Financial big data secure transmission method based on cloud computing

Abstract

The present invention relates to the technical field of predicting financial big data transmission, and specifically to a method for secure transmission of financial big data based on cloud computing. The method obtains the amount of financial data transmitted to the cloud at each moment within a preset time period; divides the amount of financial data within the preset time period into local sequences, and obtains the initial cycle according to the magnitude of the frequency corresponding to the amplitude in the frequency spectrum corresponding to the amount of financial data in the local sequence; selects high-quality cycles according to the appearance of the initial cycle in the local sequence; obtains the first regression cycle according to the difference between the moment corresponding to the peak of the high-quality cycle and the current moment, predicts the next peak of the amount of financial data at the current moment, and deploys cloud devices for financial big data transportation. The present invention improves the stability and security of financial big data transmission by accurately predicting the next peak of the amount of financial data at the current moment and deploying cloud devices.

Description

Technical Field

[0001] The present invention relates to the technical field of predictive financial big data transmission, and in particular to a method for secure transmission of financial big data based on cloud computing. Background Art

[0002] Financial big data is a kind of sensitive data. For example, securities trading data is a kind of large-scale data with sharp fluctuations. It needs to be stored securely to ensure the security and reliability of the data, and backup and recovery operations need to be performed at any time to prevent data loss or damage. The cloud provides large-scale data storage and backup capabilities, as well as powerful computing power and real-time data analysis tools, which enable rapid processing and analysis of financial big data, thereby timely discovering market trends and investment opportunities. In order to ensure that the cloud can accurately analyze financial big data, it is necessary to ensure that financial big data is accurately transmitted to the cloud.

[0003] In the existing method, financial big data is directly transmitted through cloud devices, but in reality, financial big data is unstable during the transportation process. When transported through fixed cloud devices, it is easy for some financial big data to fail to be transmitted in time, which in turn leads to the risk of loss or confusion of financial big data during transmission. It is impossible to accurately analyze financial big data, and thus it is impossible to discover market trends and investment opportunities in a timely and accurate manner. Summary of the invention

[0004] In order to solve the technical problem that financial big data is unstable during transportation, and when it is transported through fixed cloud devices, it is easy to cause some financial big data to fail to be transmitted in time, which in turn causes the risk of loss or confusion of financial big data during transmission, the purpose of the present invention is to provide a financial big data security transmission method based on cloud computing, and the technical scheme adopted is as follows:

[0005] The present invention provides a financial big data secure transmission method based on cloud computing, the method comprising the following steps:

[0006] Obtain the amount of financial data transmitted to the cloud at each moment within a preset time period;

[0007] The amount of financial data in a preset time period is divided into local sequences, and the initial period of each local sequence is obtained according to the amplitude corresponding to each frequency in the frequency spectrum corresponding to the amount of financial data in each local sequence;

[0008] According to the appearance of each initial cycle in each local sequence, high-quality cycles are screened out;

[0009] According to the difference between the time corresponding to the peak value of each high-quality period and the current time, the first regression period is obtained;

[0010] Based on the earliest regression cycle, the next financial data volume peak at the current moment is predicted, and cloud devices are deployed to transport financial big data.

[0011] Furthermore, the method for obtaining the local sequence is:

[0012] Obtain a frequency spectrum of the amount of financial data within a preset time period as an overall frequency spectrum;

[0013] The inverse of the frequency corresponding to the maximum amplitude in the overall spectrum is taken as the main period;

[0014] Divide the amount of financial data within a preset time period into local series by major cycles.

[0015] Furthermore, the method for obtaining the initial period is:

[0016] According to the difference between the amplitude corresponding to each frequency in the spectrum graph corresponding to the amount of financial data in each local sequence and the amplitude corresponding to other frequencies, the degree of isolated transcendence of each frequency is obtained;

[0017] Obtaining a target frequency in a frequency spectrum corresponding to the amount of financial data in each local sequence based on the degree of isolation transcendence;

[0018] The amount of financial data in each local sequence corresponds to the inverse of each target frequency in the spectrum diagram, which is used as the initial period of the corresponding local sequence.

[0019] Furthermore, the method for obtaining the isolation transcendence degree is:

[0020] For any frequency in the frequency spectrum corresponding to the amount of financial data in any local sequence, obtain the ratio of the amplitude corresponding to the frequency to the mean of all amplitudes in the frequency spectrum as the first eigenvalue;

[0021] Obtain the ratio of the amplitude corresponding to the frequency to the mean of the amplitude corresponding to the frequency of its neighboring regions as the second eigenvalue;

[0022] The isolation transcendence degree of the frequency is obtained according to the first eigenvalue and the second eigenvalue.

[0023] Furthermore, the method for acquiring the target frequency is:

[0024] For any local sequence, the financial data in the local sequence is clustered according to the isolation degree values ​​of all frequencies in the frequency spectrum to obtain clusters;

[0025] Get the mean of the isolation degree values ​​within each cluster as the cluster label of each cluster;

[0026] The frequencies corresponding to the isolation degree values ​​within the clusters corresponding to the maximum cluster labels are taken as the target frequencies in the frequency spectrum corresponding to the financial data volume in the local sequence.

