Data transmission optimization method, system and equipment based on optical fiber communication and medium

Abstract

The invention discloses a data transmission optimization method, system and device based on optical fiber communication and a medium, and relates to the technical field of data processing, and the method comprises the steps: obtaining an optical fiber communication line and a monitoring time slot, and obtaining the number n for monitoring according to the optical fiber communication line; acquiring unit time slot average flows of the first one, two,..., n monitoring time slots at the current time point in real time, acquiring a time sequence weight corresponding to each unit time slot average flow based on a time sequence model, and acquiring a fluctuation index based on a fluctuation model, the n unit time slot average flows and the time sequence weight corresponding to each unit time slot average flow; judging whether the fluctuation index exceeds a preset fluctuation threshold value, if so, obtaining a bandwidth adjustment increment of the optical fiber communication line according to the fluctuation index, and obtaining an adjustment cost index according to the bandwidth adjustment increment; and completing bandwidth optimization according to the adjustment cost index and the bandwidth adjustment increment. The method has the advantages of fast dynamic response, accurate transmission bandwidth matching and collaborative optimization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a data transmission optimization method, system, device and medium based on optical fiber communication. BACKGROUND

[0002] With the popularity of FTTRB (fiber to the room / building), high-bandwidth and low-latency indoor optical fiber networks have become the key infrastructure to support 4K / 8K ultra-high-definition video, cloud gaming, virtual reality and other time-sensitive services.

[0003] However, the existing FTTRB system exposes significant technical shortcomings in dynamic traffic scenarios. First, the traditional bandwidth allocation mechanism is based on fixed cycle polling or static threshold triggering adjustment, which is difficult to cope with traffic patterns with strong burst characteristics. When multiple high-code rate services (such as concurrent transmission of multiple 4K video streams and cloud gaming data streams) produce superposition effects in the time domain, the instantaneous traffic peak can reach 2-3 times the average load, causing rapid accumulation of transmission queues and increasing video stuttering rate, which degrades user experience. Second, the existing traffic prediction model generally uses linear weighted average or simple sliding window algorithm, which fails to effectively capture the non-stationary characteristics of traffic sequences, especially in the prediction error during the start / termination of service streams, resulting in a serious mismatch between pre-allocated bandwidth and actual demand. This mismatch not only wastes bandwidth resources, but also causes frequent bandwidth adjustment operations, further exacerbating transmission jitter. Third, the existing technology evaluates the adjustment cost of bandwidth in a very rough way, usually only considering the instantaneous adjustment amplitude, while ignoring the cumulative impact of historical adjustment trajectories on system stability and the trend of bandwidth changes. More seriously, the existing bandwidth adjustment strategy lacks the ability to optimize multiple targets in the adjustment process, often sacrificing energy efficiency while reducing latency. SUMMARY

[0004] In view of the defects in the prior art, the present application provides a data transmission optimization method, system, device and medium based on optical fiber communication.

[0005] A data transmission optimization method based on optical fiber communication, comprising: acquiring an optical fiber communication line and a monitoring time slot, and obtaining a number n of monitoring from the optical fiber communication line; real-time acquiring the average traffic of each unit time slot before the current time point, and obtaining the time sequence weight corresponding to each unit time slot average traffic based on the time sequence model, and obtaining the fluctuation index based on the fluctuation model, the n unit time slot average traffic and the time sequence weight corresponding to each unit time slot average traffic; determining whether the fluctuation index exceeds the preset fluctuation threshold, if it exceeds, obtaining the bandwidth adjustment increment of the optical fiber communication line according to the fluctuation index, and obtaining the adjustment cost index according to the bandwidth adjustment increment; completing bandwidth optimization according to the adjustment cost index and the bandwidth adjustment increment.

[0006] Optionally, the bandwidth optimization according to the adjustment cost index and the bandwidth adjustment increment comprises: obtaining a preset cost threshold, and determining a correction value according to the preset cost threshold and the adjustment cost index; obtaining a target adjustment amount according to the correction value and the bandwidth adjustment increment; and completing the bandwidth optimization according to the target adjustment amount.

[0007] Optionally, the correction value determined according to the preset cost threshold and the adjustment cost index is represented as: ; wherein, the correction value is, the adjustment cost index is, the preset cost threshold is, the first scaling coefficient is.

[0008] Optionally, the time sequence model for obtaining the time sequence weight corresponding to the average traffic of each unit time slot is represented as: ; wherein, the time sequence weight corresponding to the average traffic of each unit time slot is, the dynamic base value is, the number of unit time slot average traffic is.

[0009] Optionally, the fluctuation model for obtaining the fluctuation index based on the fluctuation model, the n unit time slot average traffic and the time sequence weight corresponding to the average traffic of each unit time slot is represented as: ; wherein, the fluctuation index is, the time sequence weight corresponding to the average traffic of each unit time slot is, the average traffic of each unit time slot is, the average traffic of each unit time slot is, the number of unit time slot average traffic is.

[0010] Optionally, the bandwidth adjustment increment of the optical fiber communication line according to the fluctuation index is represented as: ; wherein, the bandwidth adjustment increment is, the current bandwidth is, the second scaling coefficient is, the preset fluctuation threshold is, the fluctuation index is.

[0011] Optionally, the adjustment cost index according to the bandwidth adjustment increment is represented as: ; wherein, the adjustment cost index is, the bandwidth adjustment increment is, the current bandwidth is.

