CN110769453B - Multi-modal monitoring data dynamic compression control method under unstable network environment - Google Patents
Multi-modal monitoring data dynamic compression control method under unstable network environment Download PDFInfo
- Publication number
- CN110769453B CN110769453B CN201911054546.4A CN201911054546A CN110769453B CN 110769453 B CN110769453 B CN 110769453B CN 201911054546 A CN201911054546 A CN 201911054546A CN 110769453 B CN110769453 B CN 110769453B
- Authority
- CN
- China
- Prior art keywords
- network
- data
- compression
- compression ratio
- value
- 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
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/06—Testing, supervising or monitoring using simulated traffic
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
- H04L1/0002—Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the transmission rate
- H04L1/0003—Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the transmission rate by switching between different modulation schemes
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
A multi-mode monitoring data dynamic compression method in an unstable network environment is characterized in that three stages of compression, transmission and decompression are used as a target for modeling, the unstable network state is predicted, a dynamic compression ratio adjusting process is introduced, and the time consumption of data compression, transmission and decompression is saved. Compared with a transmission optimization method based on lossless data compression, the transmission time consumption optimization method based on the dynamic compression ratio has a better effect of reducing transmission time consumption.
Description
Technical Field
The invention relates to the technical field of wireless network data compression, in particular to a dynamic compression control method for monitoring data in an unstable network environment.
Background
At present, various system objects (such as industrial production, large-scale equipment, systems and the like) are increasingly complex in industrial production scale, structure, function and the like, multi-source perception, multi-terminal acquisition and multi-channel network transmission are generally needed for overall monitoring of the objects, multi-mode monitoring data are collected and distributed to a decision layer in time, and the overall decision level of the objects has a very close correlation relation with the data scale, the characteristics and the network transmission environment of the system objects. There are the following problems in monitoring data transmission efficiency:
1. due to the fact that the number of monitoring data source nodes is large, the collection frequency of various types of data is unequal, and functional point positions have different spatial position distributions, the data needing to be transmitted is large in scale and can be in intermittent outbreak, and continuous remote network data transmission pressure is very large.
2. Most of media used for transmitting monitoring data are distributed networks, the uncertainty of network transmission environment conditions is high, the influence of irregular fluctuation of data transmission rate in a system operation period is caused, data transmission delay is unstable, and adverse conditions such as delay, disorder, packet loss and the like are easily caused.
CN103957582A discloses an invention patent named as "adaptive compression method for wireless sensor network", which discloses technical means such as selecting a compression algorithm according to data type and precision requirements, predicting an average compression ratio, predicting an average time for performing compression, building a model with optimal energy consumption as a target, and solving an optimal compression strategy. The purpose of the comparison file is to establish a mathematical model and solve with the optimal energy consumption as a target, and the mathematical model is not modeled with the optimal time as a target, so that a mathematical model formula consisting of compression, transmission and decompression is not disclosed, and the constraint conditions are set differently; in the calculation process of the comparison file, the condition that the network condition is unstable is not considered, the current network is not predicted, and the comparison between the predicted value and the true value of the network transmission rate is not performed, so that the comparison file cannot be applied to the unstable network.
Disclosure of Invention
The invention aims to provide a multi-mode monitoring data dynamic compression control method under an unstable network environment, which takes three stages of compression, transmission and decompression as a target for modeling, predicts the unstable network state, introduces a dynamic compression ratio adjustment process and saves the time consumption of data compression, transmission and decompression.
