A narrowband cluster time slot conflict prediction optimization method and system

Abstract

The application discloses a narrowband cluster time slot conflict prediction optimization method and system, and particularly relates to the technical field of wireless communication; the application constructs a conflict feature vector by fusing historical behavior characteristics and associated group characteristics, accurately predicts a time slot conflict probability in combination with a weighted fusion logic and a mapping rule, breaks through the limitation of traditional allocation only according to a real-time idle state, effectively avoids high conflict risk time slots, significantly reduces the probability of problems such as data transmission interruption and bit error rate increase, improves the reliability and stability of key business transmission, and is suitable for complex communication scenes such as multi-subject competition and random initiation of business.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wireless communication, more particularly, the present application relates to a narrowband cluster time slot conflict prediction optimization method and system. BACKGROUND

[0002] In the field of wireless communication, narrowband cluster communication system is widely used in public safety, transportation, energy scheduling and other key industry scenarios due to its high spectrum utilization, wide coverage and strong anti-interference ability. The core of such system is to realize efficient concurrent transmission of multi-user and multi-service through dynamic allocation of time slot resources, and the rationality of time slot allocation directly determines the stability and service quality of the communication system.

[0003] However, the existing narrowband cluster time slot conflict prediction optimization has the following technical problems in actual application process:

[0004] On the one hand, the traditional time slot allocation strategy is mostly based on the simple judgment of the current idle state of the time slot, lacking in-depth analysis of the historical data of the time slot, so as to realize the prediction ability of the conflict risk in the future use process. Since the initiation time and occupation time of different services are random, and multiple communication subjects may compete for the same time slot resource at the same time, only according to the real-time idle state to allocate the time slot, it is easy to lead to time slot conflict, causing data transmission interruption, high error rate and even service failure, which seriously affects the reliability of key services;

[0005] On the other hand, the existing technology often only considers the basic conditions such as bandwidth matching in the time slot screening process, ignoring the historical use characteristics, current signal state, service adaptability and other multi-dimensional key factors of the time slot. For example, although some time slots are currently in idle state, they have high historical conflict rate, poor signal stability, or low adaptability to current service type, and allocating such time slots will significantly reduce the communication quality;

[0006] At the same time, there is lack of feedback mechanism for the actual use effect after time slot allocation, which cannot dynamically optimize the subsequent allocation strategy according to the actual situation such as conflict events and transmission failure, resulting in insufficient intelligent level of time slot allocation, and it is difficult to adapt to the complex and changeable communication scene demand.

[0007] Therefore, a narrowband cluster time slot conflict prediction optimization method and system are proposed. SUMMARY

[0008] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a narrowband cluster time slot conflict prediction optimization method and system.

[0009] In order to achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0010] A narrowband cluster time slot conflict prediction optimization method comprises:

[0011] S1: Analyze the attributes of a new service request, filter all time slots that meet the service bandwidth requirements and have a current state marker of idle, and form an initial candidate time slot set;

[0012] S2: For each candidate time slot in the initial candidate time slot set, extract its associated historical behavior features and associated group features for conflict prediction from the real-time feature database, and construct a standardized conflict feature vector;

[0013] S3: For any candidate time slot, after weighting and fusion processing of its current conflict feature vector, output the probability coefficient of the candidate time slot occurring conflict in the future prediction window and convert it into a conflict probability;

[0014] S4: According to the conflict probability of all candidate time slots, after screening the safe candidate set, for each time slot in the safe candidate set, extract its associated real-time state features and allocation performance features for time slot evaluation from the real-time feature database, and construct an allocation feature vector. Based on the allocation feature vector, determine the optimal time slot allocation to the current service request;

[0015] S5: After completing the time slot allocation, continuously monitor whether the time slot has occurred conflict or transmission failure in actual use, and if conflict or transmission failure occurs, mark it as allocation error conflict event and transmission failure record respectively and store it in the real-time feature database.

[0016] Specifically, the conflict feature vector in S2 step includes a conflict performance coefficient;

[0017] Identify the time zone in which the future prediction time window is located, and extract the historical occupancy rate and historical conflict rate of the target time slot in the past time zone from the historical behavior features;

[0018] Statistically, the ratio of the total occupancy time of the candidate time slot in the past time zone to the total time of the time zone is the historical occupancy rate;

[0019] Traverse all use records of the candidate time slot in the past time zone, and calculate the proportion of the number of conflict occurrences in the total number of time slot occupancy requests, which is the historical conflict rate;

[0020] After weighting and fusion processing of the historical occupancy rate and the historical conflict rate, the conflict performance coefficient of the candidate time slot is output.

