CN110769000B - Dynamic compression prediction control method of continuous monitoring data in unstable network transmission - Google Patents

Abstract

A dynamic compression prediction control method of continuous monitoring data in unstable network transmission includes S1 data acquisition step, S2 modeling calculation compression ratio step, S3 compression transmission step and S4 decompression step. Compared with the transmission optimization method based on lossless data compression, the transmission time-consuming optimization method based on the dynamic compression ratio is adopted to reduce the transmission time-consuming, which greatly saves the time-consuming of compression, transmission and decompression; the data is processed in time periods. , calculate the compression ratio by predicting the network speed, which is suitable for the unstable network environment; establish constraints by calculating the direct transmission time of the original data in the predicted network speed environment, and combine the function with the least time-consuming as the goal to calculate The predicted network speed compression ratio is compared with the current network speed compression ratio to ensure the accuracy of the optimal compression ratio.

Description

Dynamic Compression Predictive Control of Continuous Monitoring Data in Unstable Network Transmission method

technical field

The invention relates to the technical field of data compression, in particular to a dynamic compression prediction control method in an unstable network.

Background technique

Data compression is a widely used processing method to ensure the quality of the original data and reduce the size of the data. However, compression and decompression are time-consuming and closely related to the compression ratio. If data compression is used to improve network transmission efficiency, the following issues need to be considered:

1. The transmission task of the entire monitoring data includes the whole process from the acquisition of the underlying sensor to uploading to the target object and restoring it to the original data. In this process, the optimal compression ratio is only selected according to the current network environment and the proportion of data. Unreliable, ignoring the influence of the fluctuation frequency of the data transmission rate on the compression control decision, the subsequent change of the transmission rate will be greatly affected by the current compression strategy, and subject to the continuous characteristics of continuous time series data, this influence may or As a result, the transmission efficiency does not increase but decreases.

2. If a dynamic compression ratio is used, the update frequency of the compression ratio needs to be fully considered, which is related to the change of the transmission rate. Once the data is sent, its internal structure and size can no longer be changed, that is, the proportion of the data to be transmitted is determined in the entire transmission process. If the subsequent transmission rate changes frequently at this time, it may make the optimal The compression ratio cannot meet the overall time-consuming optimization requirements of the current transmission stage, which directly leads to the asymmetry between the compression dynamic control adjustment and the actual data transmission requirements, and the optimization decision lags or even fails.

CN103957582A discloses an invention patent named "wireless sensor network adaptive compression method", which discloses selecting compression algorithms according to data types and precision requirements, predicting the average compression ratio, predicting the average time to perform compression, and The technical means such as establishing a model for the goal of optimal consumption and solving the optimal compression strategy.

The purpose of this comparison document is to establish a mathematical model and solve it with the goal of optimal energy consumption. It does not model with the goal of time optimization. Therefore, it does not disclose the mathematical model formula composed of compression, transmission and decompression, and the constraints are established. It is not the same; the comparison document does not take into account the unstable network conditions during the calculation process, and does not predict the current network, so it cannot be applied to unstable networks.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a dynamic compression prediction control method for continuous monitoring data in unstable network transmission, which compresses the data in cycles, predicts the network speed of the next cycle, and takes the shortest transmission time as Target modeling calculates the optimal compression ratio, saving a lot of time in data compression, transmission and decompression.

The purpose of the present invention is achieved through such a technical solution, which includes S1 data acquisition step, S2 modeling calculation compression ratio step, S3 compression transmission step and S4 decompression step:

内待传输的连续型数据和网络速率历史样本；时间周期

针对信息系统在一段时间

内积累的数据包需要发送，其发送时间要小于数据积累周期

The S1 data collection step obtains the current time period

Continuous data and network rate historical samples to be transmitted; time period

For information systems over a period of time

The accumulated data packets need to be sent, and the sending time should be less than the data accumulation period

的预测网速；S2-2、根据预测网速以耗时最少为目标建模求解最优压缩比；The step of S2 modeling and calculating the compression ratio includes: S2-1, calculating the next time period according to the network rate historical sample

The predicted network speed; S2-2, according to the predicted network speed, modeling and solving the optimal compression ratio with the least time consuming as the goal;

