A network resource switching method and system based on a wireless fingerprint grid algorithm

Abstract

The present application discloses a network resource switching method and system based on a wireless fingerprint grid algorithm, including: periodically performing co-frequency measurements on a specified number of terminals in a serving cell randomly to collect the wireless characteristics of the terminals; performing blank grid division based on the wireless characteristics of the serving cell and the cells of at most N co-frequency strongest neighboring cells; obtaining the wireless characteristic information of the serving cell and the neighboring cells of different frequencies and different systems collected by each terminal in the geographical area corresponding to each blank grid; scoring the service perception of the serving cell in each grid, and when the service service perception score is lower than the target value, performing the judgment of the optimal cell, and switching the network of the user to the neighboring cell network with the highest final score. Through the present application, the optimal cell of each terminal can be obtained, ensuring that the network of the terminal user is always stable, increasing the trust and satisfaction with the 5G network, and thus improving the 5G traffic splitting ratio.

Description

Technical Field

[0001] This application relates to the field of communication technologies, and in particular, to a network resource switching method and system. Background Art

[0002] With the increasing complexity of the current communication network, 5G network coverage and user perception are the key directions that need to be considered in the wireless network. How to balance the improvement of the 5G offloading ratio and user perception is a current hot topic.

[0003] Currently, traditional 5G offloading ratio improvement solutions mainly rely on manual analysis and judgment, and adopt interoperability strategies such as unified reselection priorities and handover parameters, and adjust the entire network in a "one-size-fits-all" manner, that is, use a unified method to adjust the entire network. This method may ignore the specific service requirements and network usage conditions of different users. The traditional traffic prediction scheme has the following problems: the entire network triggers frequency band migration in a unified manner, unable to take into account the user service perception, resulting in a poor network experience for users and low network resource utilization. This problem will reduce users' trust and satisfaction with the 5G network, thereby leading to a decrease in the 5G offloading ratio. Summary of the Invention

[0004] Based on this, in view of the above technical problems, a network resource switching method and system based on a wireless fingerprint grid algorithm are provided to solve the problem that the existing technology leads to a poor network experience for users.

[0005] In a first aspect, a network resource switching method based on a wireless fingerprint grid algorithm, the method includes:

[0006] Step S1: Periodically perform co-frequency measurements on target terminals in a target area randomly, and obtain wireless feature information of serving cells and neighboring cells of the same frequency as the serving cell collected by multiple target terminals respectively; divide the geographical area where the serving cell obtained by each target terminal and the N neighboring cells with the strongest wireless features of the same frequency as its own serving cell are located into an independent blank grid, thereby completing the division of the blank grid of the target area; the wireless features include reference signal received power (RSRP) and signal-to-downlink interference plus noise ratio (SINR);

[0007] Step S2: Obtain wireless feature information of serving cells and neighboring cells of different frequencies and different systems collected by each terminal in the geographical area corresponding to each blank grid;

[0008] Step S3: Score the service perception of the serving cell of each terminal in each grid. When the service perception score of the serving cell is lower than the target value, judge the neighboring cells of different frequencies and different systems obtained by the terminal to obtain the optimal cell corresponding to each terminal. The judging the neighboring cells of different frequencies and different systems obtained by the terminal to obtain the optimal cell corresponding to each terminal includes:

[0009] Step S31: Determine the inter-frequency and inter-system neighboring cells of each terminal, evaluate the inter-frequency and inter-system neighboring cells of each terminal, calculate the spectral efficiency and suppression distance of the inter-frequency and inter-system neighboring cells, so as to calculate the user perception value of each inter-frequency and inter-system neighboring cell; when evaluating the inter-frequency and inter-system neighboring cells of each terminal, the radio feature information of the inter-frequency and inter-system neighboring cells of each terminal should be the radio feature information collected when the current inter-frequency and inter-system neighboring cell is the serving cell of other terminals;

[0010] Step S32: Compare the radio feature information of the inter-frequency and inter-system neighboring cells of each terminal with a set threshold to screen out the neighboring cells that meet the set threshold, and then score the neighboring cells that meet the set threshold according to the network frequency band and radio feature information to obtain the 4G / 5G-based residence value of each inter-frequency and inter-system neighboring cell;

[0011] Step S33: Obtain the final score of each neighboring cell according to the user perception value and the 4G / 5G-based residence value of each neighboring cell; the neighboring cell with the highest score is the optimal cell;

[0012] Step S4: Switch the network of each terminal to the network of the optimal cell.

[0013] In the above solution, optionally, the service perception of the serving cell of the terminal is scored in the following way:

[0014] Obtain the playback data when the terminal plays a video, and the playback data includes video source quality, buffering duration, number of freezes, and proportion of freeze recovery duration;

[0015] Calculate the first service perception score of the serving cell through the following formula:

[0016] VMOS = f(vQuality, vLoading, vStallingfreq, vStallingrec, W n )

[0017] where VMOS is the video user perception score, vQuality is the video source quality, vLoading is the buffering duration, vStallingfreq is the number of freezes, vStallingrec is the proportion of freeze recovery duration, and W n is the weight ratio of each factor;

[0018] When the first service perception score of the serving cell is less than the first set value, then evaluate the neighboring cells of the inter-frequency and inter-system of the terminal.