[0012] The application further provides a data transmission optimization system based on optical fiber communication, characterized in that the system comprises: an acquisition module configured to acquire an optical fiber communication line and a monitoring time slot, and acquire a number n of monitoring time slots based on the optical fiber communication line; a data processing module configured to acquire unit time slot average flows of 1, 2, …, n monitoring time slots before a current time point in real time, acquire time sequence weights corresponding to each unit time slot average flow based on a time sequence model, and acquire a fluctuation index based on a fluctuation model, the n unit time slot average flows and the time sequence weights corresponding to each unit time slot average flow; a judgment module configured to judge whether the fluctuation index exceeds a preset fluctuation threshold, and if so, acquire a bandwidth adjustment increment of the optical fiber communication line based on the fluctuation index, and acquire an adjustment cost index based on the bandwidth adjustment increment; and a transmission line optimization module configured to complete bandwidth optimization based on the adjustment cost index and the bandwidth adjustment increment.

[0013] The application further provides an electronic device, comprising: a memory having a computer program stored thereon; and a processor configured to execute the computer program in the memory to implement the data transmission optimization method based on optical fiber communication.

[0014] The application further provides a non-transitory computer readable storage medium having a computer program stored thereon, the program being executed by a processor to implement the data transmission optimization method based on optical fiber communication.

[0015] The application has the following beneficial effects: In the entire data transmission optimization method based on optical fiber communication, first, the monitoring granularity based on dynamic adaptation of service characteristics breaks through the response bottleneck of traditional fixed cycle polling, and by capturing traffic mutations in each scenario in real time, combining exponential decay weighting of historical data with gradient nonlinear amplification of the fluctuation model, a traffic prediction system with short-term sensitivity and long-term stability is constructed, so that traffic transition trends can be identified in advance and pre-expansion is started, effectively reducing the queue accumulation delay caused by prediction lag in existing solutions; further, a cooperative decision mechanism of dynamic threshold triggering and cost perception is introduced, and by adaptive fluctuation threshold of service priority, adjustment cost constraint and implicit memory of historical adjustment trajectory, a three-level stabilization protection is constructed, when multiple bandwidth demand surges are detected, bandwidth expansion can be quickly completed, and by adjusting the cost constraint, the amplitude of expansion operation is inhibited, so that the transmission jitter coefficient is within a normal range. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the specific embodiments of the application or the technical solutions in the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or prior art description. In all the drawings, similar elements or parts are generally identified by similar reference signs. In the drawings, the elements or parts are not necessarily drawn according to the actual proportions.

[0017] Figure 1 A flow chart of steps of the data transmission optimization method based on optical fiber communication of the present application; Figure 2 A flow chart of part of the steps of S4 in the data transmission optimization method based on optical fiber communication of the present application; Figure 3 A block diagram of an electronic device according to an embodiment of the present application.

[0018] Reference signs: 700 - electronic device, 701 - processor, 702 - memory, 703 - multimedia component, 704 - I / O interface, 705 - communication component. DETAILED DESCRIPTION

[0019] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations.

[0020] Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative labor are within the scope of protection of the present application.

[0021] It should be noted that: similar reference signs and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. In addition, the terms "first", "second", etc. are only used to distinguish description, and cannot be understood as indicating or implying relative importance.

[0022] As Figure 1 shown, a data transmission optimization method based on optical fiber communication is provided, comprising: S1, acquiring an optical fiber communication line and a monitoring time slot, and acquiring a number n for monitoring according to the optical fiber communication line; S2, acquiring unit time slot average flows of 1st, 2nd, …, nth monitoring time slots before a current time point in real time, acquiring time sequence weights corresponding to each unit time slot average flow based on a time sequence model, and acquiring a fluctuation index based on a fluctuation model, the n unit time slot average flows and the time sequence weights corresponding to each unit time slot average flow; S3, judging whether the fluctuation index exceeds a preset fluctuation threshold, if yes, obtaining a bandwidth adjustment increment of the optical fiber communication line according to the fluctuation index, and obtaining an adjustment cost index according to the bandwidth adjustment increment; S4, completing bandwidth optimization according to the adjustment cost index and the bandwidth adjustment increment.

[0023] In the embodiment, it should be noted that in S1, a basic parameter framework is established for subsequent dynamic bandwidth optimization, the key of which is to determine the monitoring granularity by combining the physical network topology and the service characteristics. First, the target lines to be optimized in the current FTTRB network need to be identified, which are usually composed of physical connections between optical network units (ONUs) and main optical terminals, and the selection criteria can be based on historical traffic peaks, user density or service type priority. The setting of the monitoring time slot is essentially to discretize continuous time into observable dynamic windows, and the length needs to balance the real-time performance and the computational overhead, for example, for a line carrying 8K video stream, the time slot can be set to millisecond level to capture frame-level traffic jitter, while a second-level time slot can be used for ordinary data service lines. At this time, the determination of the monitoring number n needs to consider the line transmission capacity and the service burst period, for example, for a backbone line supporting concurrent multi-path cloud gaming, the value of n will be dynamically expanded to cover the typical traffic surge period of the game loading stage.

[0024] In specific implementation, the value of n reflects the depth of memory of historical traffic patterns. Taking the smart home scenario as an example, when the FTTRB detects that a line is connected to a smart gateway supporting VR devices, it will automatically shorten the monitoring time slot according to the maximum rendering interval of the device, while increasing the value of n to capture the burst pulse characteristics unique to VR motion capture data. Conversely, for a line carrying only smart home control signals, n can be set to a smaller value to avoid resource waste caused by excessive monitoring. This dynamic adaptation mechanism can not only identify short-term traffic spikes (such as multiple devices starting 4K video conferences at the same time), but also avoid false positives caused by excessive attention to transient fluctuations, essentially establishing an adaptive balance between time series observation dimensions and computational complexity.