的预测网速，具体步骤如下：In step S2-1, the time series prediction method is used to calculate the next time period according to the historical samples of the network rate

The specific steps are as follows:

S2-1. Stationarity test of historical network series: judge whether the time series is a stationary time series according to the mean, variance and correlation coefficient of the network series;

S2-2. Sequence difference operation: if the sequence is a non-stationary time series, perform a difference operation on these sequences until the sequence satisfies stationarity;

S2-3. Model fitting of the sequence: According to the correlation coefficient ACF, partial correlation coefficient PACF and periodicity of the sample, the differenced sequence is summed and the autoregressive moving average model ARIMA(p, d, q) is fitted to obtain the time sequence function d(t);

S2-4. Test of residual sequence: perform white noise test on the residual sequence of the time series function d(t), if the residual sequence does not satisfy the white noise sequence, return to step S2-3 to re-fit the model until the residual until the difference sequence is white noise;

S2-5. Model optimization: the time series fitting model function d(t) is optimized by using the minimum information criterion to obtain the updated model function d(t);

来预测出下一时间段

的网络速率。S2-6. Prediction of series: fitting function d(t+l) with time series

to predict the next time period

network speed.

Further, according to the predicted network speed described in step S2-2, the specific method for modeling and solving the optimal compression ratio with the least time consuming as the target is as follows:

S2-2-1. Calculate the direct transmission time of the original data in the predicted network speed environment

设立约束条件；S2-2-2, establish an optimization model with the overall transmission time as the optimal goal, and combine the direct transmission time calculated in S2-2-1

establish constraints;

S2-2-3. Under the current network speed environment, solve the compression ratio and the overall transmission time consumption t;

S2-2-4, in the prediction network speed environment, solve the compression ratio and the overall transmission time consumption t;

S2-2-5, compare the total transmission time consumption t of S2-2-3 and S2-2-4, and select the compression ratio corresponding to the smaller overall transmission time consumption t as the optimal compression ratio.

Further, the specific method for calculating the time-consuming direct transmission of the original data in the predicted network speed environment described in step S2-2-1 is:

可以直接求出原始传输耗时

Determine whether the data transmission rate prediction sequence d(t) is integrable, if it is, according to the formula

The original transmission time can be directly calculated

求解出n，进而获取原始传输耗时

If it is not integrable, since the change of the rate of continuous data within the sampling period Δt is small and can be ignored, the discrete sequence set of d(t) is integrated in pieces, namely

Solve n, and then obtain the original transmission time

In the formula, t _i represents the sampling time, D _i represents the rate in the ith sampling period, and n represents the number of sampling time points.

设立约束条件的具体方法为：Further, as described in step S2-2-2, an optimization model with the overall transmission time consumption as the optimal target is established, combined with the direct transmission time consumption calculated in S2-2-1

The specific method of setting up constraints is as follows:

For many experiments, it can be obtained that the relationship between the unit original data packet compression time t ₁ and the compression ratio p is t ₁ =T ₁ (p) and the unit original data packet decompression time t ₂ and the compression ratio p The relationship between the compression ratio p is t ₂ =T ₂ (p); build the objective function

mint=t ₁ +t*+t ₂ =QT ₁ (p)+t*+QT ₂ (p)

Set the maximum compression ratio and the minimum compression ratio as P _max and P _min respectively, and set up constraints:

为未压缩数据原始传输耗时，

网速序列预测周期，t^*为压缩后数据传输耗时，Q为待传输数据的规模大小。In the formula,

Time consuming for raw transmission of uncompressed data,

The prediction period of the network speed sequence, t ^* is the time taken for data transmission after compression, and Q is the size of the data to be transmitted.