[0021] Specifically, the conflict feature vector in S2 step also includes a group performance coefficient;

[0022] Extract the historical occupation subject ID of the candidate time slot from the association group characteristics, identify the subject ID in the online state at the current time point, and match it with the historical occupation subject ID of the candidate time slot, count the number of matching successful subject IDs as the logical competition number;

[0023] For the matching successful subject ID, extract the number of times it sends a service request to the candidate time slot in the past time zone, and accumulate it as the group request number of the candidate time slot;

[0024] Take the group request number as the numerator and the total time length of the time zone as the denominator to calculate the ratio as the request competition frequency;

[0025] For the logical competition number and the request competition frequency, perform weighted fusion processing to output the group performance coefficient of the candidate time slot.

[0026] Specifically, the output conflict probability process in S3 step;

[0027] Extract the conflict performance coefficient and the group performance coefficient from the conflict feature vector;

[0028] Combine the reference coefficient set constructed in advance to perform weighted fusion processing using the weighted fusion logic to output the probability coefficient;

[0029] Use the pre-constructed probability coefficient-conflict probability mapping rule to convert the probability coefficient of the candidate time slot into the conflict probability.

[0030] Specifically, the process of screening the safe candidate set in S4 step;

[0031] Set an acceptable highest conflict probability, and first filter out all time slots with a conflict probability lower than the highest conflict probability from the initial candidate time slot set to form a safe candidate set.

[0032] Specifically, the logic of determining the optimal time slot using the allocation feature vector in S4 step;

[0033] For each time slot in the safe candidate set, output the time slot evaluation coefficient after fusing the allocation feature vector;

[0034] The allocation feature vector includes the conflict degree coefficient, the signal state coefficient, the service matching coefficient, and the actual feedback coefficient.

[0035] Specifically, the signal state coefficient construction logic in S4 step;

[0036] For each time slot in the safe candidate set, extract the signal strength and adjacent channel interference strength of the time slot in the latest statistical period from the real-time state characteristics;

[0037] The mean value and variance of the signal strength of the time slot at each time point in the statistical period are calculated respectively as the signal mean value and signal variance.

[0038] The adjacent channel interference strength of the time slot at each time point in the statistical period is identified, and the maximum value in the period is taken as the adjacent channel interference value of the time slot.

[0039] Based on the signal mean value, signal variance and adjacent channel interference value of the time slot in the statistical period, the signal state coefficient of the time slot is output.

[0040] Specifically, the service matching coefficient construction logic in S4 step;

[0041] The service type of the current service request is identified, the number of times that the time slot successfully carries a service of the same type as the current service type in history is extracted from the allocation performance characteristics, and the proportion of the number of times in the total number of successful carries is calculated as the service type adaptation rate of the time slot.

[0042] Meanwhile, the historical transmission failure record of the time slot is extracted, and the number of failures is counted after screening and searching in the failure record for the same service type as the current service request.

[0043] The failure rate of the service type of the time slot is calculated by taking the number of failures as the numerator and the total number of historical transmission failures as the denominator.

[0044] Based on the service type adaptation rate and the service type failure rate of the time slot, the service matching coefficient is output.

[0045] Specifically, the conflict degree coefficient and the actual feedback coefficient construction logic in S4 step;

[0046] The conflict degree coefficient is obtained by taking the conflict probability of the time slot as the numerator and the highest conflict probability as the denominator.

[0047] The selection record of the time slot as the optimal time slot in history is extracted, and the monitoring result after each selection is identified, i.e. the number of allocation conflict events is counted, and the proportion of the number of allocation conflict events in the total number of selections is calculated as the actual feedback coefficient.