The invention aims to realize the technical scheme, which comprises an S1 compression algorithm selection step, an S2 compression ratio modeling and calculation step, an S3 compression transmission step and an S4 decompression step, wherein the S2 compression ratio modeling and calculation step models and calculates the compression ratio in real time according to the predicted value of the current network transmission rate, the S3 compression transmission step models and calculates the compression ratio in real time and dynamically adjusts the compression ratio according to the compression ratio calculated in the S2 step for transmission, and the S2 compression ratio modeling and calculation step comprises the following steps:
s2-1, according to the compression algorithm selected in step S1, for the data size of QiMultiple compression experiments are carried out on data of the ith (i-1, …, n) mode, and the compression time t is obtained through a scheme of data fittingPressure iThe relation with the compression ratio p is tPressure i＝fi(p,Qi)；
According to the compression algorithm selected in step S1, a statistical analysis experiment is performed on data of i (i ═ 1, …, n) modalities for multiple sets of decompression time and data compression ratio, and the decompression time t is obtained through a data fitting schemeSolution iThe relation with the compression ratio p is tSolution i＝gi(p,Qi)；
S2-2, establishing a time-consuming mathematical model including compression time, transmission time after coding and decompression time according to the relation between compression time and compression ratio and the relation between decompression time and compression ratio obtained in the step S2-1, wherein the formula of the time-consuming mathematical model is as follows:
in the formula, tPressure i-data compression time of the ith modality; t is t0-compressed data transmission time; t is tSolution i-data decompression time of the ith modality; qi-data size of the ith modality; Q-Overall data Scale; p-data compression ratio;
-a current network rate prediction value;
controlling the compression ratio P at a preset maximum compression ratio PmaxAnd a minimum compression ratio of PminAnd ensuring that the optimal time consumption is less than the original transmission time consumption, and establishing a constraint condition:
in the formula, Qi-data size of the ith modality; Q-Overall data Scale; p-data compression ratio;
-a current network rate prediction value;
s2-3, solving the optimal compression ratio under the current network environment according to the time-consuming mathematical model formula and the constraint conditions in the step S2-2.
Further, the current network transmission rate predicted value is calculated by adopting a learning-training-feedback algorithm with a neural network as a main algorithm, and the specific steps of calculating the current network transmission rate predicted value are as follows:
b1, selecting network monitoring data of a plurality of historical intervals as Uk(k is 1, …, m), and each segment of historical network interval has a plurality of network data Vj (k)(j-1, …, l) as input variables, and the last network data V in each historical network intervall (k)As a desired output variable;
b2, selecting N before (N)>m/2) segment historical network interval Uk(k-1, …, N) as training sample set, segment (m-N) th historical network interval Uk(k + N +1, …, m) as a test sample to test the accuracy of the estimate;
b3, for one section of historical transmission rate interval UkFirst (l-1) rate data Vj (k)(j ═ 1, …, l-1) as training samples, the l th as expected output, and the weight of (l-1) neurons is obtained by adopting the algorithm of the artificial neural network;
b4, continuously adjusting the weight value of the historical transmission rate interval of the remaining test samples according to the step S22, and finally obtaining the relation between the network speeds in each transmission rate interval as follows:
(j ═ 1, …, l-1); in the formula, Vj (k)(j ═ 1, …, l-1) — the jth wire speed value in the kth segment historical network interval; a isj(j ═ 1, …, l-1) -the weight corresponding to the jth wire speed value;
-the ith wire speed prediction value in the kth segment of historical network interval;
b5 obtaining U according to the relation obtained by b4kAnd (k is 1, …, N), obtaining the predicted value of the current network transmission rate as follows:
in the formula (I), the compound is shown in the specification,
-a current network rate prediction value; vj (m)(j ═ 1, …, l-1) — the jth wire speed value in the mth segment historical network interval; a isj(j ═ 1, …, l-1) -the weight corresponding to the jth wire speed value;
b6, obtaining U according to the algebraic relation obtained from b4kThe predicted value of the network transmission rate in (k ═ N +1, …, m-1) is as follows:
in the formula, Vj (k)(j ═ 1, …, l-1) — the jth wire speed value in the kth segment historical network interval; a isj(j ═ 1, …, l-1) -the weight corresponding to the jth wire speed value;
-the ith wire speed prediction value in the kth segment historical network interval.
Further, after the step of modeling and calculating the compression ratio in S2, according to the real value of the network speed and the predicted value of the network transmission rate of the test sample, the prediction error can be obtained as follows:
in the formula (I), the compound is shown in the specification,
-the ith network transmission rate prediction value in the kth segment of the historical network interval; the error is used as the threshold value of the actual value of the current network speed and the predicted value of the network transmission rate;
if the difference between the two is within the range of the set error threshold, the optimal compression ratio calculated by adopting the network transmission rate predicted value is adopted; and if the deviation is large, substituting the real value of the network speed into a formula of the time-consuming mathematical model to obtain the updated optimal compression ratio and the overall transmission time consumption.