[0019] In the above solution, further optionally, the service perception of the serving cell of the terminal is also scored in the following way:

[0020] Obtain the terminal resource load and channel quality, and obtain the overall load of the serving cell of the terminal. Then, calculate the upper limit of the achievable rate that the terminal can currently obtain through the following formula:

[0021] R A = α × BW × (1 - Load1 + Load2) × log2(1 + β × SINR(down));

[0022] Where, α and β are efficiency factors, which can be obtained by learning and fitting the historical data between the rate and each parameter; BW is the system bandwidth; Load1 is the overall resource load of the serving cell, Load2 is the current terminal resource load, and R A is the achievable rate of the serving cell; SINR(down) is the downlink signal-to-interference-plus-noise ratio;

[0023] Calculate the rate margin between the achievable rate of the user in the serving cell and the service demand rate R T through the following formula:

[0024] R GAP = MAX{(R A - R T ) / R A , 0};

[0025] The rate margin between the achievable rate of the user in the serving cell and the service demand rate R T is the second service perception score;

[0026] When the second service perception score of the serving cell is greater than the second set value, then evaluate the inter-frequency and inter-system neighboring cells of the terminal.

[0027] In the above solution, optionally, the calculation of the spectral efficiency of the inter-frequency and inter-system neighboring cells includes:

[0028] Calculate the path loss obtained according to the downlink received RSRP and transmit power of the neighboring cell of the inter-frequency and inter-system, and combine the interference information obtained from the interface between the base stations to calculate the uplink SINR of the neighboring cell of the inter-frequency and inter-system. Then, calculate the spectral efficiency through the uplink SINR of the neighboring cell of the inter-frequency and inter-system; the calculation of the spectral efficiency through the uplink SINR of the neighboring cell of the inter-frequency and inter-system is specifically calculated through the following formula:

[0029] SE = α × log2(1 + β × SINR(up))

[0030] Where, α and β are efficiency factors, and SINR(up) is the uplink signal-to-interference-plus-noise ratio.

[0031] In the above solution, optionally, the calculation of the suppression distance of the inter-frequency and inter-system neighboring cells includes:

[0032] Obtain the traffic growth depression point value of the neighboring cell and the traffic value of the current state, and calculate the depression distance of the neighboring cell according to the traffic growth depression point value of the neighboring cell and the traffic value of the current state through the following formula:

[0033] D s = P s - P a

[0034] Where: P s is the traffic growth depression point value, and P a is the traffic value of the current state.

[0035] In the above solution, further optionally, the calculating the user perception value of each inter-frequency and inter-system neighboring cell includes:

[0036] Calculate the normalized spectral efficiency according to the spectral efficiency of the neighboring cell of the inter-frequency and inter-system through the following formula:

[0037] SE scale = SE / SE max ;

[0038] Where, SE max is the maximum spectral efficiency of the neighboring cell of the inter-frequency and inter-system;

[0039] Calculate the user perception value according to the reachable rate, spectral efficiency and depression distance of the neighboring cell of the inter-frequency and inter-system through the following formula:

[0040] P = {Max(R A , 0)}β × SE scale θD S Φ

[0041] Where β and θ are weighting factors, R A is the reachable rate calculated by taking the neighboring cell of the current terminal as the serving cell of other terminals, SE max is the maximum spectral efficiency, D S is the depression distance, and Φ is a variable of D S .

[0042] In the above solution, optionally, the setting threshold is set as SINR ≥ 3 and RSRP ≥ -110 dBm.

[0043] In the above solution, optionally, the scoring the neighboring cells that meet the setting threshold according to the frequency range and technology generation of the network to obtain the 4G / 5G-based residence value of each inter-frequency and inter-system neighboring cell includes:

[0044] Obtain the priority score of the neighboring cell according to the frequency band of the neighboring cell;

[0045] Calculate according to the priority score of the neighboring cell and the radio feature information by the following formula:

[0046]

[0047] Where Q is the residence value of this cell, Y is the frequency band priority score of the neighboring cell, and ε is the residence level weight.

[0048] In the above solution, optionally, the final score of each neighboring cell obtained according to the user perception value of each neighboring cell and the residence value based on 4G / 5G is calculated by the following formula:

[0049] C = γ×P + δ×Q

[0050] = γ{max(R A , 0)}β×γSEscaleθD S Φ + δ×Q

[0051] Where γ and δ are specific gravity factors, β and θ are weighting factors, P is the user perception value, and Q is the residence value based on 4G / 5G.