[0027] Furthermore, the method for obtaining the high-quality cycle is:

[0028] According to the occurrence of each initial period in each local sequence, the universality of each initial period is obtained;

[0029] When the universality meets the preset universality range, the corresponding initial period is a high-quality period.

[0030] Furthermore, the method for obtaining the universality is:

[0031] For any initial period, obtain the average of all consecutive occurrences of the initial period between local sequences as the occurrence frequency value of the initial period;

[0032] Obtain the number of local sequences between two consecutive times when the initial period appears discontinuously between local sequences as the number of intervals;

[0033] The universality of the initial cycle is obtained according to the occurrence probability, the occurrence frequency value and the maximum number of intervals of the initial cycle in all local sequences.

[0034] Furthermore, the method for obtaining the first regression period is:

[0035] For any high-quality cycle, obtain the time corresponding to the peak of the high-quality cycle closest to the current time in time series as the first time of the high-quality cycle;

[0036] Determine, according to the first moment, a moment corresponding to the next peak of the high-quality period as the second moment;

[0037] The difference between the second moment and the current moment is taken as the regression duration of the high-quality cycle;

[0038] According to the regression time of each high-quality cycle, the earliest regression cycle is obtained.

[0039] Furthermore, the method for obtaining the second moment is:

[0040] For any high-quality cycle, the sum of the first moment of the high-quality cycle and the corresponding duration of the high-quality cycle is used as the second moment of the high-quality cycle.

[0041] Furthermore, the method for obtaining the first regression period is:

[0042] The high-quality period corresponding to the shortest regression duration is taken as the first regression period.

[0043] The present invention has the following beneficial effects:

[0044] Dividing the amount of financial data within a preset time period into local sequences is conducive to accurately and efficiently obtaining each cycle of the amount of financial data within the preset time period; in order to avoid the influence of abnormal and unstable financial data on the prediction of the next peak value of financial data at the current moment, the initial cycle of each local sequence is obtained according to the amplitude corresponding to each frequency in the spectrum diagram corresponding to the amount of financial data in each local sequence, and the period with stable distribution in each local sequence is determined; considering that there may be specific initial cycles in individual local sequences, which will lead to inaccurate predicted peak values ​​of financial data, high-quality cycles are screened out according to the appearance of each initial cycle in each local sequence, and each cycle with stable distribution within the preset time period is accurately analyzed, which is conducive to the subsequent accurate prediction of the next peak value of financial data at the current moment; according to the difference between the moment corresponding to the peak of each high-quality cycle and the current moment, the first regression cycle is accurately obtained, and then the next peak value of financial data at the current moment is accurately and timely predicted based on the first regression cycle, and the cloud equipment for transmitting financial data is accurately deployed in advance to ensure accurate and stable transmission of financial big data, avoid confusion and loss of financial big data, and ensure the integrity and security of financial big data, so as to accurately analyze financial big data and discover market trends and investment opportunities in a timely and accurate manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 A schematic diagram of a process flow of a financial big data secure transmission method based on cloud computing provided by an embodiment of the present invention;

[0047] Figure 2 A flow chart of a method for obtaining an initial period provided by an embodiment of the present invention;

[0048] Figure 3 A flow chart of a method for obtaining a high-quality cycle provided by an embodiment of the present invention;

[0049] Figure 4 A flowchart of a method for obtaining universality provided by an embodiment of the present invention;

[0050] Figure 5A diagram of a financial big data security transmission system based on cloud computing provided by an embodiment of the present invention;

[0051] Figure 6 A schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a financial big data security transmission method based on cloud computing proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0053] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0054] The following is a detailed description of a specific solution of a financial big data secure transmission method based on cloud computing provided by the present invention in conjunction with the accompanying drawings.

[0055] Embodiment 1:

[0056] This invention proposes a financial big data secure transmission method based on cloud computing. Figure 1 , which shows a flow chart of a method for secure transmission of financial big data based on cloud computing provided by an embodiment of the present invention, the method comprising the following steps:

[0057] Step S1: Obtain the amount of financial data transmitted to the cloud at each moment within a preset time period.

[0058] Specifically, the data transmission volume of financial big data in the securities market fluctuates dramatically. For example, the data transmission volume of financial big data will increase significantly during the time periods corresponding to the opening and closing of each day, week, month, quarter, and year. In order to ensure the stable transmission of the data transmission volume of financial big data, the number of cloud transmission devices also needs to increase significantly. In the time periods corresponding to non-opening and non-closing, the financial big data in the securities market is relatively stable, and the data transmission volume of financial big data is relatively stable and small. In order to ensure the accurate transmission of financial big data and save the transportation cost of cloud devices, it is necessary to predict the data transmission volume of financial big data, and then deploy and adjust the cloud devices in advance.