[0048] A narrowband cluster time slot conflict prediction optimization system, comprising:

[0049] A candidate set generation module: analyzing the service type, priority and required bandwidth attributes of a new service request, screening time slots that meet the bandwidth requirements and have a current state mark of idle, and forming an initial candidate time slot set;

[0050] A conflict vectorization module: for each candidate time slot in the initial candidate time slot set, extracting historical behavior characteristics and associated group characteristics from a real-time feature database to construct a standardized conflict feature vector;

[0051] Conflict risk prediction module: based on the conflict feature vector, the probability coefficient is calculated through the weighted fusion logic combined with the reference coefficient set, and the conflict probability is converted through the mapping rule;

[0052] Slot screening allocation module: according to the conflict probability, the safe candidate set is screened, the real-time state feature and the allocation performance feature are extracted, the allocation feature vector is constructed, the time slot evaluation coefficient is calculated, and the optimal time slot is allocated to the current service;

[0053] Data updating module: monitoring the actual use state after time slot allocation, recording conflict and transmission failure events, updating real-time feature database and calculating actual feedback coefficient.

[0054] Technical effects and advantages of the present application:

[0055] By fusing historical behavior features and associated group features to construct a conflict feature vector, combining weighted fusion logic and mapping rules to accurately predict time slot conflict probability, breaking through the limitations of traditional allocation based on real-time idle state, effectively avoiding high conflict risk time slots, significantly reducing the probability of data transmission interruption, error rate increase and other problems, improving the reliability and stability of key business transmission, adapting to complex communication scenarios with multiple subjects competing and random business initiation;

[0056] By constructing an allocation feature vector containing conflict degree, signal state, business matching and actual feedback coefficient, multi-dimensional comprehensive evaluation of time slots is realized, which not only considers real-time and historical performance such as signal stability and business adaptation, but also selects the optimal time slot through quantitative evaluation, avoiding the communication quality short board caused by single-dimensional screening, making the time slot allocation more suitable for business needs, and improving the resource utilization efficiency and service quality of narrowband trunking system. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 A narrowband trunking time slot conflict prediction optimization method flow chart of the present application;

[0058] Figure 2 A narrowband trunking time slot conflict prediction optimization system principle diagram of the present application. DETAILED DESCRIPTION

[0059] 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 only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0060] Embodiment 1

[0061] As Figure 1As shown, a narrowband cluster time slot conflict prediction optimization method comprises:

[0062] Request analysis and candidate set generation: analyze the attributes of the new service request, including service type, priority, and required bandwidth; according to the current frequency-time slot resource state diagram, filter out all time slots that meet the service bandwidth requirements and have current state marks as idle, forming an initial candidate time slot set;

[0063] Supplementary explanation: first analyze the new service request to determine the service type, priority, and required bandwidth, retrieve the system real-time frequency-time slot resource state diagram, extract the bandwidth capacity and current state of each time slot, and check the time slots one by one: first determine whether the bandwidth meets the service requirements (if a single time slot is insufficient, filter out continuous idle time slot combinations to ensure that the total bandwidth meets the requirements), then confirm that the state mark is "idle" and there is no reservation or occupation record, and then aggregate all time slots (or time slot combinations) that meet the bandwidth requirements and idle state at the same time, format the time slot ID, frequency block, bandwidth, and other information, and form an initial candidate time slot set.

[0064] Conflict performance vectorization: for each candidate time slot in the initial candidate time slot set, extract its associated historical behavior characteristics and associated group characteristics for conflict prediction from the real-time feature database, and construct a standardized conflict feature vector;

[0065] Specifically:

[0066] Identify the time zone in which the future prediction time window is located, and extract the historical occupation rate and historical conflict rate of the target time slot in the past time zone from the historical behavior characteristics;

[0067] For example, if the time zone of the future prediction time window is Monday, the historical data extracted is the past Monday, and the data of multiple past Mondays can be extracted for mean processing as the final extracted data.

[0068] Calculate the total occupation time length of the candidate time slot in the past time zone (the cumulative value of all valid use periods), and the ratio of the total occupation time length to the total time length of the time zone is the historical occupation rate;

[0069] The calculation formula is: ;

[0070] Example: the total time length of the time zone is 24 hours, the total occupation time length of the time slot is 12 hours, then , reflecting the "busy degree" of the time slot in the past.

[0071] Traverse all use records of the candidate time slot in the past time zone, filter out conflict events caused by multiple subject competition and resource overlap, and calculate the proportion of the number of conflict occurrences in the total number of time slot occupation requests, which is the historical conflict rate;

[0072] Supplement: Conflict events need to meet the judgment standard of "initiating occupation request and not successfully accessing", excluding access failure caused by business initiative abandonment and device failure;

[0073] Example: The time slot in the window is requested to be occupied 60 times, of which 12 times are not successfully accessed due to conflict, so the historical conflict rate is 20%, reflecting the "intensity of competition" in the past.