Further, the method of selecting the compression algorithm in S1 is as follows:
classifying the collected multi-modal data according to data types, and dividing the data into n modes; and selecting a plurality of lossless compression algorithms for each data mode, and testing the compression time consumption of different algorithms of unit data under the same compression ratio, wherein the time consumption is the shortest, namely the optimal compression algorithm.
Due to the adoption of the technical scheme, the invention has the following advantages:
compared with a transmission optimization method based on lossless data compression, the transmission time consumption optimization method based on dynamic compression ratio has a better effect of reducing transmission time consumption; firstly, because extra encoding/decoding time is introduced when transmission data is compressed, and the extra encoding/decoding time is related to the data compression degree, the fixed compression ratio cannot ensure the minimum time consumption of system transmission under any system transmission environment; and secondly, aiming at the condition that the monitored data accounts for basically stable, the fixed data compression ratio represents that the compression/decompression time consumption is not adjusted, which means that the influence of the improvement of the transmission rate on the compression/decompression time is not obvious, however, the transmission optimization method based on the dynamic compression ratio can more effectively improve the data transmission efficiency.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof.
Drawings
The drawings of the present invention are described below.
FIG. 1 is a structural level diagram of multimodal data transfer;
FIG. 2 is a schematic flow chart of the present invention.
Detailed Description
The invention is further illustrated by the following figures and examples.
The patent selects multi-modal monitoring data of the chemical plant as example objects. As a complex industry, the chemical industry has various collection equipment nodes, so that the system collects a large amount of monitoring data of different types at an indefinite frequency; meanwhile, the network transmission environment condition of the chemical production workshop is uncertain, so that the data transmission delay is unstable and the packet loss problem is easy to generate, and the problems are very serious for the time-sensitive scene of the chemical industry. Therefore, by adopting the optimization control method, the transmission efficiency of the multi-modal data in the unstable environment can be effectively improved.
As shown in fig. 1 and fig. 2, a method for dynamically compressing and controlling multi-modal monitoring data in an unstable network environment includes the following steps:
s1: selecting a lossless compression algorithm and a decompression algorithm which are suitable for multi-modal data characteristics in a chemical plant environment;
s2, predicting the current network condition according to the historical network transmission rate set of the chemical plant;
s3: establishing an optimization model taking the overall transmission time consumption as an optimal target according to the current network rate predicted value;
s4: the compression ratio is used as a decision variable to control the compression/decompression time of the compression ratio, so that the optimal compression ratio is solved;
s5: judging whether to adjust the optimal compression ratio by taking the actual transmission rate of the chemical plant as a standard;
s6: and obtaining the adjusted optimal compression ratio, and transmitting the compressed multi-mode detection data.
The selection of the applicable compression algorithm is to select the applicable lossless compression algorithm to reduce the information redundancy of the compression algorithm by combining various factors such as the characteristics of the compressed data to be coded, the data scale, the application scene and the like, and to restore the data information losslessly by the decompression algorithm. The method comprises the following specific steps:
s11: and classifying the collected multi-modal data according to the data types, and dividing the data into n types of modes.
S12: and selecting a plurality of lossless compression algorithms for each data mode, and testing the compression time consumption of different algorithms of unit data under the same compression ratio, wherein the time consumption is the shortest, namely the optimal compression algorithm. Therefore, the data of n modes respectively correspond to an optimal compression algorithm.
S13: the optimal decompression algorithm for each data modality is obtained according to the method of step S12, and thus, the data of n modalities corresponds to one optimal decompression algorithm.
And acquiring the current network environment, namely acquiring the current network transmission rate. Because the current transmission environment is under the condition of unstable network speed, the network speed in a section of history interval is adopted to perform weighted average to obtain an average network speed to approximate the average network speed as the current network condition, and the current network transmission speed is estimated
If the current rate is directly used as a decision variable to perform the optimal compression ratio operation, the overall transmission time consumption is increased due to the extra operation time consumption. If the optimal compression ratio can be obtained in advance by using the estimated value of the current network speed before the current network speed is acquired, the decision efficiency of data compression transmission can be greatly improved, and the time consumption of the whole transmission is reduced. Therefore, in order to obtain an accurate and effective network transmission rate predicted value, the method adopts an algorithm of an artificial neural network to predict the current transmission rate value. The method comprises the following specific steps:
s21: selecting network monitoring data of a plurality of historical intervals as Uk(k is 1, …, m), and each segment of historical network interval has a plurality of network data Vj (k)(j-1, …, l) as input variables, and the last network data V in each historical network intervall (k)As the desired output variable.
