A display resource allocation system based on fuzzy logic

Abstract

The application belongs to the field of computer system resource allocation based on specific calculation model, and discloses a display resource allocation system based on fuzzy logic, which comprises: a system obtains an acceleration vector of user interaction, and performs authenticity discrimination based on the direction consistency of the acceleration vector to generate a weighted interaction acceleration; a fuzzy logic reasoning engine maps the weighted acceleration to a resource allocation priority, and the running period of the engine is dynamically adjusted according to the system core temperature to compensate for the calculation delay; finally, resource allocation is performed according to the priority, the application changes the resource scheduling basis from the application state to the physical intention of the user verified by authenticity, realizes the deep synchronization of system response and user attention, avoids the delay and lag of key interaction, and makes the human-computer interaction experience reach the qualitative change of smoothness and harmony.

Description

TECHNICAL FIELD

[0001] The present application relates to a display resource allocation system based on fuzzy logic, belonging to the technical field of computer system resources based on specific computing models. BACKGROUND

[0002] In the current mainstream electronic whiteboard collaborative interaction large screen display system, the allocation of internal computing resources such as central processor time or graphics processor rendering permission usually follows a mature and widely accepted running logic, that is, according to the system state presented by the application program itself to determine the priority of its resource acquisition. The logic takes the state that whether the application program is in the foreground and whether it is performing high-load data processing or content rendering, which can be directly identified by the operating system, as the core basis for judging the urgency of resource demand. This way, when dealing with traditional computing tasks with clear boundaries, it embodies the simplicity and universality of its design.

[0003] However, with the evolution of interactive modes towards high dynamics, high density and multi-task parallelism, especially in scenarios such as remote teaching design review that require users to perform frequent and fine operations, the inherent limitations of the aforementioned resource allocation logic begin to appear. In a typical teaching interactive scenario, the teacher may be playing a high-definition video that occupies most of the screen as a background, while using a brush tool to make small but forceful circles and points on the video to emphasize key content. At this time, according to the existing logic, the system will determine that the video playback application is a high-load foreground task and allocate most of the resources to it. The brush tool, due to its small rendering area and small pixel change, is identified as a low-load operation. The system's resource allocation result has fundamentally deviated from the user's true intention focus, resulting in slow response and broken strokes of the brush, which seriously affects the smoothness of the core teaching task. This phenomenon reveals a deep contradiction in the existing technical paradigm: it equates the computing characteristics of the application with the user's intention priority, but ignores the unencoded dynamic information that the user's physical interaction behavior itself contains, which can most directly reflect the intensity and urgency of their intention.

[0004] In the face of such problems, an intuitive improvement idea is to indiscriminately improve the response priority of all user input events, but this way does not touch the essence of the problem, it cannot effectively distinguish between a calm and gentle drag and a quick writing with an intent to be urgent, and in multi-point or multi-application concurrent interaction, it may cause new resource competition chaos due to frequent indiscriminate priority improvement, making the overall system operation more disordered. Therefore, how to establish a resource allocation mechanism that can go beyond the limitations of the application layer state, directly perceive and understand the dynamic characteristics of user physical interaction behavior, and convert this understanding into a fine-grained resource scheduling strategy that is real-time synchronized with the strength of the user's real intention, so as to solve the problem of mismatch between system response logic and user attention focus, becomes the technical problem to be solved by the present application. SUMMARY

[0005] The present application provides a display resource allocation system based on fuzzy logic, which mainly aims to solve the problem that the prior art relies on application state for resource allocation, thereby causing the inconsistency between system response and user real intention focus.

[0006] To achieve the above-mentioned purpose, the present application provides a display resource allocation system based on fuzzy logic, comprising:

[0007] An interaction dynamic characteristic acquisition module configured to obtain a coordinate time sequence generated by user interaction at a sampling frequency of no less than 120 Hz, and calculate an acceleration vector sequence based on the coordinate time sequence;

[0008] An intention authenticity discrimination unit configured to calculate a quantitative index representing the consistency of the direction of the acceleration vector sequence based on the acceleration vector sequence output by the interaction dynamic characteristic acquisition module within a time window, and adjust the weight of the acceleration amplitude of the interaction behavior according to the quantitative index to generate a weighted interaction acceleration;

[0009] A fuzzy logic reasoning engine module configured to take the weighted interaction acceleration output by the intention authenticity discrimination unit as a first fuzzification input, and map the first fuzzification input to a resource allocation priority value within its running cycle according to the fuzzy rule base stored in the memory;

[0010] A temperature perception module for obtaining a real-time temperature value of the system core processor;

[0011] A cycle adaptive adjustment unit configured to dynamically adjust the running cycle of the fuzzy logic reasoning engine module according to the real-time temperature value and a temperature and calculation delay characteristic model stored in the memory;

[0012] a resource dynamic scheduling module configured to allocate computing resources of the system according to the resource allocation priority value output by the fuzzy logic inference engine module.

[0013] Preferably, the intention authenticity discrimination unit calculates a quantified indicator of direction consistency, specifically: calculates the cosine value of the included angle between each pair of consecutive acceleration vectors in the acceleration vector sequence within the time window; and arithmetically averages all the calculated cosine values to obtain an average cosine value as the quantified indicator of direction consistency.