[0025] In S2, the non-steady-state characteristics of network traffic are dynamically captured through a multi-dimensional traffic feature fusion mechanism. First, the historical traffic data is differentially weighted based on a time series model, the core of which is to adaptively adjust the weight distribution of each time slot according to the traffic fluctuation intensity. When a traffic mutation is detected (such as the cloud game loading stage), the weight proportion of recent time slots is automatically enhanced, and an exponential weight decay curve is used to make the model focus more on the short-term traffic surge characteristics. At the same time, a dynamic base value is used to adjust the decay rate in real time - when the business flow appears frequently, the base value will decrease with the increase of traffic variance, so as to compress the memory window of historical data and ensure the rapid response to sudden traffic patterns. This design can not only avoid the prediction lag caused by the static sliding window, but also prevent misjudgment caused by excessive attention to instantaneous noise.

[0026] Further, combined with the fluctuation model, the traffic mutation intensity is quantified through the nonlinear mapping of the traffic gradient of adjacent time slots. The model uses a composite function structure to couple the time series weight and the traffic difference of adjacent time slots: for the traffic jump of high weight period (such as the continuous traffic jump caused by the simultaneous start of multiple 8K video streams), the model enhances its contribution through an exponential amplification mechanism; while for the early stable fluctuation with lower weight, it is naturally attenuated. This design enables the fluctuation index to accurately distinguish between normal fluctuations and real traffic mutations, for example, when the VR device suddenly renders high frame rate, the model can capture the exponential growth characteristics of the traffic of adjacent time slots, even if the absolute traffic value does not reach the threshold, it can still trigger the early warning mechanism through the nonlinear accumulation of gradient changes, providing a forward-looking decision basis for pre-allocated bandwidth.

[0027] In S3, the intelligent decision of bandwidth adjustment is realized through dynamic threshold triggering and cost-aware mechanism. When the fluctuation index breaks through the preset threshold, the bandwidth adjustment increment is generated based on the nonlinear response function: for bursty traffic (such as the traffic steep rise caused by the simultaneous start of multiple VR devices), according to the exceeding amplitude of the fluctuation index relative to the threshold, an exponential amplification coefficient is used to calculate the required bandwidth increment, ensuring that bandwidth expansion is completed before the traffic peak is formed. At the same time, according to the sensitivity difference of different business scenarios (such as the tolerance of cloud games to delay is much lower than that of ordinary video streams), the preset fluctuation threshold is dynamically adapted - when high-priority business features are detected, the triggering threshold is automatically lowered to improve the response speed. This mechanism effectively avoids the "overreaction" or "response lag" problem of traditional fixed threshold when facing heterogeneous business mixed transmission.

[0028] Further, the stability optimization of the adjustment strategy is realized by a multi-dimensional cost evaluation model. The bandwidth adjustment increment is input into a cost function, which focuses on the cost of two dimensions: one is the square effect of the instantaneous adjustment amplitude, which ensures that bandwidth mutation will not cause transmission jitter; the other is the cumulative effect of the historical adjustment trajectory, which imposes an exponential penalty on frequent adjustment behavior through an implicit memory mechanism. For example, when a line repeatedly triggers bandwidth adjustment due to periodic cloud game packet retransmission, the subsequent adjustment amplitude will be automatically inhibited through the cumulative cost factor, and queue optimization and other supplementary measures will be adopted. This dual-perspective cost evaluation can ensure the quality of service of key businesses while maintaining the energy balance of the overall network topology when dealing with sudden traffic, ultimately achieving multi-objective collaborative optimization of transmission delay and energy efficiency ratio.

[0029] In S4, the precise control and stability of bandwidth optimization are realized by a dynamic damping adjustment mechanism. When the adjustment cost index exceeds the preset threshold, an exponential decay function is used to nonlinearly compress the original bandwidth adjustment increment: for example, in the smart home scenario, if a line triggers a sharp traffic fluctuation due to the simultaneous start of multiple 8K televisions for on-demand, the initial bandwidth expansion may produce a large adjustment amplitude, but after the cost perception module identifies that the line has a high historical adjustment frequency, it will automatically reduce the correction coefficient, so that the actual adjustment amount decays exponentially with the cumulative cost, thereby suppressing the "bandwidth shock" phenomenon. This mechanism not only retains the agility in dealing with sudden traffic, but also prevents resource allocation imbalance caused by frequent short-term adjustments through an implicit feedback loop, especially in mixed business loads (such as concurrent cloud gaming and smart home devices), which can dynamically balance the instantaneous expansion demand and long-term network stability.

[0030] Further, in combination with the multi-objective optimization strategy, when the corrected target adjustment amount is mapped to the physical resource layer, the transmission queue optimization and energy efficiency adjustment modules will be triggered simultaneously. For example, in the VR multi-player collaboration scenario, when the bandwidth expansion amplitude is constrained by the adjustment cost, the transmission time slot of key data packets will be prioritized, and the dynamic frequency reduction of the optical module power consumption will compensate for the local gap in bandwidth resources. This collaborative optimization mechanism allows the quality of service of high-priority businesses to be met while maintaining the overall energy efficiency ratio in the optimal range, especially for mixed business scenarios of long-running high-definition video monitoring and intermittent cloud rendering, achieving optimal network resource utilization and user experience.