[0059] In order to accurately predict the data transmission volume of financial big data, it is necessary to accurately analyze each cycle of financial big data during the cloud transmission process, and then accurately predict the next data transmission volume peak of financial big data, and accurately deploy the cloud equipment for transmitting financial big data according to the predicted data transmission volume peak to ensure the accurate transmission of financial big data. This embodiment takes any securities trading company as an example for analysis. The implementer can take one or more securities trading companies as examples for analysis according to the actual situation, and it is not limited here. It should be noted that the securities trading companies that appear later are all securities trading companies selected in this embodiment. This embodiment needs to analyze the financial big data transmitted to the cloud by the securities trading company in history, in order to accurately predict the transmission of the financial big data of the securities trading company, which is conducive to adjusting and deploying the cloud equipment in advance, and then transmitting the financial data of the securities trading company safely and stably. Among them, the financial big data of the securities trading company is the securities trading data of the securities trading company. Therefore, this embodiment dynamically monitors the amount of financial data transmitted to the cloud at each moment within the preset time period of the securities trading company, that is, obtains the amount of financial data transmitted to the cloud at each moment within the preset time period, and then analyzes the transmission cycle of the financial data in the preset time period, which is conducive to the subsequent accurate prediction of the next financial data volume peak at the current moment, timely adjustment and deployment of cloud servers, and ensures that the financial big data of the securities trading company is accurately transmitted to the cloud. Among them, this embodiment sets the preset time period to one quarter, wherein the last moment of the preset time period is the current moment. The time interval between two adjacent moments is set to 1 second, and the implementer can set the preset time period and the time interval between two adjacent moments according to the actual situation, which is not limited here.

[0060] Step S2: Divide the amount of financial data within a preset time period into local sequences, and obtain the initial period of each local sequence according to the amplitude corresponding to each frequency in the frequency spectrum diagram corresponding to the amount of financial data in each local sequence.

[0061] Specifically, there are many cycles in the transmission process of financial big data. For example, there are many different cycles in the amount of financial data transmitted every day, every week, every month, every quarter and every year, and each cycle corresponds to a peak value of financial data. It is known that there are many financial data in the preset time period. In order to analyze the various cycles of the amount of financial data transmitted in the preset time period, and then accurately predict the next peak value of financial data at the current moment, timely adjust and deploy the cloud equipment for transmitting financial big data, and ensure the accurate transmission of financial big data, this embodiment first divides the amount of financial data in the preset time period into local sequences, and then analyzes the amount of financial data in each local sequence separately, so as to improve the accuracy of accurately obtaining different cycles in the preset time period. In order to accurately analyze the various cycles of the amount of financial data transmitted in each local sequence, this embodiment determines the various cycles of the amount of financial data transmitted in each local sequence by frequency domain analysis, and preliminarily determines the various cycles of the amount of financial data transmitted in the preset time period, so as to accurately predict the peak value of the amount of financial data to be transmitted in the subsequent period. Among them, the frequency domain is a well-known technology and will not be described in detail.

[0062] It is known that the amount of financial data transmitted within the preset time period is an overall continuous data set, so it is impossible to accurately divide the amount of financial data within the preset time period, which leads to inaccurate local sequence division. For example, conventional securities trading financial big data is affected by the trading activities and trading time of the securities market, and the amount of financial data transmitted has different change rules, that is, the amount of financial data transmitted is a function of multiple cycles in time series. In order to accurately divide the amount of financial data within the preset time period, and then accurately analyze the various cycles of the amount of financial data within the preset time period, this embodiment first fits the amount of financial data within the preset time period into a continuous function. Secondly, the sampling frequency of the continuous function is determined, and the continuous function is accurately mapped to the spectrum diagram. It is known that the amount of financial data is collected once per second. According to the Nyquist sampling theorem, the minimum sampling frequency needs to be greater than or equal to 2. This embodiment uses 2 as the sampling frequency. The implementer can set the size of the sampling frequency according to the actual situation, and it is not limited here. Among them, the spectrum diagram and the Nyquist sampling theorem are both well-known technologies and will not be repeated. Finally, the continuous function is sampled according to the sampling frequency to obtain the discrete data points of the continuous function, and then the frequency domain data is obtained and plotted into a spectrum diagram of the amount of financial data in a preset time period as an overall spectrum diagram. Among them, the horizontal axis of the overall spectrum diagram represents the frequency, and the vertical axis represents the amplitude. It is known that the larger the amplitude in the spectrum diagram, the more stable the corresponding period is in the continuous data. In order to accurately analyze each period of the amount of financial data in the preset time period, this embodiment uses the inverse of the frequency corresponding to the maximum amplitude in the overall spectrum diagram as the main period, which is the most stable period of the amount of financial data in the preset time period, and then the amount of financial data in the preset time period is taken as a whole, starting from the first amount of financial data in the preset time period, and the amount of financial data in the preset time period is divided into local sequences through the main period. Among them, adjacent local sequences are connected in time sequence. It should be noted that when the last sequence of the amount of financial data in the preset time period does not meet a main period, it is also defaulted to a local sequence.