[0014] Preferably, the intention authenticity discrimination unit generates the weighted interaction acceleration through the following relationship: , wherein, is the original amplitude of the acceleration of the interaction behavior, is the quantified indicator of direction consistency, i.e. the average cosine value, is a weight function stored in the memory and monotonically increasing with the number of interaction behaviors, and the value range of the weight function is zero to one.

[0015] Preferably, the period self-adaptive adjustment unit dynamically adjusts the running period of the fuzzy logic inference engine module, specifically: determines an expected computation delay according to the real-time temperature value and the temperature-computation delay characteristic model; and subtracts the expected computation delay from a baseline running period to obtain a new, shortened running period.

[0016] Preferably, the fuzzy logic inference engine module is further configured to: take a value representing the content rendering complexity associated with the interaction behavior as a second fuzzification input, and determine the resource allocation priority value in combination with the weighted interaction acceleration and the value representing the content rendering complexity.

[0017] Preferably, the fuzzy rule base contains the following rule: when the fuzzified value of the weighted interaction acceleration falls into a high-intensity interaction interval, the resource allocation priority value is determined as a highest priority value, and the execution of this rule is independent of the state of any other fuzzification input.

[0018] Preferably, the system further comprises a collaborative intention aggregation unit configured to: when the interaction dynamic feature acquisition module detects at least two interaction behaviors with different spatial positions within a set time window, and the intention authenticity discrimination unit outputs a high-intensity weighted interaction acceleration for at least two interaction behaviors, confirm that a collaborative event has occurred; and based on the confirmation of the collaborative event, trigger a collaborative allocation strategy that concentrates computing resources in the area where the at least two interaction behaviors are located.

[0019] Preferably, the system further comprises a cognitive load state tracking unit configured to: calculate an exponentially moving average value representing the user's persistent cognitive load based on a series of resource allocation priority values outputted by the fuzzy logic inference engine module within a period of time; and automatically adjust the presentation mode of all notification information in the system except the alert information to reduce the cognitive interference to the user when the exponentially moving average value exceeds a load threshold value stored in the memory.

[0020] Preferably, the interactive dynamic feature acquisition module is further configured to: set the acceleration of the interactive behavior to a zero value during a period when no user interaction is detected, so that the weighted interactive acceleration outputted by the intention authenticity discrimination unit is also a zero value, thereby switching the basis for the decision of the fuzzy logic inference engine module to other non-interactive related fuzzification inputs.

[0021] Preferably, the resource dynamic scheduling module is configured to: realize the allocation of computing resources by adjusting the central processing unit time slice allocation and the graphic processing unit rendering queue priority of the application process corresponding to the interactive behavior.

[0022] Compared with the prior art, the present application has the following beneficial effects:

[0023] 1、The present application introduces the physical acceleration generated when the user interacts with the display interface as a continuous dynamic physical feature into the resource allocation decision process, and establishes a direct correlation between the physical dynamic feature and the internal resource allocation priority of the system through a fuzzy logic inference mechanism. This way changes the core basis of system resource scheduling from relying on the relatively static or slowly changing running state of the application program itself to directly responding to the intention intensity revealed by the user in the physical operation in an instant. When the user performs high-speed force interaction, the system no longer ignores the resource request of the application to which the interaction belongs due to the application being in the background or seemingly low load, avoiding the delay or lag of key operations caused by the disconnection between the system cognition and the user intention, and keeping the response logic of the system and the focus of the user's attention in time synchronization.

[0024] 2、The present application further introduces the consideration of the direction stability of the acceleration vector of the interactive behavior within a continuous time window based on the acceleration vector as a measure of intention intensity. By coupling the size feature of the acceleration with the continuity feature of its direction, the system obtains the ability to discriminate the authenticity of the intention, which can distinguish the directionally stable fast strokes driven by the explicit intention of the user from the directionally scattered non-intentional tremors caused by physiological or environmental factors. In this way, the system avoids misinterpreting high-frequency tremor false signals as high-intensity user intention, thereby suppressing unnecessary resource priority promotion and system load fluctuation, making its perception of the user's real intention have better adaptability and reliability in different users and variable environments.

[0025] 3、The fuzzy logic reasoning engine constructed by the present application, the resource allocation priority output sequence continuously generated in the running is further utilized as a new information source, through the analysis of the statistical characteristics of the priority sequence in a period of time, the system can indirectly measure the persistence level of the current cognitive load of the user, when the user is identified to be in a high intensity interaction state for a long time, i.e. in a high cognitive load state, the system will actively adjust the presentation mode of non-critical notifications to reduce cognitive interference, this design will be used for the control output of micro-operation response, the perception input for understanding the macro user state is reused, without increasing any additional hardware or perception module, the system is extended from ensuring the smoothness of the operation level to protecting the immersive work flow of the user, and the experience of human-computer interaction is more coherent and harmonious. BRIEF DESCRIPTION OF DRAWINGS

[0026] Fig. 1 The system perception decision and adaptive adjustment integrated architecture of the present application is shown in the figure.