[0052] In a second aspect, a network resource switching system based on a wireless fingerprint grid algorithm, the system includes:

[0053] Blank grid division module: used to obtain the radio feature information of the serving cell and the neighboring cells with the same frequency as the serving cell collected by multiple target terminals through randomly performing periodic co-frequency measurements on the target terminals in the target area; divide the geographical area where the serving cell and the N neighboring cells with the strongest radio features and the same frequency as its own serving cell obtained by each target terminal are located into an independent blank grid, so as to complete the division of the blank grid of the target area; the radio features include the reference signal received power RSRP and the signal-to-downlink interference plus noise ratio SINR;

[0054] Blank grid data filling module: used to obtain the radio feature information of the serving cell and the neighboring cells with different frequencies and different systems collected by each terminal in the geographical area corresponding to each blank grid;

[0055] Service business perception scoring module: used to score the service business perception of the serving cell of each terminal in each grid. When the service business perception score of the serving cell is lower than the target value, judge the neighboring cells with different frequencies and different systems obtained by the terminal to obtain the optimal cell corresponding to each terminal;

[0056] Optimal cell determination module: For the inter-frequency and inter-system neighboring cells of each terminal, evaluate the inter-frequency and inter-system neighboring cells of each terminal, calculate the spectral efficiency and suppression distance of the inter-frequency and inter-system neighboring cells, so as to calculate the user perception value of each inter-frequency and inter-system neighboring cell; when evaluating the inter-frequency and inter-system neighboring cells of each terminal, the radio feature information of the inter-frequency and inter-system neighboring cells of each terminal should be the radio feature information collected when the current inter-frequency and inter-system neighboring cell is the serving cell of other terminals, compare the radio feature information of the inter-frequency and inter-system neighboring cells of each terminal with the set threshold to screen out the neighboring cells that meet the set threshold, and then score the neighboring cells that meet the set threshold according to the network frequency band and radio feature information to obtain the 4G / 5G-based residence value of each inter-frequency and inter-system neighboring cell; obtain the final score of each neighboring cell according to the user perception value and 4G / 5G-based residence value of each neighboring cell; the neighboring cell with the highest score is the optimal cell;

[0057] Network switching module: Used to switch the network of each terminal to the network of the optimal cell.

[0058] This application has at least the following beneficial effects:

[0059] After the fingerprint grid is established, after the terminal reports the RSRP and SINR measurement information of the serving cell and neighboring cells, match the logical grid information. For example, for the terminal, if the current serving cell is not the optimal cell for 4 / 5G residence and perception in this grid, directly perform inter-frequency / inter-system neighboring cell handover to the target cell to achieve accurate and fast carrier selection / handover and carrier coordination; if the current serving cell is the optimal cell in this grid, the terminal occupies this cell without change. This application can quickly switch to the cell with the optimal 4 / 5G residence and perception: the information reported by the terminal is only the RSRP and SINR of the same frequency or inter-frequency / inter-system, and the information reported by the terminal is less, avoiding unnecessary terminal measurements; thus ensuring that the terminal user network is always stable, increasing the trust and satisfaction with the 5G network, and thereby increasing the 5G traffic splitting ratio. Description of the Drawings

[0060] Figure 1 It is a flowchart of a network resource switching method based on a wireless fingerprint grid algorithm provided by an embodiment of this application;

[0061] Figure 2 It is a schematic diagram of blank grid division provided by an embodiment of this application;

[0062] Figure 3 It is a schematic diagram of suppression distance prediction provided by an embodiment of this application;

[0063] Figure 4 It is an improvement effect diagram of the 5G traffic splitting ratio optimization scheme in the prior art;

[0064] Figure 5Effect diagram of the 5G traffic splitting ratio optimization solution of this application;

[0065] Figure 6 Effect diagram of the improvement of the FTP upload rate by using the method of this application;

[0066] Figure 7 Effect diagram of the improvement of the Douyin video service by using the method of this application;

[0067] Figure 8 Effect of the improvement of the video service by using the method of this application;

[0068] Figure 9 A detailed flowchart of a network resource switching method based on a wireless fingerprint grid algorithm is provided for an embodiment of this application. Detailed implementation manners

[0069] In order to make the objectives, technical solutions and advantages of this application clearer and more understandable, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0070] In one embodiment, as Figure 1 shown, a network resource switching method based on a wireless fingerprint grid algorithm is provided. The method includes:

[0071] Step S1: Periodically perform co-frequency measurements on target terminals in a target area randomly to obtain wireless feature information of serving cells and neighboring cells with the same frequency as the serving cell collected by multiple target terminals respectively; divide the geographical area where the serving cell obtained by each target terminal and the N neighboring cells with the strongest wireless features and the same frequency as its own serving cell are located into an independent blank grid, so as to complete the division of the blank grid of the target area; the wireless features include reference signal received power (RSRP) and signal-to-downlink interference plus noise ratio (SINR).

[0072] In step S1, a specified number of UEs in the cell are randomly selected to initiate periodic co-frequency measurements, and models are established for wireless features (RSRP reference signal received power, SINR signal interference plus noise ratio). And perform blank grid division according to the RSRP and SINR values of the serving cell and at most two co-frequency strongest neighboring cells. Each grid represents the wireless environment characteristics where the user is located; as Figure 2 shown is a schematic diagram of blank grid division.

[0073] Step S2: Obtain the wireless feature information of the serving cell and neighboring cells with different frequencies and different systems collected by each terminal in the geographical area corresponding to each blank grid.

[0074] In step S2, the blank grids are filled with inter-frequency and inter-system neighbor cell information, mainly including inter-frequency and inter-system measurement report information, which includes the RSRP and SINR information of each inter-frequency and inter-system neighbor cell. When the cumulative number of neighbor cell measurement data of inter-frequency and inter-system in a grid reaches 3, it indicates that the grid is available.