[0031] To sum up, in the whole data transmission optimization method based on optical fiber communication, first, the monitoring granularity based on the dynamic adaptation of service characteristics breaks through the response bottleneck of traditional fixed cycle polling, by capturing the traffic mutation of each scene in real time, combining the exponential decay weighting of historical data with the gradient nonlinear amplification of the fluctuation model, a traffic prediction system with short-term sensitivity and long-term stability is constructed, which can identify the traffic transition trend in advance and start the pre-expansion, effectively reducing the queue accumulation delay caused by prediction lag in existing solutions; further, the dynamic threshold triggering and cost-aware collaborative decision mechanism is introduced, through the adaptive fluctuation threshold of service priority, the adjustment of cost constraints and the implicit memory of historical adjustment trajectory, a three-level stabilization protection is constructed, when multiple bandwidth demand surges are detected, bandwidth expansion can be quickly completed, and by adjusting the cost constraints, the amplitude of expansion operation is suppressed, so that the transmission jitter coefficient is within the normal range.

[0032] As shown in Figure 2 In one embodiment, the bandwidth optimization in S4 according to the adjustment cost index and the bandwidth adjustment increment includes: S41, obtaining a preset cost threshold, and determining a correction value according to the preset cost threshold and the adjustment cost index; S42, obtaining a target adjustment amount according to the correction value and the bandwidth adjustment increment; S43, completing bandwidth optimization according to the target adjustment amount.

[0033] In this embodiment, it should be noted that in S41, the intelligent suppression of adjustment amplitude is realized through a dynamic damping coefficient generation mechanism. When the adjustment cost index exceeds the preset threshold, an exponential decay function is used to nonlinearly compress the original adjustment demand: for example, in the smart home scenario, if a line frequently triggers bandwidth adjustment due to the periodic start of multiple 8K televisions for on-demand, the initial expansion may produce a large amplitude change, but the cost evaluation module will identify that the historical adjustment frequency of the line has broken through the stable interval, at this time the correction value will be exponentially decayed with the degree of cost overrun. This mechanism can automatically smooth the amplitude difference between adjacent adjustment periods when the service flow is continuously fluctuating (such as sudden multi-path high-definition backhaul of home security cameras), which not only retains the necessary response capability to deal with sudden traffic, but also avoids the "bandwidth shock" caused by frequent short-term operations, especially in mixed service concurrent scenarios (such as interleaving of cloud office and smart appliance control instructions), this mechanism can effectively maintain the transmission stability of the main line.

[0034] In S42, the final optimization value is generated through the fusion of multi-dimensional constraints. For example, if a sudden increase in transmission data flow is detected, requiring urgent capacity expansion, and if the line has recently undergone multiple adjustments due to image rendering, the theoretical capacity expansion requirement is compressed to an acceptable range by reducing the correction value. Simultaneously, the transmission priority remapping module is triggered to intelligently allocate the compressed adjustment value to critical data channels. This dynamic compromise mechanism ensures transmission quality through traffic feature identification when bandwidth resources are limited (such as when optical module power consumption reaches its upper limit). For example, it prioritizes millisecond-level latency while appropriately relaxing the transmission interval for environmental map data.

[0035] In S43, the mapping process of the target adjustment amount to the physical layer is generally a simple bandwidth value modification, specifically increasing the loan value corresponding to the target adjustment amount to adjust the optical transmission power of the interface to match the actual bandwidth requirement, and ultimately achieving quality lossless transmission of multiple business flows while maintaining the optimal energy efficiency ratio of the optical module.

[0036] In one embodiment, the correction value determined in S41 according to the preset cost threshold and the adjustment cost index is expressed as: ;in, is the correction value, To adjust the cost index, is the preset cost threshold, is the first scaling factor.

[0037] In this embodiment, it should be noted that is an exponential decay function. When the cost index is adjusted Exceeding the preset cost threshold hour, , exponential term Follow The increase decays exponentially, forming a nonlinear inhibition curve. This decay characteristic makes it possible to slightly exceed the limit (such as =0.1, =2), maintain a high correction value, e^{-0.2}≈0.818, allowing for moderate adjustments; or seriously exceed the limit (such as =0.5, =2), a sharp attenuation is triggered, e^{-1}≈0.368, which greatly compresses the adjustment range. In summary, this feature accurately corresponds to For example, if a home gateway's cost index exceeds the limit by 80%, 50% or 30% due to periodic cloud gaming adjustment, the correction amount is completely different, effectively suppressing the problem of large fluctuations caused by large bandwidth adjustments.

[0038] Furthermore, the scaling factor The larger the value, the faster the exponential curve decays, which is suitable for highly sensitive scenarios (such as surgical-level VR remote collaboration, where any excessive behavior needs to be quickly suppressed); The smaller the value, the smoother the decay, which is suitable for scenarios with high tolerance (such as home NAS backup tasks). Dynamically adapt the value (such as automatically adjusting it according to the service quality level of business transmission) to implement differentiated suppression strategies, breaking through the limitations of the one-size-fits-all threshold setting of traditional solutions.

[0039] Furthermore, min{1,•} implements a threshold protection mechanism. hour, 0, exponential term 1, in which case k = 1, fully preserving the original adjustment requirements. This design fully responds to burst traffic within the safety threshold, avoids unnecessary suppression of reasonable adjustments, and addresses the response lag caused by the "indiscriminate suppression" of existing solutions, ensuring that transmission service quality is prioritized while maintaining controllable costs.