[0063] In actual situations, each local sequence has a large time span and contains a large amount of financial data. Therefore, when predicting the next financial data volume peak at the current moment through the main cycle, there are inaccuracies. In order to more accurately predict the next financial data volume peak at the current moment, it is necessary to obtain and analyze smaller cycles within a large stable cycle. Therefore, this embodiment analyzes the amount of financial data in each local sequence and obtains an initial cycle with a smaller time span than the main cycle to prepare for predicting the next financial data volume peak at the current moment.

[0064] Preferably, in some possible implementation methods of this embodiment, the method for obtaining the initial period can be found in Figure 2, which shows a flow chart of a method for obtaining an initial period provided by an embodiment of the present invention, the method comprising:

[0065] Step S201: Obtain the degree of isolation transcendence of each frequency according to the difference between the amplitude corresponding to each frequency in the frequency spectrum corresponding to the amount of financial data in each local sequence and the amplitude corresponding to other frequencies.

[0066] First, according to the method of obtaining the overall spectrum, the spectrum corresponding to the amount of financial data in each local sequence is obtained. It is known that the larger the amplitude, the more obvious the corresponding period. Therefore, the amplitude corresponding to each frequency in the spectrum corresponding to the amount of financial data in each local sequence is analyzed. When the amplitude corresponding to a certain frequency is more prominent in the spectrum, the degree of isolation of the frequency is greater, the period corresponding to the frequency is more obvious in the corresponding local sequence, and the period corresponding to the frequency is more likely to be the initial period in the corresponding local sequence.

[0067] In some possible methods of this embodiment, the method for obtaining the degree of isolated transcendence is: for any frequency in the spectrum graph corresponding to the amount of financial data in any local sequence, obtain the ratio of the amplitude corresponding to the frequency to the mean of all amplitudes in the spectrum graph where it is located as the first eigenvalue; when the first eigenvalue is larger, the amplitude corresponding to the frequency is larger, the amplitude corresponding to the frequency is more prominent in the spectrum graph where it is located, the degree of isolated transcendence of the frequency is larger, and the corresponding period of the frequency is more obvious in the local sequence. In order to more accurately obtain the degree of isolated transcendence of the frequency, further obtain the ratio of the amplitude corresponding to the frequency to the mean of the amplitude corresponding to its neighboring frequencies as the second eigenvalue; wherein the neighboring frequencies of the frequency are the frequencies adjacent to the left and right of the frequency, that is, there are two neighboring frequencies of the frequency. It should be noted that if the frequency is the endpoint frequency in the spectrum graph, that is, there is only one neighboring frequency for the frequency, at this time, the ratio of the amplitude corresponding to the frequency and its neighboring frequencies is used as the second eigenvalue. When the second eigenvalue is larger, it means that the corresponding period of the frequency is more obvious, and the degree of isolation of the frequency is greater. Then, according to the first eigenvalue and the second eigenvalue, the degree of isolation of the frequency is obtained; wherein, the first eigenvalue and the second eigenvalue are both positively correlated with the degree of isolation. At this point, the degree of isolation of each frequency is obtained. In order to more accurately analyze the degree of obviousness of the corresponding period of each frequency, this embodiment specifically quantifies the degree of isolation of each frequency to obtain the value of the degree of isolation of each frequency. The larger the value of the degree of isolation, the more obvious and stable the period corresponding to the corresponding frequency. Among them, the expression of the value of the degree of isolation is: ; In the formula, The amount of financial data in the ath local sequence corresponds to the ath in the spectrum graph. The isolated exceedance value of each frequency; The amount of financial data in the ath local sequence corresponds to the ath in the spectrum graph. The amplitude corresponding to the frequency; is the mean value of all amplitudes in the spectrum corresponding to the amount of financial data in the a-th local sequence; The amount of financial data in the ath local sequence corresponds to the ath in the spectrum graph. The amplitude corresponding to the frequency; The amount of financial data in the ath local sequence corresponds to the ath in the spectrum graph. The amplitude corresponding to the frequency; is the first eigenvalue; is the second eigenvalue.

[0068] Step S202: obtaining the target frequency in the frequency spectrum corresponding to the amount of financial data in each local sequence based on the degree of isolation transcendence.

[0069] There are multiple periods in each local sequence at the same time. In actual situations, there may be abnormal financial data in the local sequence, and the abnormal financial data in the corresponding frequency spectrum may also have corresponding frequencies and amplitudes, but the amplitude corresponding to the abnormal financial data must be small, because the abnormal financial data is unstable in the local sequence. Therefore, this embodiment screens the frequencies in the frequency spectrum corresponding to the financial data in each local sequence based on the degree of isolation transcendence, and determines the target frequency in the frequency spectrum corresponding to the financial data in each local sequence, wherein the period corresponding to the target frequency is stably and obviously distributed in the local sequence.