[0075] Step S3: Score the service perception of the serving cell of each terminal in the blank grid. When the service perception score of the serving cell is lower than the target value, judge the neighbor cells obtained by the terminal to obtain the optimal cell corresponding to each terminal. The judging the inter-frequency and inter-system neighbor cells obtained by the terminal to obtain the optimal cell corresponding to each terminal includes:

[0076] Step S31: Determine the inter-frequency and inter-system neighbor cells of each terminal, evaluate the inter-frequency and inter-system neighbor cells of each terminal, calculate the spectral efficiency and suppression distance of the inter-frequency and inter-system neighbor cells, so as to calculate the user perception value of each inter-frequency and inter-system neighbor cell; when evaluating the inter-frequency and inter-system neighbor cells of each terminal, the wireless feature information of the inter-frequency and inter-system neighbor cells of each terminal should be the wireless feature information collected when the current inter-frequency and inter-system neighbor cell is the serving cell of other terminals;

[0077] Step S32: Compare the wireless feature information of the inter-frequency and inter-system neighbor cells of each terminal with the set threshold to screen out the neighbor cells that meet the set threshold, and then score the neighbor cells that meet the set threshold according to the network frequency band and wireless feature information to obtain the 4G / 5G-based residence value of each inter-frequency and inter-system neighbor cell;

[0078] Step S33: Obtain the final score of each neighbor cell according to the user perception value and the 4G / 5G-based residence value of each neighbor cell; the neighbor cell with the highest score is the optimal cell;

[0079] Step S4: Switch the network of each terminal to the network of the optimal cell.

[0080] In the above network resource switching method based on the wireless fingerprint grid algorithm, after the fingerprint grid is established, after the terminal reports the RSRP and SINR measurement information of the serving cell and neighboring cells, the logical grid information is matched. For example, for the terminal, if the current serving cell is not the optimal cell for 4 / 5G residence and perception in this grid, it directly performs inter-frequency / inter-system neighboring cell handover to the target cell to achieve accurate and fast carrier selection / handover and carrier coordination; if the current serving cell is the optimal cell in this grid, the terminal occupies this cell without change. This application can quickly switch to the cell with the optimal 4 / 5G residence and perception: the information reported by the terminal is only the RSRP and SINR of the same frequency or inter-frequency / inter-system, and the information reported by the terminal is less, avoiding unnecessary terminal measurements; thereby ensuring that the terminal user network is always stable, making the user experience stable, increasing the trust and satisfaction with the 5G network, and thus increasing the 5G traffic splitting ratio.

[0081] In one embodiment, the service perception of the serving cell of the terminal is scored in the following manner:

[0082] Obtain the playback data when the terminal plays a video, where the playback data includes video source quality, buffering duration, number of freezes, and the proportion of freeze recovery duration;

[0083] Calculate the first service perception score of the serving cell through the following formula:

[0084] VMOS = f(vQuality, vLoading, vStallingfreq, vStallingrec, W n )

[0085] where VMOS is the video user perception score, vQuality is the video source quality, vLoading is the buffering duration, vStallingfreq is the number of freezes, vStallingrec is the proportion of freeze recovery duration, and W n is the weight proportion of each factor;

[0086] When the first service perception score of the serving cell is greater than or equal to a first set value, the neighboring cells of the terminal with different frequencies and different systems are evaluated.

[0087] In one embodiment, the service perception of the serving cell of the terminal is also scored in the following manner:

[0088] Obtain the terminal resource load and channel quality, and obtain the overall load of the serving cell of the terminal, and calculate the upper limit of the achievable rate that the terminal can currently obtain through the following formula:

[0089] R A= α × BW × (1 - Load1 + Load2) × log2(1 + β × SINR(down));

[0090] Wherein, α and β are efficiency factors, which can be obtained by learning and fitting historical data between the rate and various parameters; BW is the system bandwidth; Load1 is the overall resource load of the serving cell, Load2 is the resource load of the current terminal, R A is the achievable rate of the serving cell; SINR(down) is the downlink signal-to-interference-plus-noise ratio;

[0091] Calculate the rate margin between the achievable rate of the user in the serving cell and the service demand rate R T as follows:

[0092] R GAP = MAX{(R A - R T ) / R A , 0};

[0093] The rate margin between the achievable rate of the user in the serving cell and the service demand rate R T is the second service perception score;

[0094] When the second service perception score of the serving cell is greater than the second set value, evaluate the inter-frequency and inter-system neighboring cells of the terminal.

[0095] In one embodiment, calculating the spectrum efficiency of the inter-frequency and inter-system neighboring cells includes:

[0096] Based on the path loss calculated from the downlink received RSRP and transmit power of the neighboring cells of the inter-frequency and inter-system, and combining with the interference information obtained from the interface between base stations, the uplink SINR of the neighboring cells of the inter-frequency and inter-system can be calculated, and then the spectrum efficiency is calculated through the uplink SINR of the neighboring cells of the inter-frequency and inter-system; The specific calculation of the spectrum efficiency through the uplink SINR of the neighboring cells of the inter-frequency and inter-system is as follows:

[0097] SE = α × log2(1 + β × SINR(up))

[0098] Wherein, α and β are efficiency factors, and SINR(up) is the uplink signal-to-interference-plus-noise ratio.