[0074] After the historical occupation rate and the historical conflict rate are comprehensively processed, the conflict performance coefficient of the candidate time slot is output;

[0075] That is, the historical occupation rate and the historical conflict rate are multiplied by the preset occupation weight and conflict weight respectively, and then summed to obtain the conflict performance coefficient, and the sum of the occupation weight and the conflict weight is one.

[0076] From the terminal registration information in the associated group characteristics, the historical occupation subject ID (such as device ID, user ID, service ID) of the candidate time slot is extracted, the subject ID in the online state at the current time point is identified, and is matched with the historical occupation subject ID of the candidate time slot. The number of subject IDs that match successfully is counted as the logical competition number;

[0077] The number of subjects online at the current time is counted: real-time monitoring of the total number of subjects in the competitive terminal group that are "online and have occupied the corresponding time slot", reflecting the current group's competition base for the time slot.

[0078] For the matched subject ID, the number of times it sends a service request to the candidate time slot in the past time zone is extracted, and is accumulated as the group request number of the candidate time slot;

[0079] The group request number is the numerator and the total time length of the time zone is the denominator to calculate the ratio as the request competition frequency;

[0080] After the logical competition number and the request competition frequency are comprehensively processed, the group performance coefficient of the candidate time slot is output;

[0081] That is, after the logical competition number and the request competition frequency are normalized, the logical competition number and the request competition frequency are multiplied by the preset number weight and frequency weight respectively, and then summed to obtain the group performance coefficient.

[0082] Conflict risk prediction: for any candidate time slot, after extracting its current conflict feature vector, the pre-constructed prediction evaluation logic is used for comprehensive processing, and the probability coefficient of the candidate time slot occurring conflict in the future prediction window is output and converted into the conflict probability;

[0083] Specifically:

[0084] extracting conflict performance coefficients and group performance coefficients from the conflict feature vector;

[0085] After using the weighted fusion logic to comprehensively process in combination with the pre-constructed reference coefficient set, output the probability coefficient;

[0086] Probability coefficient processing logic:

[0087] Calculate using the formula Output the probability coefficient ; wherein represents the conflict performance coefficient and the group performance coefficient; is the preset conflict allowed coefficient and the group allowed coefficient in the reference coefficient set; respectively, and .

[0088] The reference coefficient set sets the highest value that the conflict performance coefficient and the group performance coefficient can accept based on the expected current service request.

[0089] Using the pre-constructed probability coefficient-conflict probability mapping rule, convert the probability coefficient of the candidate time slot into the conflict probability;

[0090] Supplementary explanation of the mapping rule: that is, by setting the probability coefficient corresponding to each group of coefficient intervals, and each group interval corresponds to a conflict probability; the conflict probability range is set to 1-100%, and the higher the probability coefficient, the higher the corresponding matching conflict probability.

[0091] Dynamic time slot screening and allocation: according to the conflict probability of all candidate time slots, screen the safe candidate set, and for each time slot in the safe candidate set, extract its associated, real-time state features and allocation performance features for time slot evaluation from the real-time feature database, and construct a standardized allocation feature vector, based on the allocation feature vector Determine the optimal time slot allocation to the current service request;

[0092] Specifically:

[0093] Set an acceptable highest conflict probability, and first filter out all time slots with a conflict probability lower than the highest conflict probability in the initial candidate time slot set to form a safe candidate set;

[0094] Supplementary explanation, if the safe candidate set is empty (i.e. all idle time slots have too high conflict probability), then start the suboptimal strategy: select the time slot with the lowest conflict probability for allocation, and pop up a "high-risk warning" prompt, or start a delay admission control, and temporarily store the service request in the waiting queue, and then allocate it after the network state improves (if there is a low-risk time slot released).

[0095] For each time slot in the security candidate set, the signal strength and adjacent channel interference strength of the time slot in the latest statistical cycle are extracted from the real-time state features;

[0096] Data validity screening: abnormal data (such as RSSI mutation value caused by equipment failure, empty value of collection failure) is eliminated, 3σ criterion is used to identify extreme values (such as abnormal points with RSSI lower than -120 dBm or higher than -30 dBm) beyond the reasonable range, and linear interpolation method is used to complete the missing data, so as to ensure the integrity and accuracy of the parameter data.