[0027] Fig. 2 The user physical intention to system resource priority mapping flowchart of the present application is shown in the figure.

[0028] Fig. 3 The system model two-stage adaptation procedure chart for hardware and user of the present application is shown in the figure. DETAILED DESCRIPTION

[0029] In order to make the technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described in detail below in combination with the drawings. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0030] The display resource allocation system based on fuzzy logic provided by the application is built on a specific computing model, which aims to directly map the dynamic characteristics contained in the user's physical interaction behavior into the resource allocation decision of the system. The system mainly includes an interactive dynamic characteristic acquisition module, an intention authenticity discrimination unit, a fuzzy logic reasoning engine module, a temperature perception module, a period adaptive adjustment unit and a resource dynamic scheduling module. Under certain configurations, the system can also integrate a collaborative intention aggregation unit and a cognitive load state tracking unit. The top-level data flow starts from the perception of the user's physical operation by the interactive dynamic characteristic acquisition module. The generated acceleration vector sequence is sent to the fuzzy logic reasoning engine module as the core input after being weighted and adjusted by the intention authenticity discrimination unit. The engine generates resource allocation priority values according to the built-in fuzzy rule base combined with auxiliary inputs such as content rendering complexity. At the same time, the running period of the engine itself is dynamically regulated by the compensation loop composed of the temperature perception module and the period adaptive adjustment unit. Finally, the resource dynamic scheduling module adjusts the resources of the central processor and the graphics processor in real time according to the priority values.

[0031] In a typical remote teaching or collaborative design scenario with high requirements for interaction fluency, the existing resource allocation method often causes mismatch between system response and user's real intention focus due to its dependence on application layer state. To solve this technical bottleneck, the interactive dynamic characteristic acquisition module of the application is configured to continuously obtain the coordinate time sequence generated when the user interacts with the display interface through a medium such as a stylus or a finger at a sampling frequency of no less than 120 Hz This frequency ensures that the details of the fast strokes can also be captured without omission. To convert this raw coordinate data into a physical quantity that can represent the user's intention intensity, the module calculates an acceleration vector sequence in real time by performing second-order difference operation on consecutive coordinate points, for example, the acceleration vector at any time can be calculated by the procedure , where represents the coordinate vector at the current time, is the sampling time interval. In this way, the system converts the user's interaction force that was originally not perceptible into a measurable and analyzable acceleration physical characteristic, thereby building an input basis based on physical reality for subsequent intention understanding.

[0032] In real use environment, user's physiological hand jitter or vibration caused by environmental factors, can generate high amplitude acceleration pseudo signal, if not to distinguish, the system will be these meaningless jitter misinterpretation as high intensity of interaction intention, for coping with this challenge, the scheme introduces an intention authenticity discrimination unit, the core mechanism of this unit, in order to draw lessons from the signal processing field for measuring the stability of time series principles, to quantify the authenticity of interaction intention, its specific procedures are, first in a containing , for example The latest acceleration vector sliding time window, calculate the cosine value of the direction between each pair of continuous acceleration vector And , then, all the cosine values calculated in the window Arithmetic average, thus obtaining a quantitative index that can represent the direction consistency of the acceleration vector sequence in the short term, namely the average cosine value , in view of this, the unit further through a preset weight function Adjust the original acceleration amplitude To generate the final weighted interaction acceleration , the relationship is , this weight function Designed as a monotonically increasing function of , its value range is defined between zero and one, and is preset in the form of lookup table or analytical expression in the system memory, when The value of 1, representing the user's direction stable intention, the value of Also tends to 1, thus retaining the original acceleration amplitude, on the contrary, when The value is low, representing the interaction direction confusion may be unintentional jitter, The value tends to 0, thus effectively suppressing the strength of the acceleration pseudo signal, through this kind of dynamic weight adjustment based on the direction stability of acceleration vector, the system has the ability to distinguish the authenticity of user's intention, provides a high confidence input signal for subsequent fuzzy logic reasoning.

[0033] After obtaining the weighted interaction acceleration Reflecting the intensity of user's intention, how to convert it into resource allocation instructions that the operating system can understand and execute, is the core link of this calculation model, in view of the mapping relationship from physical interaction intensity to resource priority demand has fuzzy and nonlinear characteristics, the system uses a fuzzy logic reasoning engine module to undertake this task, the fuzzy stage of the engine will be the weighted interaction acceleration As a first fuzzification input, and to map its value to a fuzzy linguistic set such as slow, medium, fast, etc. according to a membership function, the engine can also be configured to take as a second fuzzification input a value representing the content rendering complexity associated with the interactive behavior, which can be obtained from the graphics processor driver interface, and also mapped to a fuzzy set such as light, medium, heavy, etc. Based on this, the fuzzy logic inference engine module performs inference according to a fuzzy rule base pre-stored in the memory, which consists of a series of IF-THEN rules, for example, IF is fast THEN priority is high, which is set to have a high execution weight, i.e. when the system perceives that the user is performing high intensity fast interaction, the resource allocation priority is directly determined as a highest priority value, which is independent of the state of any other fuzzification input, other rules such as IF is medium AND is heavy THEN priority is high, which covers all other working conditions, after inference and defuzzification, the engine finally outputs a well-defined resource allocation priority value, which provides a direct quantitative basis for subsequent resource scheduling.