[0070] In some possible implementation methods of this embodiment, the method for obtaining the target frequency is: for any local sequence, the isolation exceedance values ​​of all frequencies in the frequency spectrum corresponding to the amount of financial data in the local sequence are clustered to obtain cluster clusters; this embodiment clusters the isolation exceedance values ​​of all frequencies in the frequency spectrum corresponding to the amount of financial data in the local sequence through an adaptive K-means clustering algorithm. Among them, the adaptive K-means clustering algorithm is a well-known technology and will not be described in detail. The mean of the isolation exceedance values ​​in each cluster is obtained as the cluster label of each cluster; the frequencies corresponding to the isolation exceedance values ​​in the cluster corresponding to the maximum cluster label are all used as the target frequencies in the frequency spectrum corresponding to the amount of financial data in the local sequence. At this point, the target frequency in the frequency spectrum of the amount of financial data corresponding to each local sequence is determined.

[0071] Step S203: The inverse of each target frequency in the frequency spectrum corresponding to the amount of financial data in each local sequence is used as the initial period of the corresponding local sequence.

[0072] It is known that the period corresponding to the target frequency is relatively stably distributed in the local sequence. Therefore, in this embodiment, the inverse of each target frequency in the spectrum diagram corresponding to the amount of financial data in each local sequence is used as the initial period of the corresponding local sequence, eliminating the influence of unstable periods, which is conducive to the subsequent accurate prediction of the next peak value of the amount of financial data at the current moment.

[0073] Step S3: Filter out high-quality cycles according to the occurrence of each initial cycle in each local sequence.

[0074] In actual situations, some initial cycles only appear in specific local sequences and are not obvious in the entire preset time period. For example, in the trading activities of the securities market, some specific initial cycles will appear in the time period when the trading activities continue. Because the trading activities are not always continuous, the specific initial cycles caused by the trading activities do not exist in other time periods. The cycle corresponding to the next peak value of the amount of financial data at the current moment must be a cycle that is obviously distributed in the entire preset time period to avoid deviations caused by chance, which leads to inaccurate transmission of financial big data. Therefore, this embodiment obtains the universality of each initial cycle based on the appearance of each initial cycle in each local sequence, where the greater the universality, the more obvious the distribution of the corresponding initial cycle in the preset time period, and then selects high-quality cycles with stable distribution in the preset time period according to the universality.

[0075] Preferably, in some possible implementation methods of this embodiment, the method for obtaining the high-quality cycle can be found in Figure 3 , which shows a flow chart of a method for obtaining a high-quality cycle provided by an embodiment of the present invention, the method comprising:

[0076] Step S301: Obtain the universality of each initial period according to the occurrence of each initial period in each local sequence.

[0077] When a certain initial period appears more times in a local sequence, the initial period is more obviously distributed in a preset time period. Therefore, this embodiment obtains the universality of each initial period according to the appearance of each initial period in each local sequence.

[0078] Preferably, in some possible implementation methods of this embodiment, the method for obtaining the universality degree can be found in Figure 4 , which shows a flow chart of a method for obtaining universality provided by an embodiment of the present invention, the method comprising:

[0079] Step S301 - 1 : For any initial period, obtain the average of all consecutive occurrences of the initial period between local sequences as the occurrence frequency value of the initial period.

[0080] As an example, taking the kth initial period as an example, assume that the preset time period is divided into 20 continuous local sequences, and the numbers are numbered from 1, 2, 3, ..., 20. When the kth initial period appears in the local sequences numbered 1, 2, 3, 4, 5, 6, 9, 10, 11, 15, 16, 17, 18, 20, the first consecutive occurrence of the kth initial period is 6, corresponding to the local sequences numbered 1-6; the second consecutive occurrence of the kth initial period is 3, corresponding to the local sequences numbered 9-11; the third consecutive occurrence of the kth initial period is 4, corresponding to the local sequences numbered 15-18; the fourth consecutive occurrence of the kth initial period is 1, corresponding to the local sequence numbered 20. Therefore, the frequency of occurrence of the kth initial period is The larger the frequency value, the more obvious the kth initial period is in the preset time period. It should be noted that the frequency value is expressed as a score.

[0081] Step S301 - 2 : obtaining the number of local sequences between two consecutive times when the initial period appears discontinuously between local sequences, as the interval number.

[0082] According to the local sequence appearing in the kth initial period in step S301-1, the number of intervals corresponding to the kth initial period is 2, 3, and 1. Among them, the number of intervals 2 corresponds to the local sequences numbered 7 and 8, the number of intervals 3 corresponds to the local sequences numbered 12, 13, and 14, and the number of intervals 1 corresponds to the local sequence numbered 19. Among them, the smaller the number of intervals, the more stable the distribution of the kth initial period within the preset time period.

[0083] Step S301 - 3 : Obtain the universality of the initial cycle according to the occurrence probability, occurrence frequency value and maximum number of intervals of the initial cycle in all local sequences.