[0099] In this embodiment, the neighboring cell information is interacted through the 5G inter-station X n interface and the 4 / 5G inter-station X e interface, including: load, interference, DC / CA cooperation relationship, etc.

[0100] In one embodiment, calculating the suppression distance of the inter-frequency and inter-system neighboring cells includes:

[0101] Obtain the traffic growth depression point value of the neighboring cell and the traffic value of the current state, and calculate the depression distance of the neighboring cell according to the traffic growth depression point value of the neighboring cell and the traffic value of the current state through the following formula:

[0102] D s =P s -P a

[0103] Where: P s is the traffic growth depression point value, and P a is the traffic value of the current state.

[0104] In this embodiment, as Figure 3 shown, with the growth of the number of users and resource utilization rate, due to factors such as resource squeezing and increased interference, the process of the increase in network traffic is as follows: first, it shows linearity, then reaches the depression point and reaches the maximum value, and finally gradually decreases.

[0105] In one embodiment, the calculation of the user perception value of each inter-frequency and inter-system neighboring cell includes:

[0106] Calculate the normalized spectral efficiency according to the spectral efficiency of the neighboring cell of the inter-frequency and inter-system through the following formula:

[0107] SE scale =SE / SE max ;

[0108] Where, SE max is the maximum spectral efficiency of the neighboring cell of the inter-frequency and inter-system;

[0109] Calculate the user perception value according to the achievable rate, spectral efficiency and depression distance of the neighboring cell of the inter-frequency and inter-system through the following formula:

[0110] P = {Max(R A ,0)}β×SE scale θD S Φ

[0111] Where, β and θ are weighting factors, R A is the achievable rate calculated by using the neighboring cell of the current terminal as the serving cell of other terminals, SE max is the maximum spectral efficiency, D S is the depression distance, and Φ is a variable of D S .

[0112] In one embodiment, the set threshold is set as SINR≥3 and RSRP≥-110dBm.

[0113] In one embodiment, the method of obtaining the 4G / 5G-based residence value for each inter-frequency and inter-system neighbor cell by scoring neighbor cells that meet a set threshold according to the frequency range and technology generation of the network includes:

[0114] Obtain the priority score of the neighbor cell according to the frequency band of the neighbor cell;

[0115] Calculate according to the following formula based on the priority score of the neighbor cell and the radio feature information:

[0116]

[0117] where Q is the residence value of the cell, Y is the frequency band priority score of the neighbor cell, and ε is the residence level weight.

[0118] In this embodiment, the cells in the grid are scored first according to their frequency bands. If the frequency bands are 5G high frequency, 5G cell medium frequency, 5G low frequency, 4G high frequency, 4G medium frequency, and 4G low frequency, the priority scores are set to 7, 6, 5, 4, 3, and 1 in sequence.

[0119] In one embodiment, the method of obtaining the final score for each neighbor cell according to the user perception value and the 4G / 5G-based residence value of each neighbor cell is calculated by the following formula:

[0120] C = γ×P + δ×Q

[0121] = γ{max(R A , 0)}β×γSEscaleθD S Φ + δ×Q

[0122] where γ and δ are specific gravity factors, β and θ are weighting factors, P is the user perception value, and Q is the 4G / 5G-based residence value.

[0123] In one embodiment, a region is selected for verification: 97 physical stations and a total of 317 cells in a certain city are selected for intelligent traffic splitting pilot verification. The construction period of the logical fingerprint grid construction library is set to 1 hour, the MR reporting period is set to 5 seconds, the number of MRs required for constructing the grid is set to 1000, and the actual constructed situation is viewed to display the current number of grids, which is a total of 2774 fingerprint grids.

[0124] Based on the intelligent traffic splitting ratio improvement scheme of the wireless fingerprint grid algorithm, the 5G traffic splitting ratio is 1.75pp higher than the traditional scheme, and the perceived rate is 20% higher than the traditional scheme. As Figure 4 shown, it is the improvement effect diagram of the 5G traffic splitting ratio optimization scheme of the prior art. As Figure 5 shown, it is the improvement effect diagram of the intelligent traffic splitting ratio scheme of this application.

[0125] As Figure 6As shown in the figure, it is the effect diagram of the improvement of the FTP upload rate of the present invention. To the left of the dotted line, the UE stays in the 5G cell and slowly moves to a poor area; to the right, the UE switches to a better 4G cell in time, and the uplink rate is significantly improved. The uplink rate is increased by 20% compared with the traditional optimization scheme.

[0126] As Figure 7 shown in the figure, it is the effect diagram of the improvement of the Douyin video service of the present application. In the downlink medium packet service, handover is triggered at the perception edge. Before the handover, the service stays in 5G for a long time. After switching to a better 4G cell in time, the video service remains smooth. The downlink rate is increased by 19% compared with the traditional optimization scheme.

[0127] As Figure 8 shown in the figure, it is the effect diagram of the improvement of the video service of the present application. In the downlink small packet service, handover is triggered at the perception edge. The service stays in 5G for a long time. After switching to a better 4G cell in time, the video service remains smooth. The downlink rate is increased by 22% compared with the traditional optimization scheme.