S22: selecting the first N (N)>m/2) segment historical network interval Uk(k-1, …, N) as training sample set, segment (m-N) th historical network interval Uk(k N +1, …, m) was used as a test sample to test the accuracy of the estimates.
S23: for one section of historical transmission rate interval UkFirst (l-1) rate data Vj (k)(j-1, …, l-1) as training sample, and the l-th as expected output, and using the algorithm of artificial neural network to obtain the weight of (l-1) neurons.
S24: and continuously adjusting the weight value of the historical transmission rate interval of the remaining test samples according to the step of S22, and finally obtaining the relation between the network speeds in each transmission rate interval as follows:
s25: u can be obtained from the relationship obtained at S24kAnd (k is 1, …, N), obtaining the predicted value of the current network transmission rate as follows:
s26: similarly, U can be obtained from the algebraic relation obtained in S24kThe predicted value of the network transmission rate in (k ═ N +1, …, m-1) is as follows:
and obtaining a prediction error according to the real value and the predicted value of the network speed of the test sample as follows:
the error is used as the threshold value of the actual value and the predicted value of the current network speed.
And (3) establishing an optimization model, and establishing a target model by taking a single compression time or data transmission time or decompression time as a decision factor. The optimization model is established by taking the overall transmission time consumption as an optimal target, wherein the overall transmission time consumption means that the data transmission time after the compression algorithm is introduced consists of 3 parts of compression time, time spent in the transmission process of the monitoring data after encoding and decompression time, namely the time consumption of data processing caused by the compression algorithm and the decompression algorithm is also considered. And establishing a target optimization model by multi-factor comprehensive decision. The method comprises the following specific steps:
s31: for data size of QiPerforming multiple compression experiments on data of the ith (i-1, …, n) mode, adjusting compression ratio parameters of a compression algorithm, combining the trend of statistical results on compression time along with the change of the compression ratio, and obtaining the compression time t by a least square fitting schemePressure iThe relation with the compression ratio p is tPressure i＝fi(p,Qi)；
S32: similarly, a statistical analysis experiment of multiple sets of decompression time and data compression ratio was performed on data of i (i ═ 1, …, n) modes, and the compression time t was obtained by the method described in S21Solution iThe relation with the compression ratio p is tSolution i＝gi(p,Qi)；
S33: according to the predicted value of the current transmission rate, the relation between the whole transmission time consumption composition of the system after a compression algorithm, the data compression/decompression time and the compression ratio is introduced, on the basis, the whole transmission time consumption of the system is taken as an optimization target, and the following target functions can be obtained:
wherein Q and
the method comprises the steps that the method represents the ratio of monitoring data to be transmitted currently of a system and a transmission rate predicted value under the current environment respectively, so that the optimization model only has one decision variable p under the condition that the data ratio and the transmission rate are determined;
s34: in combination with the actual test result, the invention controls the compression ratio p within the range of [2,100] to avoid the time consumption of the whole transmission caused by the sudden increase of the data processing time; on the other hand, it needs to be ensured that the optimal time consumption of the model is less than the original transmission time consumption of the system, and an 'invalid' optimization decision is avoided, so that the system has the following constraint conditions in the transmission optimization process:
and according to the optimization model and the constraint conditions established in the S3, the compression ratio is used as a decision variable to control the whole transmission time, and the optimal compression ratio under the current approximate network speed is solved.