[0040] In summary, the transmission jitter caused by frequent adjustments can be suppressed. For example, in a video conferencing scenario, suppose a line triggers 6 adjustments per hour due to participants frequently turning on and off their cameras. In the existing solution, all 6 adjustments are made, and the delay fluctuation reaches ±15ms. In this solution, the 1st and 2nd adjustments ( )k=1, full response, the third time ( ), k=0.89; the 6th time ( =2), k=0.135, the adjustment range is reduced by 86.5%, and the delay fluctuation is compressed to ±3ms. Furthermore, it has achieved a breakthrough in the zero-sum game between delay and energy efficiency. Taking the machine vision inspection in smart factories as an example, when the burst transmission of quality inspection images causes a surge in bandwidth demand, the existing solution, in order to ensure real-time continuous full-load adjustment, the power consumption of the optical module soars by 45%. In this solution, by dynamically suppressing the k value, =0.8, triggering e^{-1.6}≈0.202. Only 20% increments are enabled. Combined with adaptive optical power regulation, power consumption is reduced by 32% with a 2ms increase in latency (still below 25% of the 8ms threshold). Furthermore, the hidden risks of historical traces are eliminated. For the periodic load of the cloud gaming platform during the evening peak period (e.g., 20:00-22:00 every day), traditional solutions ignore historical data and experience bandwidth overload during the same period for three consecutive days. This solution is passed Cross-period accumulation (such as the 4th day of the same period =2.3), suppress the k value to below 0.1, and link other load modules to intelligently migrate the transmission traffic to other lines.

[0041] For example, on a 4K online education platform, the teacher initiates 5-way screen sharing simultaneously (burst traffic 300 Mbps→1.2 Gbps), parameters: =0.4 Gbps (corresponding to the upper limit of the single adjustment cost), =3 process: first burst: =0.38<0.4→k=1, bandwidth is expanded by the bandwidth adjustment increment from 1 Gbps→1.2 Gbps; after a certain period of time, the second burst: =0.43→k=e^{-3*(0.43-0.4)}=e^{-0.09}≈0.914, the actual expansion is to 1.2+(1.2-1)*0.914=1.383 Gbps; after a certain period of time, the third burst: =0.61→k=e^{-3*0.21}=0.532, only expanded to 1.383+0.2*0.532=1.489 Gbps. Compared with the existing solution full adjustment (cumulative expansion to 1.728 Gbps), the final bandwidth of the present solution is 14% lower, but the packet loss rate is still controlled below 0.02% through dynamic queue optimization, and the optical module power consumption is reduced.

[0042] In one embodiment, the time sequence model in S2 is represented as: ; wherein, is the time sequence weight corresponding to the unit time slot average traffic of the jth monitoring time slot before the current time point, is a dynamic base value, is the number of unit time slot average traffic.

[0043] In the present embodiment, it should be noted that the exponential decay characteristics of the numerator have time-sensitive decay characteristics. When >1, as the time slot index j increases (the time is more distant), the value of n-j decreases, resulting in exponential decay; for example, n=5, =2, j=1 (the latest time slot) has a weight of 2 4 =16, j=5 (the oldest time slot) has a weight of 2 0 =1, forming a weight distribution of 16:8:4:2:1, so that the model focuses 85% of the weight on the first three time slots. In summary, during the cloud game loading phase (traffic 0.5 Gbps→3 Gbps jump), the model allocates 72% of the weight to the last two time slots (e.g. time slots 1-2), which is 3.6 times faster than the traditional sliding window (which allocates 20% of the weight evenly), and reduces the prediction error during the business startup phase from 32% to 7%.

[0044] Furthermore, the denominator The normalization design ensures that the sum of all weights is 1 to avoid the absolute value affecting comparability; when the monitoring window n is dynamically expanded (for example, n in VR scenario increases from 5 to 8), the weight ratio is automatically recalculated to maintain the consistency of time decay. In summary, in the smart home multi-device wake-up scenario, the n value is expanded from 5 to 8 to cover the device startup cycle, and the weight distribution is automatically adjusted to 128:64:32:16:8:4:2:1 ( =2), the latest time slot still holds 41.5% of the weight, avoiding the weight dilution problem caused by window expansion in traditional solutions.

[0045] Furthermore, dynamic base value Adaptive adjustment, when =2→1.5, when a sudden increase of 30% in traffic variance is detected (such as burst rendering of 4K video stream), From 2 to 1.5, the weight distribution changes from 16:8:4:2:1 to 5.06:3.38:2.25:1.5:1 (n=5), the weight of the most recent time slot decreases from 50% to 38.8%, and the total weight of the first three time slots decreases from 87.5% to 83.7%, that is, The lower it is, the more historical data is enhanced; generally =2, when the flow fluctuation is lower than the threshold, maintain =2, retaining more historical data features.

[0046] For example, to break through the predicted bottleneck in the service startup phase, when the device is started, the traffic jumps in three consecutive time slots (0.2Gbps→1.8Gbps→4.5Gbps). =1.8, time slot 1 weight =1.8 2 / (1.8 2 +1.8 1 +1.8 0 ) = 3.24 / (3.24 + 1.8 + 1) = 56.3%; Time slot 2 weight = 1.8 1 / 6.04=29.8%; Time slot 3 weight = 1.8 0 / 6.04=13.9%.

[0047] In one embodiment, the fluctuation model in the fluctuation index obtained in S2 based on the fluctuation model, the average flow of the n unit time slots, and the time series weight corresponding to the average flow of each unit time slot is expressed as follows: ;in, is a volatility indicator, is the timing weight corresponding to the average flow rate per unit time slot of the j monitoring time slots before the current time point, is the average flow rate per time slot of the j monitoring time slots before the current time point, is the average flow rate per time slot of j+1 monitoring time slots before the current time point, is the average number of flows per time slot.