[0097] For the signal strength of each time point in the statistical cycle, the mean and variance are calculated respectively, and the signal mean and signal variance are obtained;

[0098] Signal mean ; the signal strength of each time point in the statistical cycle after screening Take the arithmetic mean; i is the time point number, N is the total number of time points, which reflects the overall level of the current signal strength of the time slot;

[0099] Signal variance ; the greater the variance, the more intense the signal strength fluctuation, and the more unstable the communication environment.

[0100] The adjacent channel interference strength of each time point in the statistical cycle is identified, and the maximum value in the cycle is taken as the adjacent channel interference value of the time slot; which directly reflects the influence degree of adjacent channel interference on the communication quality of the time slot;

[0101] Based on the signal mean, signal variance and adjacent channel interference value of the time slot in the statistical cycle, the signal state coefficient of the time slot is output;

[0102] Signal state coefficient calculation process:

[0103] According to the attributes of the service request, i.e. service type, priority and required bandwidth, the expected signal mean, acceptable signal variance and acceptable adjacent channel interference value corresponding to the signal mean, signal variance and adjacent channel interference value respectively are set;

[0104] The formula is used for calculation, and the signal state coefficient is output. represents the signal mean, signal variance and adjacent channel interference value of the time slot; represents the expected signal mean, acceptable signal variance and acceptable adjacent channel interference value; is the set weight coefficient, and the sum of the three groups of coefficients is one.

[0105] Identify the service type of the current service request, extract the number of times the service type that is consistent with the current service type in the history of the time slot from the allocation performance characteristics, and calculate the proportion of it in the total number of successful times as the service type adaptation rate of the time slot;

[0106] A higher compatibility rate indicates that the hardware characteristics and transmission protocol of the time slot are more compatible with the current business, reducing the risk of compatibility failure.

[0107] Simultaneously, extract historical transmission failure records (such as bit error rate and retransmission) from the time slots, and then filter and count the failures from the failure records to find those belonging to the same service type as the current service request.

[0108] The ratio of the number of failures to the total number of historical transmission failures is used to calculate the service type failure rate of the time slot.

[0109] Based on the service type adaptation rate and service type failure rate of the time slot, output the service matching coefficient;

[0110] Business matching coefficient calculation process:

[0111] Using formula Perform calculations and output the business matching coefficient. ;in This represents the business type compatibility rate and the business type failure rate; These are the weighting coefficients, and the sum of the two sets of coefficients is one.

[0112] The conflict level coefficient, signal state coefficient, service matching coefficient, and actual feedback coefficient are used as the allocation feature vector; for each time slot in the security candidate set, the time slot evaluation coefficient is output after fusing the allocation feature vector.

[0113] The conflict severity coefficient is obtained by calculating the ratio of the conflict probability of the time slot to the highest conflict probability.

[0114] Select the time slot with the highest time slot evaluation coefficient as the optimal time slot and allocate it to the current business request;

[0115] The calculation process of the time slot evaluation coefficient:

[0116] Assigning feature vectors ;in These represent the conflict level coefficient and the actual feedback coefficient, respectively.

[0117] After standardizing all the coefficients in the assigned feature vector, the formula is used... Calculate the time slot evaluation coefficient ;in The weighting coefficients are set, and the sum of the coefficients is one.

[0118] Slot history allocation quantification: after completing the slot allocation, continuously monitor whether the slot conflicts in actual use, whether the transmission fails, and if a conflict or transmission failure occurs, mark it as an allocation error conflict event and a transmission failure record respectively, and store it in the real-time feature database;

[0119] Actual feedback coefficient: extract the selection record of the slot as the optimal slot in history, identify the monitoring result after each selection, that is, count the number of allocation conflict events, calculate the proportion of the number of allocation conflict events in the total number of selections, and take it as the actual feedback coefficient.

[0120] Embodiment 2

[0121] Please refer to Figure 2 Based on the narrowband cluster slot conflict prediction optimization method provided in Embodiment 1 of the present application, Embodiment 2 of the present application proposes a narrowband cluster slot conflict prediction optimization system. Embodiment 2 is only a preferred mode of Embodiment 1, and the implementation of Embodiment 2 will not affect the separate implementation of Embodiment 1.

[0122] Specifically, the narrowband cluster slot conflict prediction optimization system provided by Embodiment 2 of the present application comprises:

[0123] A candidate set generation module is configured to analyze the service type, priority, and required bandwidth attribute of a new service request, filter slots that meet the bandwidth requirement and have an idle state label, and form an initial candidate slot set.