The optimal compression ratio is adjusted and fed back by comparing the actually detected network transmission rate value with the predicted value, and the purpose of feedback can be achieved by improving the decision efficiency and not influencing the decision result. If the difference between the two is within the range of the set error threshold, the original optimal compression ratio is adopted, and the decision efficiency is improved; if a large deviation exists, the actual speed value needs to be substituted into the objective function to obtain the updated optimal compression ratio and the overall transmission time, so that the decision result is not influenced even if the efficiency cannot be improved. The method comprises the following specific steps:
s51: detecting the current network transmission rate D of the chemical plant;
s52: comparing the estimated transmission rate with the actual transmission rate if the estimated transmission rate is not equal to the actual transmission rate
(E is the threshold error in S25), the two are considered to be equal, otherwise, there is a large deviation between the two;
s53: if the two are approximately equal, the original compression ratio is adopted; and if the large deviation exists, substituting the actual speed value into the objective function to obtain the updated optimal compression ratio and the overall transmission time consumption.
And finally, performing lossless compression on the multi-mode monitoring data by adopting the adjusted and decided optimal compression ratio, and transmitting the compressed data.
Compared with the transmission optimization method based on lossless data compression, the transmission time consumption optimization method based on the dynamic compression ratio has better effect of reducing the transmission time consumption. Firstly, because extra encoding/decoding time is introduced when transmission data is compressed, and the extra encoding/decoding time is related to the data compression degree, the fixed compression ratio cannot ensure the minimum time consumption of system transmission under any system transmission environment; and secondly, aiming at the condition that the monitored data accounts for basically stable, the fixed data compression ratio represents that the compression/decompression time consumption is not adjusted, which means that the influence of the improvement of the transmission rate on the compression/decompression time is not obvious, however, the transmission optimization method based on the dynamic compression ratio can more effectively improve the data transmission efficiency.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (4)
1. A multi-modal monitoring data dynamic compression method under an unstable network environment comprises a step of S1 selecting a compression algorithm, a step of S2 modeling and calculating a compression ratio, a step of S3 compression transmission and a step of S4 decompression, and is characterized in that: the S2 modeling compression ratio calculation step models and calculates the compression ratio in real time according to the predicted value of the current network transmission rate, the S3 compression transmission step dynamically adjusts the compression ratio in real time according to the compression ratio calculated in the S2 step for transmission, and the S2 modeling compression ratio calculation step comprises the following steps:
s2-1, according to the compression algorithm selected in step S1, for the data size of QiThe data of the ith mode is subjected to a plurality of compression experiments, wherein i is 1, …, n, and the compression time t is obtained through a scheme of data fittingPressure iThe relation with the compression ratio p is tPressure i＝fi(p,Qi) (ii) a Where i represents the modality class of the data, n represents the total number of modality classes, fiRepresenting the functional relation between the compression time and the compression ratio of the ith modal data;
according to the compression algorithm selected in the step S1, carrying out a statistical analysis experiment of a plurality of groups of decompression time and data compression ratio on the data of the i modes, and obtaining the decompression time t through a data fitting schemeSolution iThe relation with the compression ratio p is tSolution i＝gi(p,Qi) (ii) a Wherein, giExpressing the functional relation between the decompression time and the compression ratio of the ith mode data;
s2-2, establishing a time-consuming mathematical model including compression time, transmission time after coding and decompression time according to the relation between compression time and compression ratio and the relation between decompression time and compression ratio obtained in the step S2-1, wherein the formula of the time-consuming mathematical model is as follows:
in the formula, tPressure i-data compression time of the ith modality; t is t0-compressed data transmission time; t is tSolution i-data decompression time of the ith modality; qi-data size of the ith modality; Q-Overall data Scale; p-data compression ratio;
-a current network transmission rate prediction value;
controlling the compression ratio P at a preset maximum compression ratio PmaxAnd a minimum compression ratio of PminAnd ensuring that the optimal time consumption is less than the original transmission time consumption, and establishing a constraint condition:
s2-3, solving the optimal compression ratio under the current network environment according to the time-consuming mathematical model formula and the constraint conditions in the step S2-2.