[0048] In this embodiment, it should be noted that the index term When j is small (the nearest time slot), Approaching 1, the exponential term ≈ 2, the flow difference The amplification is squared; for example, when n = 5, the exponential term of j = 1 = 1 + e^{-0.2} = 1.818. When the traffic jumps from 200Mbps to 500Mbps, (300)^{1.818} ≈ 300^{1.8} = 4.05×10^4, which is 362 times more sensitive than the linear model. When j approaches n, exp(-j / n) → 0, the exponential term ≈ 1, and it degenerates into a linear calculation; for example, when j = 4, the exponential term = 1 + e^{-0.8} = 1.449, (300)^{1.449} ≈ 300^{1.4} = 1.26×10^3, and the sensitivity is only increased by 11 times.

[0049] Further, The non-uniform reinforcement of weight distribution is combined with the exponential decay weight of the time series model, such as when α=2 and n=5, The distribution is 16:8:4:2:1. The latest time slot with j=1 is given a weight of 51.6%, and its gradient after amplification accounts for more than 86% of the total fluctuation index. At the same time, noise interference is suppressed. When there is random fluctuation in the early time slot (such as traffic jitter ±50Mbps when j=4), =0.06 and the exponential term ≈1.45, (50)^{1.45}×0.06≈216×0.06=12.96, which only accounts for 0.3% of the total volatility index, avoiding false triggering.

[0050] In summary, accurate predictions can be achieved during the service startup phase. For example, when a 4K video stream is started, the traffic increases from 100 Mbps to 800 Mbps (100 Mbps to 300 Mbps, and then to 800 Mbps) within three time slots. When j = 1, the gradient is 500 Mbps (800-300). =0.57, calculated value = 0.57×(500)^{1.717}=24547, when j=2, the gradient is 200Mbps (300-100), =0.29, calculated value = 0.29x(200)^{1.51} = 865, total V = 24500 + 865 = 25412, exceeding the threshold, triggering pre-scaling. Compared with the traditional model (V = 1200), the prediction is 200ms earlier, and the initial buffering time is compressed from 1.2s to 0.3s.

[0051] For example, scene 4K live broadcast (sudden rendering causes inter-frame traffic jump), parameters: n = 5, a = 2, Sequence → [500, 400, 400, 300, 100]; calculation process: time sequence weight calculation: Sequence → [0.516, 0.258, 0.129, 0.065, 0.032]; gradient power calculation, j = 1, =100Mbps, exponential term = 1+e^{-1 / 5} = 1.818, 100^{1.818}≈4325, local fluctuation = 0.516x4325≈2232; j = 2, =0Mbps, exponential term = 1+e^{-2 / 5} = 1.670, 0^{1.67}≈0, local fluctuation = 0.258x0≈0; when j = 3, =100Mbps, exponential term = 1+e^{-3 / 5} = 1.549, 100^{1.549}≈1253, local fluctuation = 0.129x1253≈162; when j = 4, =200Mbps, exponential term = 1+e^{-4 / 5} = 1.449, 200^{1.449}≈2159, local fluctuation = 0.065x2159≈140, then the fluctuation index V = 2232 + 0 + 162 + 140≈2534.

[0052] Effect comparison: existing linear model V = Σ|Δ |=100+0+100+200=400, which is only 15.8% of the value of the model, the model triggers scaling in advance, while the traditional model does not trigger scaling, and the buffering delay of the model is reduced from 180ms to 50ms, and the stall rate is reduced from 12% to 0.8%.

[0053] In one embodiment, the bandwidth adjustment increment of the optical fiber communication line according to the fluctuation index in S3 is represented as: ; wherein, is the bandwidth adjustment increment, is the current bandwidth, is the second scaling coefficient, is the preset fluctuation threshold, is the fluctuation index.

[0054] In this embodiment, it should be noted that, the directional triggering mechanism, (wave fluctuation does not reach the threshold), > 0, triggering exponential term, adjusting increment decreases with the increase of difference; (wave fluctuation exceeds the threshold), = 0, exponential term e 0 = 1, B1 = 2B0, triggering bandwidth multiplication. For example, when B0 = 1Gbps, directly expand to 2Gbps to cope with 2-3 times surge of peak traffic.

[0055] wherein, continuous mapping of the exponential function, linear difference is converted into a nonlinear adjustment coefficient, realizing "small fluctuation weak response, large fluctuation strong inhibition". For example: = 50 (V is slightly lower than the threshold), 0.005, e^{-0.005×50}=e -0.25 ≈ 0.78 → B1 = B0 × 1.78 (proper adjustment); = 200, e^{-0.005×200}=e -1 ≈ 0.37 → B1 = B0 × 1.37 (small adjustment). In summary, it can inhibit the low-frequency small fluctuation error caused by prediction error (such as occasional interference of smart home control signal), and avoid the frequent adjustment caused by static threshold in traditional scheme.

[0056] Further, the dynamic adaptation ability of the scaling coefficient in high sensitivity scenarios increase, such as VR real-time collaboration (delay requirement < 10ms), set = 0.05, when = 50, e^{-0.05×50}=e -2.5 ≈ 0.08, almost prohibit expansion, and force traffic shaping priority. In low sensitivity scenarios decrease, such as file download task (high delay tolerance), set = 0.01, allow = 100 still retains e^{-1}=0.37 adjustment amount, realizing certain progressive expansion.