[0034] In the working condition of long-time high-load operation of electronic devices, the temperature rise of the core processor may trigger the thermal throttling mechanism, causing the increase of calculation delay. In order to compensate for the influence brought by this physical environmental change, the system integrates an adaptive adjustment mechanism composed of a temperature perception module and a periodic adaptive adjustment unit, wherein the temperature perception module is responsible for obtaining a real-time core processor temperature value by using the built-in sensor of the processor, and the periodic adaptive adjustment unit adjusts the running period of the fuzzy logic inference engine module dynamically according to the real-time temperature value and a temperature and calculation delay characteristic model pre-stored in the memory. The characteristic model is established through offline calibration experiments in which the environmental temperature is controllable: at different temperature step points, the fuzzy logic inference engine module is driven to perform standard inference operation continuously for thousands of times, and the average calculation delay is accurately measured, so as to obtain a discrete data pair of core processor temperature and average calculation delay of the specific hardware. By polynomial fitting of the data points, a function describing the relationship between the two is finally generated, and a lookup table is constructed and solidified in the system memory. According to this, the unit adopts a reverse compensation deterministic procedure to set the next running period of the fuzzy logic inference engine , which is calculated by subtracting the expected calculation delay from a reference running period , for example 20 milliseconds For example, when the core processor temperature is 40°C, the model calculates the expected latency. If the interval is 0.5 milliseconds, then the new running cycle is... When the core processor temperature rises to 85°C due to high load, the model calculates the expected latency. If increased to 3.0 milliseconds, the system will automatically shorten the runtime to... Thus, when the processor temperature rises and the expected latency increases, the system will proactively shorten the decision interval, using the increase in decision frequency to offset the decrease in the speed of a single decision, thereby maintaining the total response time at a stable level.

[0035] When the fuzzy logic inference engine module outputs the final resource allocation priority value, the resource dynamic scheduling module is responsible for converting this decision into the actual allocation of system computing resources. This module interacts with the kernel scheduler of the operating system, and according to the received priority value, it adjusts one or both of the central processing unit time slice allocation and the graphics processor rendering queue priority of the application process corresponding to the current interaction behavior to achieve resource allocation. For example, a paint application process with a high priority value, such as 95, can have its nice value temporarily adjusted to -20 in a Linux-like system, and its related rendering instructions can be placed at the front of the graphics processor task queue, thereby obtaining more processor resources to ensure smooth and instantaneous response of its handwriting rendering. To cope with the scenario of multiple users working together, the system can also include a collaboration intention aggregation unit. When the interaction dynamic feature acquisition module detects at least two spatially different interaction behaviors within a set time window, and the intention authenticity discrimination unit outputs a high-intensity weighted interaction acceleration for at least two interaction behaviors, the aggregation unit confirms that a collaboration event has occurred. Based on this confirmation, the system will trigger a collaborative allocation strategy, such as aggregating computing resources to the area where at least two interaction behaviors are located, to ensure the overall smoothness of the entire collaboration task. In addition, to protect the cognitive state of the user, the system can also integrate a cognitive load state tracking unit that reuses the output of the fuzzy logic inference engine. It takes a series of resource allocation priority values output by the fuzzy logic inference engine within a period of time, such as 30 seconds, as input signals, and calculates a moving average value representing the user's sustained cognitive load. When this moving average value exceeds a pre-set load threshold stored in the memory, it indicates that the user is in a high-intensity focused work state. The system will automatically adjust the presentation mode of all notification information except critical alarm information, such as downgrading pop-up notifications to static icons in the status bar, thereby reducing cognitive interference on the user without interrupting information transmission. During periods when no user interaction is detected, the interaction dynamic feature acquisition module is configured to set the acceleration of the interaction behavior to a zero value, which makes the weighted interaction acceleration output by the intention authenticity discrimination unit also zero. In this non-interaction state, the decision basis of the fuzzy logic inference engine will smoothly switch to other non-interaction-related fuzzification inputs, such as relying only on content rendering complexity for decision-making, thereby falling back to a more traditional resource allocation logic that ensures the consistency of the system's behavior in different interaction modes.

[0036] In an application scenario of distance teaching, a teacher is demonstrating a high-definition video of a physical experiment to students, which occupies most of the screen and shows the trajectory of a particle in a cloud chamber. The decoding and rendering process of the video continuously occupies system computing resources. When the video is playing at a critical moment, the teacher needs to use a stylus to circle and point on the video screen in a small range but fast and forceful manner to highlight a fleeting particle trajectory. In a system using a traditional resource allocation method, the brush tool is determined to be a low-load operation because the rendering area is small and the pixel changes are few, so it is not allocated enough computing resources, resulting in a delay and lag in the teacher's handwriting, and the transmission process of the teaching intention is significantly disturbed. When the system using the technical solution of the present application faces this working condition, the interactive dynamic feature acquisition module captures a series of coordinate time sequences generated when the teacher circles and points fast, and converts them into an acceleration vector sequence. The intention authenticity discrimination unit then processes the vector sequence, calculates the average cosine value of the direction angle between consecutive vectors, and obtains a high directional consistency quantization index close to 1 . This index represents an intention clear stroke with a stable direction, and the weight function outputs a weight value also close to 1, so that the original acceleration amplitude of the brush interaction is transmitted almost without loss, generating a high-intensity weighted interaction acceleration . This weighted interaction acceleration and the value representing the high content rendering complexity of the video playback are simultaneously input into the fuzzy logic reasoning engine module.