[0128] In one embodiment, a network resource switching system based on a wireless fingerprint grid algorithm, characterized in that the system includes:

[0129] Blank grid division module: used to obtain the wireless feature information of the serving cell and the neighboring cells of the same frequency as the serving cell collected by multiple target terminals by randomly performing periodic co-frequency measurements on the target terminals in the target area; divide the geographical area where the serving cell obtained by each target terminal and the N neighboring cells with the strongest wireless features of the same frequency as its own serving cell are located into an independent blank grid, so as to complete the division of the blank grid of the target area; the wireless features include the reference signal received power RSRP and the signal-to-downlink interference plus noise ratio SINR;

[0130] Blank grid data filling module: used to obtain the wireless feature information of the serving cell and the neighboring cells of different frequencies and different systems collected by each terminal in the geographical area corresponding to each blank grid;

[0131] Service business perception scoring module: used to score the service perception of the serving cell of each terminal in each grid. When the service business perception score of the serving cell is lower than the target value, the neighboring cells of different frequencies and different systems obtained by the terminal are judged to obtain the optimal cell corresponding to each terminal;

[0132] Optimal cell determination module: For the inter-frequency and inter-system neighboring cells of each terminal, evaluate the inter-frequency and inter-system neighboring cells of each terminal, calculate the spectral efficiency and suppression distance of the inter-frequency and inter-system neighboring cells, so as to calculate the user perception value of each inter-frequency and inter-system neighboring cell; when evaluating the inter-frequency and inter-system neighboring cells of each terminal, the wireless feature information of the inter-frequency and inter-system neighboring cells of each terminal should be the wireless feature information collected when the current inter-frequency and inter-system neighboring cell is the serving cell of other terminals, compare the wireless feature information of the inter-frequency and inter-system neighboring cells of each terminal with the set threshold to screen out the neighboring cells that meet the set threshold, and then score the neighboring cells that meet the set threshold according to the network frequency band and wireless feature information to obtain the 4G / 5G-based residence value of each inter-frequency and inter-system neighboring cell; obtain the final score of each neighboring cell according to the user perception value and the 4G / 5G-based residence value of each neighboring cell; the neighboring cell with the highest score is the optimal cell;

[0133] Network switching module: Used to switch the network of each terminal to the network of the optimal cell.

[0134] As Figure 9 shown, the overall idea of the intelligent shunt ratio improvement scheme based on the wireless fingerprint grid algorithm of this application has two parts:

[0135] First, establish a logical fingerprint grid library based on the machine self-learning algorithm. Collect the measurement data reported by the terminals in the cell, and divide the blank grids according to the serving cell and at most 2 co-frequency strongest neighboring cells; fill the RSRP and SINR information of the inter-frequency and inter-system neighboring cells in each blank grid. When the cumulative number of inter-frequency and inter-system neighboring cell measurement data in a grid reaches 3, the grid is available; when the perceived score of the serving cell is lower than the service demand target value, predict from the three dimensions of spectral efficiency, suppression distance, and 4 / 5G residence of the inter-frequency and inter-system neighboring cells, and calculate the cell with the best perception and the best 4 / 5G residence in the grid.

[0136] Second, fingerprint grid application. After the fingerprint grid is established, after the terminal reports the RSRP and SINR measurement information of the serving cell and neighboring cells, use the RSRP and SINR measurement information as the grid index in the grid library to match the logical grid information. If the current serving cell is not the cell with the best 4 / 5G residence and the best perception in the grid, directly perform inter-frequency / inter-system neighboring cell switching to the optimal cell to achieve accurate and fast carrier selection / switching and carrier coordination; if the current serving cell is the optimal cell of the grid, the terminal occupies this cell unchanged.

[0137] Compared with the existing traditional solutions, the advantages of the intelligent shunt ratio improvement scheme based on the wireless fingerprint grid algorithm of the present invention are mainly reflected in two aspects:

[0138] (1) Combine perception to improve the 5G offloading ratio in multiple dimensions. Through machine self-learning algorithms, collect the RSRP and SINR indicators of the user's current serving cell and neighboring cells of different frequencies / different systems, establish a logical fingerprint grid, predict from three dimensions of spectral efficiency, suppression distance, and 4 / 5G residence, and calculate the cell with the best perception and the best 4 / 5G residence within the grid. After the terminal matches the grid, it migrates to the optimal cell. Compared with the traditional handover that only judges from coverage, the solution of the present invention comprehensively considers multiple dimensions and better guarantees the user perception.

[0139] (2) Reduce measurement information, make handovers faster, and further improve the rate. Compared with the traditional 4 / 5G interoperability handover event based on coverage, unnecessary measurement data and handover decision processes are reduced.

[0140] Regarding the specific limitations of a network resource handover system based on a wireless fingerprint grid algorithm, reference can be made to the limitations of a network resource handover method based on a wireless fingerprint grid algorithm in the above text, which will not be elaborated here. Each module in the above network resource handover system based on a wireless fingerprint grid algorithm can be implemented in whole or in part by software, hardware, and their combinations. The above modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above respective modules.