Further, step S2-2-3 and step S2-2-4 adopt genetic algorithm to solve the optimal compression ratio, and the specific method is as follows:

1. Coding: According to the compression ratio range of each mode [P _min , P _max ], binary coding with length k is used, and there are 2 ^k different codes. The adjacent coding interval is

2. Generation of the initial population: randomly generate N string structure data as the initial population to start evolution, that is, generate N initial compression ratios p encoded in binary as the initial population;

3. Fitness evaluation: select the fitness function as t=t ₁ +t*+t ₂ =QT ₁ (p)+t*+QT ₂ (p), and calculate the fitness function value of each initial individual in the population ;

4. Natural selection: individuals whose fitness function value t satisfies the constraints and are smaller than the population average are retained as individuals with strong adaptability and added to the new population;

5. Crossover and mutation: Crossover is to exchange part of the code between individuals and individuals. Mutation is to randomly select an individual in the population to randomly change a character in the code with a small probability. The purpose of crossover and mutation are both. Get new individuals to add to the new population;

6. Whether to stop the evolution: Termination condition: the number of evolution is the set number or the variance of the population fitness is less than the set variance, if the termination condition is met, the evolution will be stopped, otherwise the updated population will return to step 3;

7. Decoding: Convert the optimal individual left by selection into the original parameters through encoding as the optimal compression ratio, and obtain the optimal fitness at the same time.

大于系统数据发送时间。Further, the time period

It is longer than the system data sending time.

Owing to adopting the above-mentioned technical scheme, the present invention has the following advantages:

1. Compared with the transmission optimization method based on lossless data compression, the present invention adopts the transmission time-consuming optimization method based on the dynamic compression ratio to reduce the transmission time-consuming, and greatly saves the time-consuming of compression, transmission and decompression;

2. The present invention processes the data in time periods, and calculates the compression ratio by predicting the network speed, which is suitable for unstable network environments;

3. The present invention sets up constraints by calculating the direct transmission time of the original data in the predicted network speed environment, and combines the function with the least time consuming as the goal to calculate the predicted network speed compression ratio and compare the current network speed compression ratio to ensure that the The accuracy of the optimal compression ratio.

Other advantages, objects, and features of the present invention will be set forth in the description that follows, and will be apparent to those skilled in the art based on a study of the following, to the extent that is taught in the practice of the present invention. The objectives and other advantages of the present invention may be realized and attained by the following description and claims.

Description of drawings

The accompanying drawings of the present invention are described below.

Fig. 1 is the schematic flow chart of the present invention;

Fig. 2 is the structural level diagram of continuous operation detection data transmission of the power system;

Figure 3 is a flow chart of time series fitting and forecasting.

Detailed ways

The present invention will be further described below with reference to the accompanying drawings and embodiments.

As shown in Figure 1, Figure 2 and Figure 3, the dynamic compression prediction control method of continuous operation detection data of the power system in the transmission of continuous monitoring data in an unstable network is described as an example, including the following steps:

S1: Periodically collect the network rate in the power system to obtain the current network data;

S2: According to the current data transmission environment, it takes T to obtain the current direct transmission of continuous data;

内传输速率的变化趋势；S3: Predict the next time period at the current moment

The change trend of internal transmission rate;

S4: Obtain future direct transmission time based on changes in future transmission rates

S5: Establish an optimization model with the overall transmission time as the optimal goal according to the current transmission rate;

S6: Use the network rate in the future time period as a decision variable to adjust the optimal compression ratio;

S7: Obtain the optimal compression ratio, and transmit the compressed continuous detection data.

Further, the network rate collection is based on the time taken for a certain proportion of data packets to be transmitted from the server to the client, so as to calculate the current transmission rate.

The specific steps are as follows:

S11: During the data transmission process, set the server to send the query signal get_time to the client at a cycle of 100 milliseconds, and record the size q of the query signal at the same time. Before sending the query signal each time, the server uses the current sending time node as the start of the receipt node, and cache it, and the start node of the receipt is recorded as start_time;

S12: After the client receives the get_time signal, it immediately records the current time node, and records it as end_time, and finally returns the end_time to the server for requesting a receipt, thus completing a data transmission rate sequence collection;

S13: By calculating the difference between end_time and start_time, the request time of the receipt signal can be obtained. Combined with the size of the receipt signal q, the data transmission rate in each collection period can be obtained as:

The direct transmission time T is calculated based on the current data transmission rate D and the data ratio Q to obtain the direct transmission time.

To predict the change trend of the transmission rate, it is necessary to predict the change trend of the transmission rate in a period of time in the future through an effective time series prediction algorithm.