[0124] A conflict vectorization module is configured to extract historical behavior features and associated group features from the real-time feature database for each candidate slot in the initial candidate slot set, and construct a standardized conflict feature vector.

[0125] A conflict risk prediction module is configured to calculate a probability coefficient based on the conflict feature vector and by using a weighting fusion logic combined with a reference coefficient set, and then convert the probability coefficient into a conflict probability by using a mapping rule.

[0126] A slot selection and allocation module is configured to select a safe candidate set according to the conflict probability, extract real-time state features and allocation performance features of the slot, construct an allocation feature vector and calculate a slot evaluation coefficient, and determine an optimal slot to allocate to the current service.

[0127] A data updating module is configured to monitor the actual use state of the slot after allocation, record conflict and transmission failure events, update the real-time feature database, and calculate an actual feedback coefficient.

[0128] The above formulas are all dimensionless values calculated, and specific dimensionless can be standardized and other means, and will not be described here, the formula is obtained by collecting a large amount of data to simulate the formula of the nearest real situation, and the preset parameters in the formula are set by the person skilled in the art according to the actual situation.

[0129] The above embodiments can be realized wholly or partially by software, hardware, firmware or any combination thereof. When realized by software, the above embodiments can be realized wholly or partially in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are wholly or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable devices. The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired or wireless (such as infrared, wireless, microwave, etc.) mode. The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center and the like containing one or more available medium sets. The available medium can be a magnetic medium (such as a floppy disk, an ATA hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid state ATA hard disk.

[0130] It should be understood that in various embodiments of the present application, the size of the sequence number of each process does not mean the order of execution, and the execution order of each process should be determined by its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0131] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0132] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. For example, the described device embodiments are merely schematic. The division of the units is merely logical function division. There can be other division manners in actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0133] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, and can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0134] In addition, each functional unit in the various embodiments of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit.

[0135] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product in essence or the part of the technical solutions that make contributions to the prior art or part of the technical solutions. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile ATA hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0136] The above is merely specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for optimizing narrowband trunking time slot collision prediction, characterized in that, Comprise: S1: Analyze the attributes of the new service request, filter out all time slots that meet the service bandwidth requirements and whose current state markers are idle, and form an initial candidate time slot set; S2: For each candidate time slot in the initial candidate time slot set, extract its associated historical behavior characteristics and associated group characteristics for conflict prediction from the real-time feature database, and construct a standardized conflict feature vector; The conflict feature vector includes a conflict performance coefficient and a group performance coefficient; Identify the time zone where the future prediction time window is located, and extract the historical occupancy rate and historical conflict rate of the target time slot in the past time zone from the historical behavior characteristics; The ratio of the total occupancy time of the candidate time slot in the past time zone to the total time of the time zone is the historical occupancy rate; Traverse all usage records of the candidate time slot in the past time zone, and calculate the proportion of the number of conflict occurrences in the total number of time slot occupancy requests, which is the historical conflict rate; After weighting and fusion processing of the historical occupancy rate and the historical conflict rate, the conflict performance coefficient of the candidate time slot is output; Extract the historical occupancy subject ID of the candidate time slot from the associated group characteristics, identify the subject ID that is in an online state at the current time point, and match it with the historical occupancy subject ID of the candidate time slot, and count the number of matching subject IDs as the logical competition quantity; For the matching subject ID, extract the number of times it sends a service request to the candidate time slot in the past time zone, and add it up as the group request number of the candidate time slot; Calculate the ratio of the group request number to the total time of the time zone as the request competition frequency; After weighting and fusion processing of the logical competition quantity and the request competition frequency, the group performance coefficient of the candidate time slot is output; S3: For any candidate time slot, extract its current conflict feature vector for weighting and fusion processing, and output the probability coefficient of the candidate time slot occurring conflict in the future prediction window and convert it into a conflict probability; Extract the conflict performance coefficient and the group performance coefficient from the conflict feature vector; The formula is used in combination with a pre-constructed reference coefficient set to perform calculation and output a probability coefficient ; wherein represents a conflict performance coefficient and a group performance coefficient; are preset conflict and group allowed coefficients in the reference coefficient set; are preset weight coefficients, respectively, and ; Use the pre-constructed probability coefficient-conflict probability mapping rule to convert the probability coefficient of the candidate time slot into a conflict probability; S4: According to the conflict probability of all candidate time slots, filter the safe candidate set, and for each time slot in the safe candidate set, extract its associated real-time state characteristics and allocation performance characteristics for time slot evaluation from the real-time feature database, and construct an allocation feature vector, and determine the optimal time slot based on the allocation feature vector. Assign it to the current service request; S5: After completing the time slot allocation, continuously monitor whether the time slot has occurred conflict or transmission failure in actual use, and if conflict or transmission failure occurs, mark it as allocation conflict event and transmission failure record respectively and store it in the real-time feature database.