[0141] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0142] The above-described embodiments only represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A network resource switching method based on a wireless fingerprint grid algorithm, characterized in that: The method comprises: Step S1: By randomly performing periodic co-frequency measurements on target terminals in the target area, wireless characteristic information of the service cell and the neighboring cells with the same frequency as the service cell collected by multiple target terminals respectively is obtained; the geographical area where the service cell obtained by each target terminal and the N neighboring cells with the same frequency as its own service cell and the strongest wireless characteristics are located is divided into an independent blank grid, thereby completing the division of the blank grid of the target area; the wireless characteristics include reference signal received power RSRP and signal to downlink interference plus noise ratio SINR; Step S2: obtaining wireless characteristic information of the serving cell and neighboring cells of different frequencies and systems collected by each terminal in the geographical area corresponding to each blank grid; Step S3: Score the service perception of the service cell of each terminal in each grid. When the service service perception score of the service cell is lower than the target value, the neighboring area of ​​different frequencies and systems obtained by the terminal is judged to obtain the optimal cell corresponding to each terminal. The judging of the neighboring area of ​​different frequencies and systems obtained by the terminal to obtain the optimal cell corresponding to each terminal includes: Step S31: determining the hetero-frequency and hetero-system neighboring area of ​​each terminal, evaluating the hetero-frequency and hetero-system neighboring area of ​​each terminal, calculating the spectrum efficiency and suppression distance of the hetero-frequency and hetero-system neighboring area, and thus calculating the user perception value of each hetero-frequency and hetero-system neighboring area; when evaluating the hetero-frequency and hetero-system neighboring area of ​​each terminal, the wireless characteristic information of the hetero-frequency and hetero-system neighboring area of ​​each terminal should be the wireless characteristic information collected when the current hetero-frequency and hetero-system neighboring area serves as a service cell for other terminals in the same grid; Step S32: Compare the wireless characteristic information of the hetero-frequency and hetero-system neighboring cells of each terminal with the set threshold to select the neighboring cells that meet the set threshold, and then score the neighboring cells that meet the set threshold according to the network frequency band and wireless characteristic information to obtain the 4G / 5G-based resident value of each hetero-frequency and hetero-system neighboring cell; Step S33: Obtain a final score for each neighboring cell according to the user perception value of each neighboring cell and the 4G / 5G-based residence value; the neighboring cell with the highest score is the optimal cell; Step S4: Switch the network of each terminal to the network of the optimal cell; The calculation of the suppression distance of the hetero-frequency and hetero-system neighboring areas comprises: Obtain the traffic growth suppression point value of the neighboring area and the traffic value of the current state, and calculate the suppression distance of the neighboring area according to the traffic growth suppression point value of the neighboring area and the traffic value of the current state using the following formula: D s =P s -P a Where: P s is the flow growth suppression point value, P a is the flow value of the current state; The calculation of the user perception value of each hetero-frequency and hetero-system neighboring cell includes: The normalized spectrum efficiency is calculated based on the neighboring cell spectrum efficiency of different frequencies and systems using the following formula: SE scale =SE / SE max ; Among them, SE max The maximum spectrum efficiency of the neighboring cells with different frequencies and systems; The user perception value is calculated using the following formula based on the achievable rate, spectrum efficiency, and suppression distance of the neighboring cells of different frequencies and systems: P={Max(R A ,0)}β×SE scale θD S F Among them, β and θ are weighting factors, R A The reachable rate and SE calculated for the neighboring cell of the current terminal as the service cell of other terminals in the same grid max is the maximum spectral efficiency, D S is the suppression distance, Φ is D S variables.

2. The method according to claim 1, characterized in that The service perception of the serving cell of the terminal is scored in the following manner: Obtain playback data when the terminal plays the video, the playback data including video source quality, buffering time, number of freezes, and percentage of freeze recovery time; The first service perception score of the serving cell is calculated by the following formula: VMOS=f(vQuality,vLoading,vStallingfreq,vStallingrec,W n ) Among them, VMOS is the video user perception score, vQuality is the video source quality, vLoading is the buffering time, vStallingfreq is the number of freezes, vStallingrec is the percentage of freeze recovery time, and W n is the weight ratio of each factor; When the first service perception score of the serving cell reaches the first set value, the neighboring cells of the terminal with different frequencies and systems are evaluated.

3. The method according to claim 2, characterized in that The service perception of the serving cell of the terminal is also scored in the following manner: Obtain the terminal resource load and channel quality, and obtain the overall load of the terminal's serving cell and calculate the upper limit of the achievable rate that the terminal can currently obtain using the following formula: R A =α×BW×(1-Load1+Load2)×log2(1+β×SINR(lower)); Among them, α and β are efficiency factors, which can be obtained by learning and fitting the historical data between the rate and various parameters; BW is the system bandwidth; Load1 is the overall resource load of the serving cell, Load2 is the current terminal resource load, and R A is the achievable rate of the serving cell; SINR (lower) is the downlink signal to interference plus noise ratio; Calculate the user's achievable rate in the service cell and the service demand rate R T The rate margin between is calculated by the following formula: R GAP =MAX{(R A -R T ) / R A ,0}; The user's achievable rate in the service cell is equal to the service demand rate R T The rate margin between is the second service perception score; When the second service perception score of the serving cell reaches the second set value, the hetero-frequency and hetero-system neighboring cells of the terminal are evaluated.