Specific steps are as follows:

内积累的数据包需要发送给控制中心，其发送时间要小于数据积累周期

否则会导致数据发送混乱，因此可将预测周期为

S31: Define the forecast period, for the power information system in a period of time

The accumulated data packets need to be sent to the control center, and the sending time should be less than the data accumulation period

Otherwise, it will cause confusion in data transmission, so the prediction period can be set as

的网络速率预测序列

S32: According to a large number of historical network rate samples in the power system, the method of time series prediction is used to predict the next time period

The network rate prediction sequence of

Obtaining the original transmission time is a combination of the predicted change trend of the transmission rate to obtain the transmission time of the uncompressed data under the unstable network speed.

倘若d(t)是可积序列，则可以直接求出原始传输耗时

倘若d(t)无法直接积分则可采用下述方法：Let d(t) represent the data transmission rate prediction sequence at this stage, and Q represents the proportion of the original monitoring data, so according to the formula

If d(t) is an integrable sequence, the original transmission time can be directly calculated

If d(t) cannot be directly integrated, the following method can be used:

其中t_i表示采样时刻；D_i表示第i个采样周期内的速率，从而求解出n来，进而获取原始传输耗时

According to the experimental analysis, the change of the rate in the acquisition period is small and can be ignored. Based on this, the discrete sequence set of d(t) can be integrated in pieces, namely

Among them, t _i represents the sampling time; D _i represents the rate in the ith sampling period, so as to solve n, and then obtain the original transmission time.

To establish an optimization model, the data transmission time after the introduction of the compression algorithm will be composed of the compression time, the time consumed by the encoded monitoring data in the transmission process, and the decompression time, that is, it is necessary to establish an optimization with the overall transmission time as the optimal goal. Model.

Specific steps are as follows:

S51: For multiple trials, the relationship between the compression time t ₁ of the unit original data packet and the compression ratio p can be obtained as t ₁ =T ₁ (p);

S52: Similarly, it can also be obtained that the relationship between the unit original data packet decompression time t ₂ and the compression ratio p is t ₂ =T ₂ (p);

S53: In view of the influence of the transmission rate on the transmission time-consuming optimization decision in an unstable network environment, the effectiveness of the optimization decision in the entire transmission stage is ensured by dynamically adjusting the compression ratio. Based on this, an optimal transmission time-consuming model based on transmission rate prediction is constructed. The objective function is:

mint=t ₁ +t*+t ₂ =QT ₁ (p)+t*+QT ₂ (p)

Among them, t* represents the time consumed in the transmission phase after the data is compressed and encoded under the current network transmission rate D.

S54: Considering the influence of the transmission rate change on the current optimization result, there are constraints:

To solve the optimal compression ratio, according to the above-mentioned transmission time-consuming optimization model and constraints based on the current rate, the compression ratio is used as a decision variable to control the overall transmission time. Optimal compression ratio, in order to ensure that the system transmission time is minimized.

Since the model is a nonlinear model, the present invention adopts an evolutionary algorithm such as a genetic algorithm to solve the nonlinear programming problem.

The specific steps for its solution are as follows:

S61: Coding: According to the compression ratio range of each mode is [2,100], binary coding with length k is used, there are 2 ^k different codes, and the adjacent coding interval is

S62: Generation of the initial population: randomly generate N string structure data as the initial population to start evolution, that is, generate N initial compression ratios p encoded in binary as the initial population;

S63: Fitness evaluation: select the fitness function as t=t ₁ +t*+t ₂ =QT ₁ (p)+t*+QT ₂ (p), and calculate the fitness function value of each initial individual in the population ;

S64: natural selection: individuals whose fitness function value t satisfies the constraint condition and is smaller than the population average value among individuals are retained as individuals with strong adaptability and added to the new population;

S65: Crossover and mutation: Crossover is to exchange part of the code between individuals and individuals, and mutation is to randomly select an individual in the population to randomly change a character in the code with a small probability. Get new individuals to add to the new population;

S66: Whether to stop the evolution: Termination condition: the number of evolutions reaches the set number or the variance of the population fitness is less than the set variance, if the termination condition is met, the evolution is stopped, otherwise the updated population is returned to step S63;

S67: Decoding: Convert the optimal individual left by selection into the original parameter as the optimal compression ratio through encoding, and obtain the optimal fitness at the same time.