[0084] As an example, taking the kth initial period in step S301-1 as an example, the ratio of the number of all local sequences in which the kth initial period appears to the total number of all local sequences is obtained as the probability of occurrence of the kth initial period. The greater the frequency of occurrence, the more stable the kth initial period is within the preset time period. When the maximum number of intervals corresponding to the kth initial period is smaller, it also means that the kth initial period is more stable within the preset time period. Therefore, this embodiment obtains the universality of the kth initial period according to the probability of occurrence, the frequency of occurrence value and the maximum number of intervals of the kth initial period. Among them, the probability of occurrence and the frequency of occurrence value are both positively correlated with the universality, and the maximum number of intervals is negatively correlated with the universality. So far, the universality of each initial period is obtained. Among them, the greater the universality, the more stable the corresponding initial period is within the preset time period. In order to more accurately analyze the distribution of each initial period within the preset time period, this embodiment specifically quantifies the universality of each initial period and obtains the universality value of each initial period, wherein the greater the universality value, the more stable the distribution of the corresponding initial period is within the preset time period. The expression of universality value is: ; In the formula, is the universality value of the kth initial period of the ath local sequence; is the occurrence probability of the kth initial period of the ath local sequence; is the occurrence frequency value of the kth initial period of the ath local sequence; is the maximum number of intervals corresponding to the kth initial period of the ath local sequence; is the first preset constant, which is greater than 0. Set to 1 to avoid the denominator being 0. Implementers can set it according to actual conditions. The size is not limited here.

[0085] Step S302: When the universality meets the preset universality range, the corresponding initial period is a high-quality period.

[0086] It is known that the larger the universality value is, the more stable the distribution of the corresponding initial cycle is within the preset time period. Therefore, in this embodiment, the universality value of each initial cycle is first normalized to obtain the normalized universality value of each initial cycle. Among them, the normalization method is a well-known technology and will not be described in detail. When the normalized universality value is greater than the preset universality value threshold, the corresponding initial cycle is regarded as a high-quality cycle and participates in the subsequent prediction of the peak value of the financial data volume of the next cycle at the current moment; when the normalized universality value is less than or equal to the preset universality value threshold, the corresponding initial cycle is no longer analyzed, that is, the distribution of the corresponding initial cycle is unstable within the preset time period, which will affect the prediction of the next financial data volume peak at the current moment, resulting in inaccurate predicted financial data volume peak. In this embodiment, the preset universality value threshold is set to 0.7. The implementer can set the size of the preset universality value threshold according to the actual situation, which is not limited here. At this point, the high-quality cycle within the preset time period is determined.

[0087] Step S4: Obtain the first regression period according to the difference between the time corresponding to the peak of each high-quality period and the current time.

[0088] Specifically, by analyzing the time corresponding to the peak of each high-quality cycle closest to the current time, the time corresponding to the next peak of each high-quality cycle can be accurately predicted. When the time corresponding to the predicted next peak of a certain high-quality cycle is closer to the current time, it means that the peak of the high-quality cycle is more likely to be the next financial data volume peak predicted at the current time, and the corresponding high-quality cycle is the first regression cycle.

[0089] Preferably, in some possible implementation methods of this embodiment, the method for obtaining the first return period is: for any high-quality period, obtain the moment corresponding to the peak of the high-quality period that is closest to the current moment in time sequence as the first moment of the high-quality period; determine the moment corresponding to the next peak of the high-quality period according to the first moment as the second moment, that is, add the first moment of the high-quality period and the corresponding duration of the high-quality period as the second moment of the high-quality period. The closer the second moment is to the current moment, the more likely the high-quality period is to be the first return period. Then obtain the difference between the second moment and the current moment as the return duration of the high-quality period; the smaller the return duration, the more likely the high-quality period is to be the first return period. Therefore, this embodiment uses the high-quality period corresponding to the smallest return duration as the first return period. So far, the first return period of the current moment is determined.

[0090] Step S5: predict the next financial data volume peak at the current moment based on the first regression cycle, and deploy cloud devices to transport financial big data.

[0091] Specifically, in actual situations, the peak data of the first regression cycle within the set time period has certain fluctuations. In order to more accurately predict the next financial data volume peak at the current moment, this embodiment inputs all the peak data that appear in the first regression cycle within the preset time period into the time series prediction algorithm ARIMA, and predicts the next financial data volume peak of the first regression cycle, that is, the predicted next financial data volume peak at the current moment. Among them, the time series prediction algorithm ARIMA is a well-known technology and will not be described in detail. Then, according to the predicted next financial data volume peak at the current moment, the cloud device is deployed and adjusted.

[0092] When the predicted next financial data volume peak at the current moment is not the financial data volume peak of the first regression cycle, the original deployment strategy of the cloud device does not change.

[0093] When the next predicted financial data volume peak at the current moment is the financial data volume peak of the first return cycle, the cloud device is deployed and adjusted in advance. The specific method is as follows:

[0094] (1) Taking the predicted next financial data volume peak at the current moment and the current cloud device deployment strategy as input, the congestion control algorithm is used to obtain the stability parameters of the current cloud device transmitting financial big data. The congestion control algorithm is a well-known technology and will not be described in detail.