2. The method of dynamically compressing multi-modal monitoring data in an unstable network environment according to claim 1, wherein: the current network transmission rate predicted value is calculated by adopting a learning-training-feedback algorithm with a neural network as a main algorithm, and the specific steps of calculating the current network transmission rate predicted value are as follows:
b1, selecting the network monitoring data of a plurality of sections of historical network intervals as UkWherein k is 1, …, m, eachA plurality of speed data V are arranged in the segment history network intervalj (k)As input variables, where j is 1, …, l, and the last rate data V in each historical network intervall (k)As a desired output variable; wherein k represents the kth historical network interval; m represents the total number of historical network intervals; j represents the jth wire speed value; l represents the total number of the network speed values in any section of the historical network interval;
b2, selecting the previous N segments of historical network intervals, N>m/2，UkAs a training sample set, k is 1, …, N, and then m-N segments of historical network intervals UkK is N +1, …, m, which is used as a test sample for testing the accuracy of estimation;
b3, for one section of historical network interval UkFirst l-1 rate data Vj (k)Taking j as a training sample, wherein j is 1, …, l-1, the l-th rate data is taken as an expected output, and an algorithm of an artificial neural network is adopted to obtain the weight of l-1 neurons;
b4, continuously adjusting the weight of the historical network interval with the rest test samples according to the step b2, and finally obtaining the relation between the network speeds in each segment of the historical network interval as follows:
wherein j is 1, …, l-1; in the formula, Vj (k)The network speed value of the jth network speed in the kth section of historical network interval is j ═ 1, …, l-1; a isjThe weight value corresponding to the jth network speed value, wherein j is 1, …, l-1;
the net speed prediction value is the l net speed prediction value in the k section of historical network interval;
b5 obtaining U according to the relation obtained by b4kAn algebraic relation of the internal transmission rate, wherein k is 1, …, N, and a predicted value of the current network transmission rate is obtained as follows:
in the formula (I), the compound is shown in the specification,
predicting a current network rate value; vj (m)The j-th network speed value in the m-th historical network interval is shown, wherein j is 1, …, l-1; a isjThe weight value corresponding to the jth network speed value, wherein j is 1, …, l-1;
b6, obtaining a test set U according to the algebraic relation obtained by b4kWherein k is N +1, …, and the predicted value of the network transmission rate in m is as follows:
wherein j is 1, …, l-1, k is N +1, …, m; in the formula, Vj (k)The network speed value is the jth network speed value in the kth section of historical network interval; a isjThe weight value is corresponding to the jth network speed value;
and predicting the net speed value of the ith section in the historical network interval.
3. The method for dynamically compressing multi-modal monitoring data in an unstable network environment as recited in claim 2, wherein: after the step of modeling and calculating the compression ratio in S2, the prediction error can be obtained according to the real network speed value and the predicted network transmission rate value of the test sample as follows:
in the formula (I), the compound is shown in the specification,
for the ith network transmission rate prediction value, V, in the qth historical network intervall (q)The real value of the net speed is the ith net speed in the qth section of the historical network interval; the error is used as the threshold value of the current real value of the network speed and the predicted value of the network transmission rate;
if the difference between the two is within the range of the set error threshold, the optimal compression ratio calculated by adopting the network transmission rate predicted value is adopted; and if the deviation is large, substituting the real value of the network speed into a formula of the time-consuming mathematical model to obtain the updated optimal compression ratio and the overall transmission time consumption.
4. The method for dynamically compressing multi-modal monitoring data in an unstable network environment according to claim 3, wherein the S1 selects the compression algorithm as follows:
classifying the collected multi-modal data according to data types, and dividing the data into n modes; and selecting a plurality of lossless compression algorithms for each data mode, and testing the compression time consumption of different algorithms of unit data under the same compression ratio, wherein the time consumption is the shortest, namely the optimal compression algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911054546.4A CN110769453B (en) | 2019-10-31 | 2019-10-31 | Multi-modal monitoring data dynamic compression control method under unstable network environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911054546.4A CN110769453B (en) | 2019-10-31 | 2019-10-31 | Multi-modal monitoring data dynamic compression control method under unstable network environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110769453A CN110769453A (en) | 2020-02-07 |