[0037] Inside the module, a rule with a high execution weight in the preset fuzzy rule base is triggered, that is, when the fuzzification value of the weighted interaction acceleration falls into a high-intensity interaction interval, the resource allocation priority value is determined as a highest priority value. The execution of this rule changes the core basis of system resource allocation from the rendering area at the application level to the user intention intensity at the physical interaction level, thereby effectively managing the resource competition between the high-load background application and the high-intention focus foreground operation. The engine outputs a highest priority value according to this rule, and the resource dynamic scheduling module immediately allocates computing resources from the video playback process that has been in the background to the interaction process associated with the brush tool after receiving this value. Finally, the teacher's every fast circle and point can instantly and without delay present the handwriting on the screen, keeping pace with the physical movement of the nib, so that the fleeting particle trajectory is clearly and completely captured and presented. The computing resources of the system are synchronized and aggregated to the focus of user attention at the moment when the focus shifts, so that the human-computer interaction process and the cognitive process of the teaching task present consistency.

[0038] Example 2: To objectively verify the improvement effect of the technical scheme of the present application on the response delay of key interactive events in the scenario of coexistence of high load and high intensity interaction, the present comparative test is set up. The test purpose is to quantify the rendering delay of high-intensity pen interaction in the simulated remote teaching scenario, comparing the test group using the resource allocation system of the present application with the control group using the standard operating system resource allocation method. The test platform is based on a computer equipped with a multi-point touch screen, with a standard four-core configuration as the core processor and a general Linux distribution as the operating system. To ensure the consistency and reproducibility of interactive input, a six-axis mechanical arm is used to replace human hands to perform touch pen operations. The touch pen held by the end of the mechanical arm is set to peak acceleration and path direction stability to draw a straight line with a length of 10 cm in the specified area of the screen, simulating an urgent writing action. The value of this acceleration is set to effectively simulate high-intensity interaction of users. This value corresponds to the typical value of the high-intensity interaction interval obtained from statistical analysis of user interaction behavior. A high-speed camera with a frame rate of 240 Hz is directed at the screen to capture the process from the physical contact of the touch pen with the screen to the first rendering of the corresponding handwriting on the screen, and to calculate the handwriting rendering delay through frame-by-frame analysis with a time accuracy of 4.17 ms. In the test, a high-load background task, i.e., a 60 Hz 4K video, is set to loop play in the 80% area of the screen to continuously occupy system computing resources.

[0039] The test process first conducts a baseline test without background load in both groups, and records that the handwriting rendering delay of both is in the interval of 8-13 ms. Then, the background task of 4K video playback is started in both groups, and after the system load stabilizes, the test is conducted, and it is recorded that the average handwriting rendering delay of the control group increases to 79.23 ms, while under the same load conditions, the average handwriting rendering delay of the test group is only 12.51 ms. This difference is directly related to the running state of the system in the test group. In the test of the test group, when the high-intensity weighted interaction acceleration generated by the brush interaction is detected by the fuzzy logic reasoning engine module and its peak reaches , the resource allocation priority value of the brush process is instantaneously increased to 96 according to the internal rules, triggering the intervention of the resource dynamic scheduling module. To further verify the effectiveness of the intention authenticity discrimination unit, an interference test is added in the test group, i.e., while the mechanical arm is drawing a line, a micro-vibration is applied through a vibration device. Under this working condition, the rendering delay of the test group for the intention brush interaction still remains at a low level of 16.68 ms, and the reason is that the intention authenticity discrimination unit gives a weight coefficient of 0.1 to the acceleration signal generated by the vibration and the direction is not stable.substantially reduced, so that its corresponding weighted interactive acceleration is inhibited at a level, thus avoiding false resource scheduling; experimental data show that the technical scheme of the present application can, in a computing environment with high-load background tasks, identify the intention authenticity by sensing the physical dynamic characteristics of user interaction behavior, and then dynamically allocate system resources using a fuzzy logic reasoning model, thereby controlling the response delay of key interactive operations at a sustained low level.