[0095] (2) The stability parameters are judged by using the AHP (Analytic Hierarchy Process) algorithm to determine whether the deployment strategy of the cloud device corresponding to the stability parameter is suitable for the transmission of the next peak value of financial data volume at the current moment, and obtain the judgment result. The AHP (Analytic Hierarchy Process) algorithm is a well-known technology and will not be described in detail.

[0096] (3) Dynamically adjust the deployment strategy of cloud devices based on the judgment results. The specific adjustment methods are as follows:

[0097] If the judgment result is adaptation, the deployment strategy of the current cloud device will not be adjusted;

[0098] If the judgment result is not suitable, the server of the cloud device is adjusted through horizontal expansion or vertical expansion. Horizontal expansion means adding servers based on the current number of servers, which is used to alleviate the problem of excessive server load or bandwidth utilization approaching the upper limit due to the increase in financial data volume. In this embodiment, the number of servers added each time is set to 1. The implementer can set the number of servers added each time according to the actual situation, which is not limited here; vertical expansion means adjusting the type or configuration of the server on the service provider's management console, such as increasing the number of CPU cores, memory capacity or storage space.

[0099] (4) The deployment strategy of the cloud device after dynamic adjustment is used as the deployment strategy of the current cloud device, and the predicted next financial data volume peak value at the current moment is used as the new input, and steps (1) to (3) are repeated until the optimal result is obtained, and the adjustment of the deployment strategy of the cloud device is stopped. The optimal result means that after adjusting the deployment strategy of the cloud device multiple times (or once), it can accurately and stably transmit the predicted next financial data volume peak value at the current moment.

[0100] In summary, this embodiment obtains the amount of financial data transmitted to the cloud at each moment within a preset time period; divides the amount of financial data within the preset time period into local sequences, and obtains the initial cycle according to the amplitude of the frequency corresponding to the frequency spectrum of the amount of financial data in the local sequence; selects high-quality cycles according to the appearance of the initial cycle in the local sequence; obtains the first regression cycle according to the difference between the moment corresponding to the peak of the high-quality cycle and the current moment, predicts the next peak of the amount of financial data at the current moment, and deploys cloud devices for financial big data transportation. The present invention improves the stability and accuracy of financial big data transmission by accurately predicting the next peak of the amount of financial data at the current moment and deploying cloud devices.

[0101] Embodiment 2:

[0102] The present invention also proposes a financial big data security transmission system based on cloud computing, please refer to Figure 5 , which shows a structure diagram of a financial big data security transmission system based on cloud computing provided by an embodiment of the present invention. The system includes: a data acquisition module 10, an initial cycle acquisition module 20, a high-quality cycle acquisition module 30, a first regression cycle acquisition module 40 and a data processing module 50.

[0103] The data acquisition module 10 is used to acquire the amount of financial data transmitted to the cloud at each moment within a preset time period.

[0104] The initial period acquisition module 20 is used to divide the amount of financial data in a preset time period into local sequences, and acquire the initial period of each local sequence according to the amplitude corresponding to each frequency in the spectrum diagram corresponding to the amount of financial data in each local sequence.

[0105] The high-quality cycle acquisition module 30 is used to screen out high-quality cycles according to the appearance of each initial cycle in each local sequence.

[0106] The first regression period acquisition module 40 is used to acquire the first regression period according to the difference between the time corresponding to the peak of each high-quality period and the current time.

[0107] The data processing module 50 is used to predict the next financial data volume peak at the current moment based on the first regression cycle, and deploy cloud devices to transport financial big data.

[0108] It should be noted that the system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the financial big data security transmission system based on cloud computing and the financial big data security transmission method based on cloud computing provided in the above embodiment belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0109] Embodiment 3:

[0110] The present invention also proposes a financial big data security transmission device based on cloud computing, which includes a memory and a processor, wherein the memory stores an executable program code, and the processor is used to call and execute the executable program code to execute a financial big data security transmission method based on cloud computing provided in an embodiment of the present application. The device can be a chip, a component or a module, and the chip may include a connected processor and a memory; wherein the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute a financial big data security transmission method based on cloud computing provided in the above embodiment.

[0111] In addition, the present application embodiment also protects a computer device, see Figure 6 The computer device includes a memory 401, a processor 402, and a computer program 403 stored in the memory 401 and running on the processor 402, wherein when the processor 402 executes the computer program 403, the computer device can execute any one of the cloud computing-based financial big data security transmission methods introduced above.

[0112] Embodiment 4:

[0113] This embodiment also provides a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer, the computer executes the above-mentioned related method steps to implement a cloud computing-based financial big data security transmission method provided in the above embodiment.

[0114] Embodiment 5:

[0115] This embodiment also provides a computer program product. When the computer program product runs on a computer, it enables the computer to execute the above-mentioned related steps to implement a cloud computing-based financial big data security transmission method provided in the above embodiment.

[0116] Among them, the device, computer-readable storage medium, computer program product or chip provided in this embodiment is used to execute the corresponding method provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method provided above, and will not be repeated here.

