used as final reconstructed data to be output;
in step S3, the computing node selects the front K with highest occurrence frequency according to the latest support set calculated by the computing node and the latest support set obtained from the neighbor computing node (c) Each being a common partial support set, where K (c) Sparsity for common parts;
in step S1, the support set of data to be reconstructed and the data residuals of each computing node are initialized by: initializing sparsity of a public part and an individual part, and obtaining a support set and a data residual error of data to be reconstructed according to the collected compressed data, a measurement matrix and the sparsity of the public part and the individual part by using an MODENPP function; wherein the common part support set is set to be empty;
in step S2, the iterative reconstruction compresses the data, i.e. constructs the objective function such that it converges:
wherein y is the collected compressed data, A is the measurement matrix, and x is the sparse signal to be reconstructed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911409497.1A CN111130557B (en) | 2019-12-31 | 2019-12-31 | Data reconstruction method based on distributed quasi-Newton projection tracking |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911409497.1A CN111130557B (en) | 2019-12-31 | 2019-12-31 | Data reconstruction method based on distributed quasi-Newton projection tracking |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111130557A CN111130557A (en) | 2020-05-08 |
| CN111130557B true CN111130557B (en) | 2023-11-17 |
Family
ID=70506801
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911409497.1A Active CN111130557B (en) | 2019-12-31 | 2019-12-31 | Data reconstruction method based on distributed quasi-Newton projection tracking |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111130557B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101895297A (en) * | 2010-07-30 | 2010-11-24 | 哈尔滨工业大学 | Compressed sensing-oriented block-sparse signal reconfiguring method |
| CN102611455A (en) * | 2012-03-05 | 2012-07-25 | 哈尔滨工业大学 | Compressed sensing-oriented sparse multiband signal reconstruction method |
| CN103300859A (en) * | 2013-05-31 | 2013-09-18 | 王勇 | High-quality rapid CS-MRI (Compressed Sensing-Magnetic Resonance Imaging) method of mixed norm |
| CN106291452A (en) * | 2016-09-12 | 2017-01-04 | 电子科技大学 | A kind of direction of arrival estimation method based on enhanced greedy algorithm |
| CN106596724A (en) * | 2016-12-08 | 2017-04-26 | 清华大学 | Method for sparse compression optimization reconstruction of narrowband Lamb wave detection data |
| CN108282788A (en) * | 2018-01-19 | 2018-07-13 | 南京航空航天大学 | A kind of resource allocation methods of the Energy Efficient based on quasi- newton interior point method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105811407B (en) * | 2016-04-18 | 2018-07-06 | 清华大学 | A kind of micro-capacitance sensor primary frequency modulation control method based on distributed Newton method |
-
2019
- 2019-12-31 CN CN201911409497.1A patent/CN111130557B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101895297A (en) * | 2010-07-30 | 2010-11-24 | 哈尔滨工业大学 | Compressed sensing-oriented block-sparse signal reconfiguring method |
| CN102611455A (en) * | 2012-03-05 | 2012-07-25 | 哈尔滨工业大学 | Compressed sensing-oriented sparse multiband signal reconstruction method |
| CN103300859A (en) * | 2013-05-31 | 2013-09-18 | 王勇 | High-quality rapid CS-MRI (Compressed Sensing-Magnetic Resonance Imaging) method of mixed norm |
| CN106291452A (en) * | 2016-09-12 | 2017-01-04 | 电子科技大学 | A kind of direction of arrival estimation method based on enhanced greedy algorithm |
| CN106596724A (en) * | 2016-12-08 | 2017-04-26 | 清华大学 | Method for sparse compression optimization reconstruction of narrowband Lamb wave detection data |
| CN108282788A (en) * | 2018-01-19 | 2018-07-13 | 南京航空航天大学 | A kind of resource allocation methods of the Energy Efficient based on quasi- newton interior point method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111130557A (en) | 2020-05-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yao et al. | Dr2-net: Deep residual reconstruction network for image compressive sensing | |
| CN111027686B (en) | Landslide displacement prediction method, device and equipment | |
| Khobahi et al. | Model-based deep learning for one-bit compressive sensing | |
| Li et al. | Compressed sensing via dictionary learning and approximate message passing for multimedia Internet of Things | |
| CN111383741B (en) | Method, device and equipment for establishing medical imaging model and storage medium | |
| CN117011673B (en) | Electrical impedance tomography image reconstruction method and device based on noise diffusion learning | |
| Zhou et al. | Network traffic prediction method based on improved echo state network | |
| Yan et al. | Stochastic collocation algorithms using l_1-minimization for bayesian solution of inverse problems | |
| Song et al. | Short-term forecasting based on graph convolution networks and multiresolution convolution neural networks for wind power | |
| CN115730509A (en) | Spacecraft in-cabin temperature field reconstruction task research benchmark method based on machine learning | |
| CN116152465A (en) | Ocean environment field reconstruction method based on multistage interpolation method | |
| Yan et al. | Adaptive sign algorithm for graph signal processing | |
| CN115859116A (en) | Marine environment field reconstruction method based on radial basis function regression interpolation method | |
| Okuno et al. | Combining multiple forecasts for multivariate time series via state-dependent weighting | |
| JPWO2021059348A5 (en) | ||
| CN111130557B (en) | Data reconstruction method based on distributed quasi-Newton projection tracking | |
| Kartheeswaran et al. | A data-parallelism approach for PSO-ANN based medical image reconstruction on a multi-core system | |
| CN115809747B (en) | Pyramid causal network-based coupling information flow long-term prediction method | |
| Hu et al. | On geometric inverse problems in time-fractional subdiffusion | |
| Shen et al. | An efficient sensor quantization algorithm for decentralized estimation fusion | |
| Zhang et al. | Distributed compressive sensing via LSTM-aided sparse bayesian learning | |
| CN114155354B (en) | Method and device for reconstructing capacitance tomography based on graph convolution network | |
| Zhang et al. | Towards Probabilistic Robust and Sparsity-Free Compressive Sampling in Civil Engineering: A Review | |
| CN110113053B (en) | Distributed compressed sensing sparse signal reconstruction method | |
| CN114492165A (en) | Parameter optimization method and system based on genetic breeding method |
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 |