4. The method according to claim 1, characterized in that: The calculating of the inter-frequency and inter-system adjacent area spectrum efficiency comprises: According to the path loss calculated based on the downlink receiving RSRP and the transmit power of the neighboring cells of different frequencies and systems, and combined with the interference information obtained from the interface between base stations, the uplink SINR of the neighboring cells of different frequencies and systems can be calculated, and then the spectrum efficiency is calculated by the uplink SINR of the neighboring cells of different frequencies and systems; the spectrum efficiency is calculated by the uplink SINR of the neighboring cells of different frequencies and systems by the following formula: SE=α×log2(1+β×SINR(upper)) Among them, α and β are efficiency factors, and SINR(up) is the uplink signal to interference plus noise ratio.

5. The method according to claim 1, characterized in that The set threshold is SINR≥3 and RSRP≥-110dBm.

6. The method according to claim 1, characterized in that Scoring neighboring cells that meet the set threshold according to the frequency range and technology generation of the network to obtain a 4G / 5G-based resident value for each hetero-frequency and hetero-system neighboring cell includes: Obtain the priority score of the neighboring cell according to the frequency band of the neighboring cell; The priority score of the neighboring cell and the wireless characteristic information are calculated using the following formula: Among them, Q is the residence value of the cell, Y is the frequency band priority score of the neighboring cell, and ε is the residence level weight.

7. The method according to claim 1, characterized in that The final score of each neighboring area is calculated based on the user perception value of each neighboring area and the 4G / 5G-based residence value by the following formula: C=γ×P+δ×Q =γ{max(R A ,0)}β×γSE scale θD S Φ+δ×Q Among them, γ and δ are weight factors, β and θ are weighting factors, P is the user perception value, and Q is the residence value based on 4G / 5G.

8. A network resource switching system based on wireless fingerprint grid algorithm, characterized in that: The system comprises: Blank grid division module: used to obtain wireless characteristic information of the service cell and the neighboring cells with the same frequency as the service cell respectively collected by multiple target terminals by randomly performing periodic same-frequency measurements on the target terminals in the target area; the geographical area where the service cell obtained by each target terminal and the N neighboring cells with the same frequency as its own service cell and the strongest wireless characteristics are located is divided into an independent blank grid, so as to complete the division of the blank grid of the target area; the wireless characteristics include reference signal received power RSRP and signal to downlink interference plus noise ratio SINR; Blank grid data filling module: used to obtain the wireless characteristic information of the service cell and the neighboring cells of different frequencies and systems collected by each terminal in the geographical area corresponding to each blank grid; Service business perception scoring module: used to score the service perception of the service cell of each terminal in each grid. When the service business perception score of the service cell is lower than the target value, the neighboring cells of different frequencies and systems obtained by the terminal are judged to obtain the optimal cell corresponding to each terminal; The optimal cell judgment module is used for the hetero-frequency and hetero-system neighboring areas of each terminal, evaluates the hetero-frequency and hetero-system neighboring areas of each terminal, calculates the spectrum efficiency and suppression distance of the hetero-frequency and hetero-system neighboring areas, and thus calculates the user perception value of each hetero-frequency and hetero-system neighboring area; when evaluating the hetero-frequency and hetero-system neighboring areas of each terminal, the wireless feature information of the hetero-frequency and hetero-system neighboring areas of each terminal should be the wireless feature information collected when the current hetero-frequency and hetero-system neighboring areas serve as the service cells of other terminals, compares the wireless feature information of the hetero-frequency and hetero-system neighboring areas of each terminal with the set threshold to select the neighboring areas that meet the set threshold, and then scores the neighboring areas that meet the set threshold according to the network frequency band and the wireless feature information to obtain the 4G / 5G-based resident value of each hetero-frequency and hetero-system neighboring area; obtains the final score of each neighboring area according to the user perception value of each neighboring area and the 4G / 5G-based resident value; the neighboring area with the highest score is the optimal cell; the calculation of the suppression distance of the hetero-frequency and hetero-system neighboring areas includes: Obtain the traffic growth suppression point value of the neighboring area and the traffic value of the current state, and calculate the suppression distance of the neighboring area according to the traffic growth suppression point value of the neighboring area and the traffic value of the current state using the following formula: D s =P s -P a Where: P s is the flow growth suppression point value, P a is the flow value of the current state; The calculation of the user perception value of each hetero-frequency and hetero-system neighboring cell includes: The normalized spectrum efficiency is calculated based on the neighboring cell spectrum efficiency of different frequencies and systems using the following formula: SE scale =SE / SE max ; Among them, SE max The maximum spectrum efficiency of the neighboring cells with different frequencies and systems; The user perception value is calculated using the following formula based on the achievable rate, spectrum efficiency, and suppression distance of the neighboring cells of different frequencies and systems: P={Max(R A ,0)}β×SE scale θD S F Among them, β and θ are weighting factors, R A The reachable rate and SE calculated for the neighboring cell of the current terminal as the service cell of other terminals in the same grid max is the maximum spectral efficiency, D S is the suppression distance, Φ is D S Variables; Network switching module: used to switch the network of each terminal to the network of the optimal cell.

Images