[0040] Embodiment 3: This embodiment combines Figs. 1 to 3 to illustrate a display resource allocation system based on fuzzy logic, as Fig. 1 shown, the data flow of this system begins with the collection of user physical interaction coordinate time series, which is processed by the interactive dynamic feature acquisition module. This module collects interaction points at a frequency of no less than 120Hz and calculates the acceleration vector sequence; this sequence is then passed to the intention authenticity identification unit, which generates weighted interactive acceleration based on the consistency of the direction of acceleration to distinguish between intentional operations and unintentional shaking. The weighted acceleration signal, as the core input, is combined with another second fuzzification input representing the complexity of content rendering, and together they are fed into the fuzzy logic reasoning engine module. This engine, as the core decision-making unit, maps the input signals to the final resource allocation priority value. The operation of the reasoning engine is also adjusted by a compensation loop, in which the temperature sensing module is responsible for obtaining the real-time temperature value of the system's core processor and providing this data to the periodic adaptive adjustment unit, which dynamically adjusts the engine's operation period according to the temperature and delay model to compensate for calculation delay. Finally, the resource dynamic scheduling module adjusts the system resources of CPU / GPU according to the priority value output by the engine. In addition, the system also includes a cognitive load state tracking unit, which calculates the user's cognitive load based on the priority sequence over a period of time and adjusts non-critical notifications accordingly to reduce cognitive interference.

[0041] As Fig. 2As shown, the interactive dynamic feature acquisition module calculates the acceleration vector by high-frequency sampling touch coordinates, generates acceleration data and outputs to the intent authenticity discrimination unit, which distinguishes real intent and jitter by analyzing direction consistency, and then generates a weighted signal, which is input into the fuzzy logic reasoning engine module along with the content rendering complexity information from the outside. The module is the core of decision-making, and its output is a quantitative priority value. At the same time, a separate temperature monitoring loop is continuously working, monitoring the processor temperature and generating temperature data, which is handed over to the cycle adaptive adjustment unit for delay compensation, and generates dynamic cycle parameters to adjust the reasoning engine's running rhythm, ensuring stable response. The priority output by the reasoning engine is obtained by the resource dynamic scheduling module, which adjusts the processor time slice and the rendering queue priority to output the final output of the system: the optimized resource allocation scheme. The figure also shows two extendable units in dashed boxes, namely the collaborative intent aggregation unit for multi-point interaction detection and collaborative resource allocation, and the cognitive load state tracking unit for monitoring user load and intelligent notification management. Both interact with the fuzzy logic reasoning engine module.

[0042] As shown in Fig. 3 , first, a temperature-computing delay feature model needs to be built, that is, under temperature control, measure and fit the reasoning computing delay under different core temperatures to generate a lookup table or function. Second, the weighted interactive acceleration membership function needs to be calibrated, that is, by simulating different intensity interactions with a mechanical arm, and performing probability density analysis on the collected acceleration data to determine the boundaries of fuzzy sets such as gentle, agile, and urgent; At the same time, the direction consistency weight function also needs to be calibrated, that is, based on standard lines and simulated jitter actions, determine the mapping relationship between direction consistency indicators and weight values. Finally, through regression analysis of the relationship between multiple GPU performance indicators and overall utilization, the optimal weight coefficient is solved to build a content rendering complexity calculation model. All models formed in this stage will be fixed in the system memory and deployed with the device. When the user device starts for the first time, it enters the first run self-calibration phase. The system will guide the user to perform high / low intensity interaction tasks and record the priority sequence during this process, and then calculate the priority mean and set the load threshold, so that the threshold is adapted to the user's personal use habits. Subsequently, the system automatically performs internal function self-checking and timing calibration, tests the logic consistency of the fuzzy rule base by injecting virtual touch events, and measures the internal data round-trip time. After the self-checking is passed, the initialization of the system is completed, and it formally enters the normal running mode.

[0043] Example 4: In a specific engineering practice, to ensure that the technical solution of this invention achieves the preset operating effect when applied to a new touch display device with a specific hardware configuration, several models and parameters within the system need to be calibrated offline in one go. This calibration process aims to match the general algorithm logic with the physical and performance characteristics of the specific hardware platform. The calibration process begins with establishing a temperature and computational latency characteristic model. The construction of this model aims to compensate for fluctuations in the computation time of the fuzzy logic inference engine caused by changes in processor temperature. The experiment is conducted in a shielded room with a controllable ambient temperature. The touch display device to be calibrated is placed in the room, and its internal temperature sensing module and an external high-precision timer are synchronously connected to a data acquisition host. The calibration procedure is as follows: First, the temperature of the shielded room is set to 25°C. After the temperature of the device's core processor stabilizes, a test script drives the fuzzy logic inference engine module to continuously execute 1000 standard inference operations. The timer records the total time required to complete these 1000 operations, and then the average computational latency of a single inference is calculated. Subsequently, the temperature of the shielded room is gradually increased in 5°C increments until the temperature of the core processor reaches its safe operating limit, which is 85°C in this embodiment. At each temperature step point, the above-described delay calculation and measurement process is repeated. In this way, a set of core processor temperatures related to this specific hardware is obtained. With average computation delay For discrete data pairs, a function representing the relationship between the two is generated by performing polynomial fitting on the set of data points. Based on this function, a lookup table is constructed and stored in the system memory as a temperature and computational delay characteristic model of the device.