| CN110769453B true CN110769453B (en) | 2020-11-06 |
Family
ID=69335087
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911054546.4A Active CN110769453B (en) | 2019-10-31 | 2019-10-31 | Multi-modal monitoring data dynamic compression control method under unstable network environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110769453B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111865326B (en) * | 2020-07-14 | 2023-06-23 | 北京灵汐科技有限公司 | Data compression method, device, equipment and storage medium |
| CN113746870B (en) * | 2021-11-05 | 2022-02-08 | 山东万网智能科技有限公司 | Intelligent data transmission method and system for Internet of things equipment |
| CN114979290A (en) * | 2022-04-14 | 2022-08-30 | 中国联合网络通信集团有限公司 | Situation awareness compression algorithm selection method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011162964A2 (en) * | 2010-06-25 | 2011-12-29 | Intel Corporation | System and method and computer program product for parameter estimation for lossless video compression |
| CN103957582A (en) * | 2014-05-17 | 2014-07-30 | 浙江大学宁波理工学院 | Wireless sensor network self-adaptation compression method |
| CN106649026A (en) * | 2016-09-26 | 2017-05-10 | 国家电网公司北京电力医院 | Monitoring data compression method applicable to operation and maintenance automation system |
| CN109243558A (en) * | 2018-08-28 | 2019-01-18 | 重庆汇邡机械制造有限公司 | Data after carrying out big data collection extract optimization method |
-
2019
- 2019-10-31 CN CN201911054546.4A patent/CN110769453B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011162964A2 (en) * | 2010-06-25 | 2011-12-29 | Intel Corporation | System and method and computer program product for parameter estimation for lossless video compression |
| CN103957582A (en) * | 2014-05-17 | 2014-07-30 | 浙江大学宁波理工学院 | Wireless sensor network self-adaptation compression method |
| CN106649026A (en) * | 2016-09-26 | 2017-05-10 | 国家电网公司北京电力医院 | Monitoring data compression method applicable to operation and maintenance automation system |
| CN109243558A (en) * | 2018-08-28 | 2019-01-18 | 重庆汇邡机械制造有限公司 | Data after carrying out big data collection extract optimization method |
Non-Patent Citations (2)
| Title |
|---|
| Proximity Operator Based Alternating Iteration Algorithm for Sparse Signal Recovery;CHAI Yi 等;《Proceedings of the 33rd Chinese Control Conference》;20140730;全文 * |
| 基于参数估计的通用工业数据在线压缩方法;沈春锋 等;《控制工程》;20110930;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110769453A (en) | 2020-02-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110769453B (en) | Multi-modal monitoring data dynamic compression control method under unstable network environment | |
| CN109934337B (en) | Spacecraft telemetry data anomaly detection method based on integrated LSTM | |
| CN108764568B (en) | Data prediction model tuning method and device based on LSTM network | |
| JP2017524205A (en) | Model predictive control using wireless process signals | |
| EP3938854A1 (en) | Method of performing a process and optimizing control signals used in the process | |
| WO2021213177A1 (en) | Dynamic compensation method for ship sectional construction curved surface deformation | |
| CN107578093A (en) | The Elman neural network dynamic Forecasting Methodologies of Landslide Deformation | |
| CN108803528A (en) | Process industry system prediction model based on multivariate correlation and time lag | |
| KR20200063338A (en) | Prediction system based on parameter improvement through learning and method thereof | |
| CN110769000B (en) | Dynamic compression prediction control method of continuous monitoring data in unstable network transmission | |
| CN109918280B (en) | Health management method and device for servo system | |
| CN108959787B (en) | Thermal deformation prediction method and system of macro-macro dual-drive system considering actual working conditions | |
| CN113326132A (en) | Information adjusting method, equipment and storage medium | |
| CN111108738B (en) | Data processing device, data analysis device, data processing system, and method for processing data | |
| CN116880191A (en) | Intelligent control method of process industrial production system based on time sequence prediction | |
| CN116859839A (en) | Industrial control method and device based on model training | |
| CN110738403A (en) | data processing method, device and computer storage medium | |
| CN116842440A (en) | Self-adaptive link switching method, system, equipment and medium based on context awareness | |
| CN111369072A (en) | Nuclear minimum mean square time sequence online prediction model based on sparsification method | |
| CN115963420A (en) | Battery SOH influence factor analysis method | |
| CN115619028A (en) | Clustering algorithm fusion-based power load accurate prediction method | |
| CN108537322A (en) | Neural network interlayer activation value quantization method and device | |
| CN103135444A (en) | Steel production energy consumption immune prediction control model | |
| CN110309472B (en) | Offline data-based policy evaluation method and device | |
| CN110570032B (en) | Initial weight optimization method and system for user behavior model |
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 |