[0057] In summary, through the asymmetric response mechanism, the bandwidth is multiplied when exceeding the threshold to ensure burst coverage, and the exponential decay is inhibited when the threshold is less than the threshold to avoid over-adjustment, breaking through the limitations of linear proportional adjustment in existing schemes; further, through value dynamically binds the quality of service level of business transmission (such as VR visual = 0.02, video conference =0.01), achieving business-aware differentiated response; further, the multiplication factor 2 implies the maximum allowed expansion range, preventing resource over-allocation due to model misjudgment (e.g. avoiding misadjustment from 1 Gbps to 5 Gbps), forming a double protection with the adjustment cost model; in summary, the entire expression is controlled by three levels of control through directional triggering, nonlinear attenuation, and dynamic adaptation, which improves the bandwidth utilization from 68% to 94% in a laboratory environment under burst traffic scenarios, controls the latency fluctuation within ±5 ms, optimizes the optical module energy efficiency ratio by 23%, and completely solves the technical short board of traditional FTTRB systems.

[0058] In one embodiment, the adjustment cost index according to the bandwidth adjustment increment in S3 is represented as: ; wherein, is the adjustment cost index, is the bandwidth adjustment increment, is the current bandwidth.

[0059] In the present embodiment, it is necessary to note that the super-linear growth characteristic of 1.5 times and the reinforcement awareness of moderate adjustment amount: when =0.5 (bandwidth expansion 50%), =0.5^{1.5}=0.353, which is reduced by 29.4% compared to the linear model (0.5), but is increased by 41.2% compared to the square model ( =0.5^{2}=0.25), forming a gradual penalty curve.

[0060] For example, in the 4K online education burst multi-screen sharing scenario (the traffic demand is calculated by S3 to be 1 Gbps→1.9 Gbps→2 Gbps), the parameters are =1 Gbps, =0.5, =2.

[0061] The first expansion ( =1.9 Gbps), =0.9 Gbps→ =(0.9 / 1)^{1.5}=0.854> , the correction factor k is triggered =min{1, e^{-2×(0.854-0.5)}}=e^{-0.708}≈0.49, the actual adjustment amount =0.9×0.49≈0.45 Gbps, and the final =1.45 Gbps, thus avoiding the 24% bandwidth waste caused by directly expanding to 1.9 Gbps in the traditional scheme.

[0062] Continue the second adjustment ( =2Gbps =0.55Gbps =0.55^{1.5}=0.41< No trigger correction factor, directly expanded to 2Gbps.

[0063] The expression reduces the number of frequent adjustments in real business scenarios through the triple mechanism of asymmetric nonlinearity, dynamic sensitivity adjustment and cross-cycle accumulation, has high burst traffic coverage, controls time delay jitter within a small range, and forms a new generation of cost evaluation paradigm that balances agility and stability.

[0064] Also provided is a data transmission optimization system based on optical fiber communication, the system comprising: An acquisition module is configured to acquire an optical fiber communication line and a monitoring time slot, and to acquire a number n of monitoring units according to the optical fiber communication line; A data processing module is configured to acquire unit time slot average traffic of 1, 2,..., n monitoring time slots before a current time point in real time, to acquire a time sequence weight corresponding to each unit time slot average traffic based on a time sequence model, and to acquire a fluctuation index based on a fluctuation model, the n unit time slot average traffics, and the time sequence weight corresponding to each unit time slot average traffic; A judgment module is configured to judge whether the fluctuation index exceeds a preset fluctuation threshold, and if so, to acquire a bandwidth adjustment increment of the optical fiber communication line according to the fluctuation index, and to acquire an adjustment cost index according to the bandwidth adjustment increment; A transmission line optimization module is configured to complete bandwidth optimization according to the adjustment cost index and the bandwidth adjustment increment.

[0065] In the embodiment, it should be noted that, regarding the above data transmission optimization system based on optical fiber communication, the specific manner of performing operations has been described in detail in the embodiment of the data transmission optimization method based on optical fiber communication, and will not be described in detail here.

[0066] Figure 3 is a block diagram of an electronic device for a data transmission optimization method based on optical fiber communication according to an exemplary embodiment. As shown in Figure 3 the electronic device 700 can include a processor 701, a memory 702. The electronic device 700 can also include one or more of a multimedia component 703, an I / O interface 704 (input / output interface), and a communication component 705.

[0067] The processor 701 is configured to control overall operations of the electronic device 700 to complete all or part of the steps of the above-described data transmission optimization method based on fiber-optic communication. The memory 702 is configured to store various types of data to support operations of the electronic device 700, which can include, for example, instructions for operating any application or method on the electronic device 700, and application-related data, such as contact data, sent and received messages, pictures, audio, video, and the like. The memory 702 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic storage, a flash memory, a magnetic disk, or an optical disk. The multimedia component 703 can include a screen and an audio component. The screen can be, for example, a touch screen, and the audio component is configured to output and / or input audio signals. For example, the audio component can include a microphone configured to receive external audio signals. The received audio signals can be further stored in the memory 702 or transmitted through the communication component 705. The audio component also includes at least one speaker configured to output audio signals. The I / O interface 704 provides an interface between the processor 701 and other interface modules, which can be a keyboard, a mouse, a button, and the like. The buttons can be virtual buttons or physical buttons. The communication component 705 is configured to perform wired or wireless communication between the electronic device 700 and other devices. The wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G, 4G, NB-IOT, eMTC, or other 5G, and the like, or a combination of one or more of them, is not limited herein. Therefore, the corresponding communication component 705 can include a Wi-Fi module, a Bluetooth module, an NFC module, and the like.

[0068] In an exemplary embodiment, the electronic device 700 can be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors, or other electronic elements for performing the above-mentioned data transmission optimization method based on fiber communication.