[0044] Next, we will analyze the core input of the fuzzy logic reasoning engine module, namely the weighted interaction acceleration. The fuzzy membership function is calibrated using a six-axis robotic arm test platform capable of performing line-drawing actions at a preset acceleration. The aim is to align the boundaries of the three fuzzy language sets (smooth, agile, and rapid) with the statistical distribution of interactive behaviors. The calibration procedure is as follows: control the robotic arm to cover from... arrive Within the scope of For a plurality of different peak accelerations of the step length, a series of standard line drawing actions are performed on the touch screen, and the interactive dynamic feature acquisition module records the original acceleration value corresponding to each action. By performing probability density analysis on a large number of collected acceleration data samples, it can be observed that the data presents a plurality of aggregation zones on the numerical axis. The right boundary of the first main aggregation zone is defined as the upper limit of the flat fuzzy set, and the left boundary of the last main aggregation zone is defined as the lower limit of the rapid fuzzy set. The region between the two boundaries belongs to the agile fuzzy set, thereby determining the inflection point of the membership function. At the same time, the weight function used to generate the weighted interactive acceleration is calibrated. This process aims to determine the mapping relationship between the direction consistency quantitative indicators and the weight value, in order to distinguish between intentional strokes and unintentional shaking. The calibration procedure is as follows. First, the mechanical arm performs a standard line drawing action with a direction consistency quantitative indicator value of , and sets the output value of in this state to . Second, by superimposing a high-frequency random disturbance signal in the motion control program of the mechanical arm, it performs a series of simulated shaking actions with the direction consistency quantitative indicator fluctuating between and , and sets the output value of in this state to . Based on these two endpoints, and setting the function to be monotonically increasing within this interval, a weight function based on an S-shaped curve can be determined. Through the above calibration procedure, several key models and parameters in the original technical solution are concretized from a general logical framework into an engineering example corresponding to this specific hardware platform.

[0045] Embodiment 5: After the technical solution of the present application is deployed on a specific device, a standardized initial running self-calibration procedure is performed to adapt its internal model parameters to the user's personal use habits and the current hardware state. When the device is started for the first time, the system guides the user to enter an interactive calibration link. This link requires the user to perform two pre-set interactive tasks, one being a low-intensity task of continuously dragging an icon at a flat speed, and the other being a focused high-intensity task of quickly and continuously clicking multiple targets within a limited time. In this process, the cognitive load state tracking unit continuously records the sequence of resource allocation priority values output by the fuzzy logic reasoning engine module, and calculates the priority mean of the low-intensity task stage and the priority mean of the high-intensity task stage, respectively. Subsequently, the load threshold a predetermined quantile between the two means, calculated as where the coefficient is a preset constant with a value between 0 and 1, in this embodiment, the value is 0.75.

[0046] After this calibration step, the system further performs an internal functional self-check and timing calibration, which does not require user involvement. The system internally generates a simulated virtual touch event with a high-intensity weighted interactive acceleration and directly injects this event into the fuzzy logic inference engine module. The system then verifies whether the output of the engine is a value in a preset highest priority interval, in order to test the logical consistency of the core fuzzy rule base. At the same time, the system measures the complete internal data round-trip time from the injection of the virtual event to the reception of the corresponding priority value by the resource dynamic scheduling module using a high-precision internal clock . The measured time is used to set the time window threshold in the collaborative intent aggregation unit for determining whether multiple interactive behaviors occur simultaneously, i.e., the window is set to .

[0047] Embodiment 6: In a specific engineering implementation, in order to enable the content rendering complexity input used by the fuzzy logic inference engine module to quantitatively reflect the real computing load of different applications on a specific hardware platform, a computing model needs to be constructed to fuse multiple graphics processor performance indicators into a single value. The construction of this model is framed as an offline optimization procedure with the goal of minimizing the prediction error, aiming to solve a set of optimal weight coefficients so that the constructed computing model can accurately predict the actual graphics processor usage. The procedure first selects a set of key performance indicators that can represent the graphics processor load. In this embodiment, the number of vertex processing per frame , the number of fragment shader execution instructions per frame , and the number of texture memory access bytes per frame are selected as the basic input variables. To eliminate the effects of different dimensions and numerical ranges of each variable, normalization processing is required. This processing procedure is as follows: a set of benchmark test programs covering multiple rendering loads are run on the target hardware platform, and the maximum values of each variable during the entire test process are recorded . , . Subsequently, in the actual operation of the system, the normalized indicator value at any time is calculated as follows: , , .

[0048] The final content rendering complexity value , which is calculated by weighted summation of the above normalized indicators, with the calculation relationship as follows: , where the weight coefficients are determined by a calibration procedure based on multiple linear regression, in which a series of test applications with different rendering loads are run, and the instantaneous values of the basic input variables and the instantaneous values of the overall usage of the graphics processor provided by the operating system or driver interface are recorded simultaneously. Through regression analysis of a large number of data points collected, a set of weight coefficients is solved, which can minimize the root mean square error between the calculated and the actually measured , and satisfy the constraint . This set of calibrated weight coefficients is fixed in the system, thereby completing the construction of the content rendering complexity calculation model.