2. The narrowband cluster time slot conflict prediction optimization method according to claim 1, characterized in that: The process of filtering the safe candidate set in step S4; Set an acceptable maximum conflict probability, and first filter out all time slots with a conflict probability lower than the maximum conflict probability from the initial candidate time slot set to form a safe candidate set.

3. The narrowband trunking time slot conflict prediction optimization method according to claim 2, characterized in that: S4 step uses the allocation feature vector to determine the optimal time slot logic; For each time slot of the safe candidate set, the time slot evaluation coefficient is output after the allocation feature vector is fused; The allocation feature vector includes the conflict degree coefficient, the signal state coefficient, the service matching coefficient, and the actual feedback coefficient.

4. The narrowband trunking time slot conflict prediction optimization method according to claim 3, characterized in that: S4 step signal state coefficient construction logic; For each time slot in the safe candidate set, the signal strength and adjacent channel interference strength of the time slot in the latest statistical cycle are extracted from the real-time state feature; For the signal strength of the time slot at each time point in the statistical cycle, the mean and variance are calculated respectively, and the signal mean and signal variance are output; For the adjacent channel interference strength of the time slot at each time point in the statistical cycle, the maximum value in the cycle is identified as the adjacent channel interference value of the time slot; Based on the signal mean, signal variance, and adjacent channel interference value of the time slot in the statistical cycle, the signal state coefficient of the time slot is output.

5. The narrowband trunking time slot conflict prediction optimization method according to claim 3, characterized in that: S4 step service matching coefficient construction logic; Identify the service type of the current service request, extract the number of times that the time slot successfully carries the service of the same type as the current service type from the allocation performance feature, calculate the proportion of the number of times in the total number of successful carries, and output the service type adaptation rate of the time slot; At the same time, extract the time slot historical transmission failure record, and search and statistically analyze the same service type in the failure record to obtain the failure number; Take the failure number as the numerator and the total number of historical transmission failures as the denominator to calculate the ratio, which is the service type failure rate of the time slot; Based on the service type adaptation rate and the service type failure rate of the time slot, the service matching coefficient is output.

6. The narrowband trunking time slot conflict prediction optimization method according to claim 3, characterized in that: S4 step conflict degree coefficient and actual feedback coefficient construction logic; The conflict degree coefficient is calculated by taking the conflict probability of the time slot as the numerator and the highest conflict probability as the denominator; Extract the selection record of the time slot as the optimal time slot in history, identify the monitoring result after each selection, count the number of allocation conflict events, calculate the proportion of the number of allocation conflict events in the total number of selections, and output the actual feedback coefficient.

7. A narrowband trunking timeslot conflict prediction optimization system, applied to the narrowband trunking timeslot conflict prediction optimization method of any one of claims 1-6, characterized in that, It includes: Candidate set generation module: analyze the service type, priority, and required bandwidth attributes of the new service request, filter the time slots that meet the bandwidth requirements and have an idle state mark, and form an initial candidate time slot set; Conflict vectorization module: for each candidate time slot in the initial candidate time slot set, extract the historical behavior features and associated group features from the real-time feature database, and construct a standardized conflict feature vector; Conflict risk prediction module: based on the conflict feature vector, combine the reference coefficient set through weighted fusion logic to calculate the probability coefficient, and then convert it into a conflict probability through mapping rules; Slot screening allocation module: screening the safe candidate set according to the conflict probability, extracting the instant state features and allocation performance features of the time slot, constructing the allocation feature vector and calculating the time slot evaluation coefficient, and determining the optimal time slot allocation to the current service; Data update module: monitoring the actual use state after time slot allocation, recording the conflict and transmission failure events, updating the real-time feature database and calculating the actual feedback coefficient.

Images