[0117] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0118] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A financial big data secure transmission method based on cloud computing, characterized in that: The method comprises the following steps: Obtain the amount of financial data transmitted to the cloud at each moment within a preset time period; The amount of financial data in a preset time period is divided into local sequences, and the initial period of each local sequence is obtained according to the amplitude corresponding to each frequency in the frequency spectrum corresponding to the amount of financial data in each local sequence; According to the appearance of each initial cycle in each local sequence, high-quality cycles are screened out; According to the difference between the time corresponding to the peak value of each high-quality period and the current time, the first regression period is obtained; Based on the first regression cycle, the next financial data volume peak at the current moment is predicted, and cloud devices are deployed to transport financial big data; The method for obtaining the initial period is: According to the difference between the amplitude corresponding to each frequency in the spectrum graph corresponding to the amount of financial data in each local sequence and the amplitude corresponding to other frequencies, the degree of isolated transcendence of each frequency is obtained; Obtaining a target frequency in a frequency spectrum corresponding to the amount of financial data in each local sequence based on the degree of isolation transcendence; The inverse of each target frequency in the frequency spectrum corresponding to the amount of financial data in each local sequence is used as the initial period of the corresponding local sequence; The method for obtaining the isolation transcendence degree is: For any frequency in the frequency spectrum corresponding to the amount of financial data in any local sequence, obtain the ratio of the amplitude corresponding to the frequency to the mean of all amplitudes in the frequency spectrum as the first eigenvalue; Obtain the ratio of the amplitude corresponding to the frequency to the mean of the amplitude corresponding to the frequency of its neighboring regions as the second eigenvalue; Obtaining the degree of isolated transcendence of the frequency according to the first eigenvalue and the second eigenvalue; The expression of the isolation transcendence degree value is: ; In the formula, The amount of financial data in the ath local sequence corresponds to the ath in the spectrum graph. The isolated exceedance value of each frequency; The amount of financial data in the ath local sequence corresponds to the ath in the spectrum graph. The amplitude corresponding to the frequency; is the mean value of all amplitudes in the spectrum graph corresponding to the amount of financial data in the a-th local sequence; The amount of financial data in the ath local sequence corresponds to the ath in the spectrum graph. The amplitude corresponding to the frequency; The amount of financial data in the ath local sequence corresponds to the ath in the spectrum graph. The amplitude corresponding to the frequency; is the first eigenvalue; is the second eigenvalue; The method for obtaining the target frequency is: For any local sequence, the financial data in the local sequence is clustered according to the isolation degree values ​​of all frequencies in the frequency spectrum to obtain clusters; Get the mean of the isolation degree values ​​within each cluster as the cluster label of each cluster; The frequencies corresponding to the isolation degree values ​​in the clusters corresponding to the maximum cluster labels are all used as the target frequencies in the frequency spectrum corresponding to the financial data in the local sequence; The method for obtaining the high-quality cycle is: According to the occurrence of each initial period in each local sequence, the universality of each initial period is obtained; When the universality meets the preset universality range, the corresponding initial period is a high-quality period; The method for obtaining the universality is: For any initial period, obtain the average of all consecutive occurrences of the initial period between local sequences as the occurrence frequency value of the initial period; Obtain the number of local sequences between two consecutive times when the initial period appears discontinuously between local sequences as the number of intervals; According to the occurrence probability, occurrence frequency value and maximum number of intervals of the initial cycle in all local sequences, the universality of the initial cycle is obtained; The expression of the universality value is: ; In the formula, is the universality value of the kth initial period of the ath local sequence; is the occurrence probability of the kth initial period of the ath local sequence; is the occurrence frequency value of the kth initial period of the ath local sequence; is the maximum number of intervals corresponding to the kth initial period of the ath local sequence; It is the first preset constant, which is greater than 0.

2. A method for secure transmission of financial big data based on cloud computing as claimed in claim 1, characterized in that: The method for obtaining the local sequence is: Obtain a frequency spectrum of the amount of financial data within a preset time period as an overall frequency spectrum; The inverse of the frequency corresponding to the maximum amplitude in the overall spectrum is taken as the main period; Divide the amount of financial data within a preset time period into local series by major cycles.

3. The method for secure transmission of financial big data based on cloud computing as claimed in claim 1, characterized in that: The method for obtaining the first regression period is: For any high-quality cycle, obtain the time corresponding to the peak of the high-quality cycle closest to the current time in time series as the first time of the high-quality cycle; Determine, according to the first moment, a moment corresponding to the next peak of the high-quality cycle as the second moment; The difference between the second moment and the current moment is taken as the regression duration of the high-quality cycle; According to the regression time of each high-quality cycle, the earliest regression cycle is obtained.

4. A method for secure transmission of financial big data based on cloud computing as claimed in claim 3, characterized in that: The method for obtaining the second moment is: For any high-quality cycle, the sum of the first moment of the high-quality cycle and the corresponding duration of the high-quality cycle is used as the second moment of the high-quality cycle.

5. A method for secure transmission of financial big data based on cloud computing as claimed in claim 3, characterized in that: The method for obtaining the first regression period is: The high-quality period corresponding to the shortest regression duration is taken as the first regression period.

Images