[0049] Embodiment 7: In order to make the allocation instructions of the resource dynamic scheduling module produce a certain execution effect under different operating system kernel scheduling mechanisms, the module also includes a closed-loop feedback correction procedure for scheduling execution effect. In the offline calibration stage, first, a mapping relationship table is established between the resource allocation priority value and a set of kernel-level thread scheduling interface call parameters provided by the operating system , where represents the mapping relationship table, which is usually implemented in the form of a lookup table or a function; is the resource allocation priority value output by the fuzzy logic reasoning engine, which is used as the input of the mapping; the mapping relationship converts the input value into a specific operating system kernel-level thread scheduling interface call and a set of corresponding parameters required to execute the call . The call parameters include thread priority, CPU core affinity setting, and graphics processor context priority. In the real-time running stage of the system, for each interactive thread set to high priority, the procedure measures the actual time to complete the rendering or calculation task, and compares it with the target response time calculated according to the current interaction acceleration, for example, for an interaction sampling rate of 120Hz, is set to 8.33 milliseconds, to calculate the response error , and a proportional-integral controller, i.e. , is used to dynamically adjust the priority value output to the mapping table next time, where is the priority adjustment value for the current time instant, whose initial value is the output of the fuzzy logic inference engine, while the gain coefficient With is a value determined in the offline calibration phase by injecting a series of step response test signals on the target hardware and targeting the critical damping response, a procedure that transforms the resource allocation from an open loop request to a closed loop control process targeting the actual response time.

[0050] It is apparent for a person skilled in the art that the present application is not limited to the details of the above-described exemplary embodiments, but can be implemented in other concrete forms without departing from the spirit or essential characteristics of the present application.

[0051] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application, and although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A fuzzy logic based display resource allocation system, characterized by, The system comprises: an interactive dynamic feature acquisition module configured to obtain a coordinate time series generated by user interaction at a sampling frequency no less than 120 Hz, and calculate an acceleration vector sequence based on the coordinate time series; an intention authenticity discrimination unit configured to calculate a quantitative indicator representing the consistency of the direction of the acceleration vector sequence based on the acceleration vector sequence output by the interactive dynamic feature acquisition module within a time window, and adjust the acceleration amplitude of the interaction behavior according to the quantitative indicator; a fuzzy logic reasoning engine module configured to take the weighted interaction acceleration output by the intention authenticity discrimination unit as a first fuzzification input, and map the first fuzzification input to a resource allocation priority value within its running cycle according to a fuzzy rule base stored in the memory; a temperature sensing module for obtaining a real-time temperature value of the system core processor; a cycle adaptive adjustment unit configured to dynamically adjust the running cycle of the fuzzy logic reasoning engine module according to the real-time temperature value and a temperature and calculation delay characteristic model stored in the memory; a resource dynamic scheduling module configured to allocate the computing resources of the system according to the resource allocation priority value output by the fuzzy logic reasoning engine module; The intention authenticity discrimination unit calculates the quantitative indicator of the direction consistency, specifically: calculates the cosine value of the included angle between each pair of consecutive acceleration vectors in the acceleration vector sequence within the time window; and takes the arithmetic average of all calculated cosine values to obtain an average cosine value as the quantitative indicator of the direction consistency. The intent authenticity discrimination unit generates the weighted interaction acceleration by the following relation , , wherein is the original amplitude of the acceleration of the interaction behavior, is the quantified indicator of the direction consistency, i.e., the average cosine value, is a weight function stored in the memory and monotonically increasing with the time, and the value range of the weight function is zero to one.

2. A fuzzy logic based display resource allocation system according to claim 1, wherein, The cycle adaptive adjustment unit dynamically adjusts the running cycle of the fuzzy logic reasoning engine module, specifically: determines an expected calculation delay according to the real-time temperature value and the temperature and calculation delay characteristic model; and subtracts the expected calculation delay from a baseline running cycle to obtain a new and shortened running cycle.

3. The fuzzy logic based display resource allocation system according to claim 1, wherein, The fuzzy logic reasoning engine module is further configured to: take a value representing the content rendering complexity associated with the interaction behavior as a second fuzzification input, and determine the resource allocation priority value in combination with the weighted interaction acceleration and the value of the content rendering complexity.

4. The fuzzy logic based display resource allocation system of claim 1, wherein, The fuzzy rule base contains the following rule: when the fuzzification value of the weighted interaction acceleration falls within a high-intensity interaction interval, the resource allocation priority value is determined as a highest priority value, and this rule is executed independently of the state of any other fuzzification input.

5. The fuzzy logic based display resource allocation system according to claim 1, wherein, The system further comprises a collaborative intention aggregation unit configured to: when the interactive dynamic feature acquisition module detects at least two interaction behaviors with different spatial positions within a set time window, and the intention authenticity discrimination unit outputs high-intensity weighted interaction acceleration for at least two interaction behaviors, confirm that a collaborative event has occurred; and based on the confirmation of the collaborative event, trigger a collaborative allocation strategy that concentrates computing resources in the area where the at least two interaction behaviors are located.

6. The fuzzy logic based display resource allocation system according to claim 1, wherein, The system further comprises a cognitive load state tracking unit configured to: calculate an exponentially moving average value representing the user's sustained cognitive load based on a series of resource allocation priority values output by the fuzzy logic inference engine module over a period of time; and automatically adjust the presentation of all notification information in the system, except for the alert information, to reduce cognitive interference to the user when the exponentially moving average value exceeds a load threshold value stored in the memory.

Images