[0069] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the above-mentioned data transmission optimization method based on fiber communication. For example, the computer-readable storage medium can be the above-mentioned memory 702 including program instructions, which can be executed by the processor 701 of the electronic device 700 to complete the above-mentioned data transmission optimization method based on fiber communication.

[0070] In another exemplary embodiment, a computer program product is also provided, which contains a computer program capable of being executed by a programmable device, and the computer program has code portions for executing the above-mentioned data transmission optimization method based on fiber communication when executed by the programmable device.

[0071] The preferred embodiments of the present disclosure are described in detail above with reference to the accompanying drawings, but the present disclosure is not limited to the specific details in the above-described embodiments. Within the technical concept scope of the present disclosure, various simple modifications can be made to the technical solutions of the present disclosure, and these simple modifications all belong to the protection scope of the present disclosure.

[0072] In addition, it should be noted that each specific technical feature described in the above-described specific embodiments can be combined in any appropriate manner without contradiction. In order to avoid unnecessary repetition, various possible combinations are not described again in the present disclosure.

[0073] In addition, any combination of various different embodiments of the present disclosure can also be made, as long as it does not deviate from the idea of the present disclosure, and it should also be considered as the disclosed content of the present disclosure.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can still be modified, or some or all of the technical features can be replaced by equivalent replacements; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should be covered in the scope of the claims and the description of the present application.

Claims

1. A data transmission optimization method based on optical fiber communication, characterized in that: include: Obtaining an optical fiber communication line and a monitoring time slot, and obtaining a monitoring number n according to the optical fiber communication line; Obtain the average flow rate per unit time slot of 1, 2, ..., n monitoring time slots before the current time point in real time, obtain the time series weight corresponding to the average flow rate of each unit time slot based on the time series model, and obtain the fluctuation index based on the fluctuation model, the average flow rate of the n unit time slots, and the time series weight corresponding to the average flow rate of each unit time slot; Determine whether the fluctuation index exceeds a preset fluctuation threshold; if so, obtain a bandwidth adjustment increment of the optical fiber communication line based on the fluctuation index, and obtain an adjustment cost index based on the bandwidth adjustment increment; Bandwidth optimization is completed based on the adjustment cost index and bandwidth adjustment increment.

2. The data transmission optimization method based on optical fiber communication according to claim 1, characterized in that: The bandwidth optimization is completed according to the adjustment cost index and the bandwidth adjustment increment, including: Obtaining a preset cost threshold, and determining a correction value based on the preset cost threshold and the adjustment cost indicator; Obtain a target adjustment amount according to the correction value and the bandwidth adjustment increment; Bandwidth optimization is completed based on the target adjustment amount.

3. The data transmission optimization method based on optical fiber communication according to claim 2, characterized in that: The correction value determined according to the preset cost threshold and the adjustment cost index is expressed as: ;in, is the correction value, To adjust the cost index, is the preset cost threshold, is the first scaling factor.

4. The data transmission optimization method based on optical fiber communication according to claim 1, characterized in that: The timing model for obtaining the timing weight corresponding to the average flow of each unit time slot based on the timing model is expressed as: ;in, is the timing weight corresponding to the average flow rate per unit time slot of the j monitoring time slots before the current time point, is a dynamic base value, is the average number of flows per time slot.

5. The data transmission optimization method based on optical fiber communication according to claim 1, characterized in that: The fluctuation model in the fluctuation index obtained based on the fluctuation model, the average flow of the n unit time slots and the time series weight corresponding to the average flow of each unit time slot is expressed as follows: ;in, is a volatility indicator, is the timing weight corresponding to the average flow rate per unit time slot of the j monitoring time slots before the current time point, is the average flow rate per time slot of the j monitoring time slots before the current time point, is the average flow rate per time slot of j+1 monitoring time slots before the current time point, is the average number of flows per time slot.

6. The data transmission optimization method based on optical fiber communication according to claim 1, characterized in that: The bandwidth adjustment increment of the optical fiber communication line obtained according to the fluctuation index is expressed as: ;in, For bandwidth adjustment increments, is the current bandwidth, is the second scaling factor, is the preset fluctuation threshold, It is a volatility indicator.

7. The data transmission optimization method based on optical fiber communication according to claim 1, characterized in that: The adjustment cost index obtained according to the bandwidth adjustment increment is expressed as: ;in, To adjust the cost index, For bandwidth adjustment increments, The current bandwidth.

8. A data transmission optimization system based on optical fiber communication, characterized in that: The system comprises: An acquisition module is used to acquire the optical fiber communication line and the monitoring time slot, and acquire the number n of monitoring units according to the optical fiber communication line; The data processing module is used to obtain the average flow rate per unit time slot of 1, 2, ..., n monitoring time slots before the current time point in real time, obtain the time series weight corresponding to the average flow rate of each unit time slot based on the time series model, and obtain the fluctuation index based on the fluctuation model, the average flow rate of the n unit time slots, and the time series weight corresponding to the average flow rate of each unit time slot; A judgment module is used to judge whether the fluctuation index exceeds a preset fluctuation threshold. If so, the bandwidth adjustment increment of the optical fiber communication line is obtained according to the fluctuation index, and the adjustment cost index is obtained according to the bandwidth adjustment increment. The transmission line optimization module is used to complete bandwidth optimization according to the adjustment cost index and the bandwidth adjustment increment.

9. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the data transmission optimization method based on optical fiber communication as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the data transmission optimization method based on optical fiber communication described in any one of claims 1 to 7 is implemented.

Images