Further, the adjustment of the optimal compression ratio is to adjust the current optimal compression ratio according to the network rate prediction trend d(t). At this time, the constraints should become:

At this time, t* represents the time spent in the transmission phase after the data is compressed and encoded when the change of the transmission rate affects the current compression decision. According to the optimization objective, the genetic algorithm in step S6 is used to obtain the optimal compression ratio and the optimal overall compression time when the data transmission rate changes dynamically.

Compare the overall transmission time consumption t ratio of the two, and take the compression ratio corresponding to the smaller time consumption as the optimal compression ratio.

After the optimal compression ratio is finally obtained, the power system data packets to be transmitted are compressed and transmitted. .

The invention is a transmission time-consuming optimization method based on predicting the transmission rate and adjusting the compression ratio, which can effectively reduce the time-consuming of data transmission in the power system. During the device adjustment period after the system task is started, the data transmission rate fluctuates more frequently. The scheme of improving the current compression ratio by predicting the data transmission rate can ensure the effective transmission of data. That is to say, when the network transmission rate changes more frequently In large cases, adjusting the current compression decision by predicting and analyzing its change trend can still ensure that the transmission status of the system monitoring data is not affected. In each transmission stage after the system task is started, the introduction of transmission rate prediction to improve the current optimization decision can control the data transmission time to be lower than the underlying acquisition period of each system, that is, the transmission state at this time can ensure the timeliness of monitoring data , accuracy, so that the upper-level client can display and feedback the operating status of the system in time, and the decision-making level can do further processing.

As will be appreciated by those skilled in the art, the embodiments of the present application may be provided as a method, a system, or a 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, etc.) 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 present application. It will be understood that each flow and/or block in the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing device produce Means for implementing the functions specified in a flow or flow of a flowchart and/or a block or blocks of a block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions The apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

These computer program instructions can also be loaded on a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process such that The instructions provide steps for implementing the functions specified in the flow or blocks of the flowcharts and/or the block or blocks of the block diagrams.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: the present invention can still be Modifications or equivalent replacements are made to the specific embodiments of the present invention, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention shall be included within the protection scope of the claims of the present invention.

Claims (2)

1. A dynamic compression prediction control method of continuous monitoring data in unstable network transmission comprises a data acquisition step S1, a modeling calculation compression ratio step S2, a compression transmission step S3 and a decompression step S4, and is characterized in that:

the step of S1 acquiring data acquires the current time period

Continuous data to be transmitted and network rate history samples; information system for a period of time

The internal accumulation of the data packets to be transmitted is less than the data accumulation period

The step of modeling and calculating the compression ratio of S2 comprises the following steps: s2-1, calculating the next time period by adopting a time series prediction method according to the network rate historical samples

The predicted network speed of (2); s2-2, modeling and solving the optimal compression ratio by taking the least time consumption as a target according to the predicted network speed;

step S2-1, calculating the next time period by adopting a time series prediction method according to the network rate historical samples

The network speed prediction method comprises the following specific steps:

s2-1, smoothness inspection of a historical network sequence: judging whether the time sequence is a stationarity time sequence according to the mean value, the variance and the correlation coefficient of the network sequence;

s2-2, sequence difference operation: if the sequence is a non-stationary time sequence, performing differential operation on the sequences until the sequence meets stationarity;

s2-3, model fitting of the sequence: according to the correlation coefficient ACF and the partial correlation coefficient PACF of the sample and the periodicity, carrying out summation autoregressive moving average model ARIMA (p, d, q) on the sequence after the difference and fitting to obtain a time sequence function d (t);

s2-4, and testing of residual sequences: carrying out white noise test on the residual sequence of the time sequence function d (t), and if the residual sequence does not meet the white noise sequence, returning to the step S2-3 to re-fit the model until the residual sequence is white noise;

s2-5, model optimization: performing model optimization on the time-setting time sequence fitting model function d (t) by adopting a minimum information criterion to obtain an updated model function d (t);

s2-6, prediction of sequence: fitting functions with time series

To predict the next time period

The network rate of (c);

the specific method for modeling and solving the optimal compression ratio based on the predicted net speed and the least time consumption as the target in the step S2-2 is as follows:

s2-2-1, calculating the time consumption of direct transmission of the original data in the predicted network speed environment

S2-2-2, establishing an optimization model taking the overall transmission time consumption as an optimal target, and combining the direct transmission time consumption calculated in the S2-2-1

Establishing a constraint condition;

s2-2-3, solving compression ratio and overall transmission time consumption t under the current network speed environment;

s2-2-4, solving compression ratio and overall transmission time consumption t under the network speed predicting environment;

s2-2-5, comparing the integral transmission time consumption t of S2-2-3 and S2-2-4, and selecting a compression ratio corresponding to the smaller integral transmission time consumption t as an optimal compression ratio;

the specific method for calculating the direct transmission time consumption of the original data in the network speed predicting environment in the step S2-2-1 is as follows:

judging whether the data transmission rate prediction sequence d (t) can be integrated or not, if so, according to a formula

Can directly calculate the original transmission time consumption

Q is the size of the scale of the data to be transmitted;

if the discrete sequence set d (t) cannot be integrated, the change of the speed of the continuous data in the sampling period △ t is small and can be ignored, namely the discrete sequence set d (t) is subjected to segmented integration

Solving n to obtain the original transmission time

In the formula, t_iRepresenting the sampling instant, D_iRepresenting the rate in the ith sampling period, and n represents the number of sampling time points;

step S2-2-2, establishing an optimization model with the overall transmission time consumption as an optimal target, and combining the direct transmission time consumption calculated in step S2-2-1

The specific method for establishing the constraint condition comprises the following steps:

the unit original data packet compression time t can be obtained by aiming at multiple tests₁The relation with the compression ratio p is t₁＝T₁(p) and unit original packet decompression time t₂The relation with the compression ratio p is t₂＝T₂(p); constructing an objective function

min t＝t₁+t*+t₂＝QT₁(p)+t*+QT₂(p)

Setting the maximum compression ratio and the minimum compression ratio to P, respectively_maxAnd P_minSetting a constraint condition:

in the formula (I), the compound is shown in the specification,

the original transmission of the uncompressed data is time consuming,

network speed sequence prediction period, t^*Q is the scale size of the data to be transmitted when the compressed data is time-consuming to transmit;

the step S2-2-3 and the step S2-2-4 adopt a genetic algorithm to solve the optimal compression ratio, and the specific method is as follows:

1. and (3) encoding: compression ratio range according to each mode is [ P ]_min,P_max]Using binary codes of length k, total 2^kDifferent codes are coded, adjacent codes are spaced by

2. Generation of an initial population: randomly generating N string structure data as an initial population to begin evolution, namely generating N initial compression ratios p which take binary as codes and serve as the initial population;

3. and (3) fitness evaluation: selecting a fitness function as t ═ t₁+t*+t₂＝QT₁(p)+t*+QT₂(p) calculating the fitness function value of each initial individual in the population;

4. and (3) natural selection: individuals of which fitness function values t meet constraint conditions and are smaller than the average population value are reserved as individuals with strong adaptability and added to a new population;

5. crossover and mutation: the crossing is to exchange partial codes between individuals, the variation is to randomly select an individual in the population and randomly change a certain character in the code with a small probability, and the purpose of the crossing and the variation is to obtain a new individual to be added into a new population;

6. whether to stop the evolution: termination conditions were as follows: if the evolution times reach the set times or the variance of the population fitness is smaller than the set variance, stopping the evolution if the termination condition is met, otherwise, returning the updated population to the step 3;

7. and (3) decoding: and converting the selected and left optimal individual into an original parameter through coding to serve as an optimal compression ratio, and obtaining optimal fitness at the same time.

2. The method as claimed in claim 1, wherein the time period is a period of time for dynamic compression prediction control of the continuous monitor data in the non-stable network transmission

Greater than the system data transmission time.

Images