At operation S420, an area interference pattern for each of the plurality of unit areas may be determined.

Specifically, in the disclosure, the area interference pattern represents a coding combination of an intensity of the intra-frequency interference of adjacent cells suffered by one area at respective time unit within a predetermined time, and in the disclosure, the time unit may be, for example, a time slot, and at this time, the above predetermined time may be, for example, a wireless frame comprising four time slots, but the disclosure is not limited thereto. In the following description, for ease of description, the time unit is taken as a time slot for example. In addition, the intensity of the intra-frequency interference of the adjacent cells suffered by one area at respective time unit within the predetermined time may be, for example, a three-level intensity division such as high, medium and low, and furthermore, high, medium and low referred to herein in a certain sense indicate that the intra-frequency interference intensity is relatively high, relatively medium and relatively low. However, the disclosure is not limited to this, and the intensity of the intra-frequency interference of the adjacent cells suffered by one area at respective time unit within the predetermined time may also be, for example, a four-level intensity division, a five-level intensity division, and the like, according to the intensity from high to low. In the following description, for ease of description, the three-level intensity division such as high, medium and low is taken as an example to describe the area interference pattern, and at this time, the area interference pattern may be represented by a coding combination shown in Table 1 below. Furthermore, in the disclosure, in order to limit adjustment of priorities of the UE only suitable for those time units in a current pattern, scheduling priorities between different area scheduling patterns of the UE are not changed, and +1 and −1 need to appear in pairs in each area interference pattern, as shown in FIG. 3B. Furthermore, in each area interference pattern, the interference suffered by the current cell at different time units is a sum of interference from at least one other neighbor cell, wherein each neighbor cell may generate, on the current cell, interference in different intensity, e.g., as shown in FIG. 3B, for pattern 16, interferences suffered by a current cell 0 at time units TS0, TS1, TS2 and TS3 is [−100 dBm, −85 dBm, −110 dBm, −90 dBm], wherein each of interference is a sum of interference from other neighbor cells 1, 2 . . . G. In FIG. 3B, the interference on the current cell 0 generated by each neighbor cell is indicated by a thick arrow representing strong interference or a thin arrow representing weak interference. In addition, the above mentioned one area may be the unit area herein or the interference measurement area, and the disclosure may employ an AI method or a rule-based non-AI method to determine the area interference pattern for the unit area and the interference measurement area. How to determine the area interference pattern will be described in detail with reference to FIGS. 8 and 12.

TABLE 1

Intensity of intra-frequency

interference of adjacent cells suffered

by one area at one time unit
Coding

High
+1

Medium
0

Low
−1

At operation S430, the plurality of unit areas may be grouped, according to area interference pattern sequence correlations between unit areas of the plurality of unit areas within a period of time, to determine the plurality of interference measurement areas, wherein the area interference pattern sequence correlation represents a correlation between area interference pattern sequences of two unit areas within a period of time (e.g., 1 minute). This is described in detail below with reference to FIG. 6.

FIG. 6 is a flowchart illustrating a process of determining a plurality of interference measurement areas by grouping a plurality of unit areas according to an embodiment of the disclosure.

Referring to FIG. 6, at operation S610, a first unit area may be determined as a pending interference measurement area. For example, a unit area 1 in panel (a) of FIG. 5 is determined as the pending interference measurement area, and since it has not yet been determined whether the unit area 1 may be merged with a next unit area 2 to form an interference measurement area, the unit area 1 may currently be determined as the pending interference measurement area.

At operation S620, with respect to the pending interference measurement area, an operation of determining the area interference pattern sequence correlation may be performed, wherein, the operation of determining the area interference pattern sequence correlation is used to determine the area interference pattern sequence correlation between each unit area in the pending interference measurement area and a next unit area adjacent to the pending interference measurement area in a predetermined direction.

Specifically, the above predetermined direction may be clockwise or counterclockwise, and in the following description it is assumed that the predetermined direction is clockwise. At the first execution of operation S620, only the first unit area (e.g., the unit area 1 in panel (a) of FIG. 5) is included in the pending interference measurement area, and the next unit area adjacent to the pending interference measurement area in the clockwise direction is the second unit area (e.g., the unit area 2 in panel (a) of FIG. 5), thus, operation S620 only needs to determine, according to the following Equation 1, the area interference pattern sequence correlation between the first unit area and the second unit area.

ρ

m
,
k

=

❘
"[LeftBracketingBar]"

A
k

⁢

A
m
H

❘
"[RightBracketingBar]"



A
k



⁢



A
m



Equation
⁢

1

- wherein, ρ_m,krepresents an area interference pattern sequence correlation between a unit area m and a unit area k within a period of time, Ak and Am represent area interference pattern sequences between the unit area k and the unit area m within the period of time, and

A
m
H

represents a transposed matrix of Am, ∥A_k∥ and ∥A_m∥ represent norms of Ak and Am, respectively.

❘
"[LeftBracketingBar]"

A
k

⁢

A
m
H

❘
"[RightBracketingBar]"

represents a solution of a determinant of the matrix

A
k

⁢

A
m
H

.

For example, there are 12 time units in the period of time, and every 4 time units constitute one radio frame, and each radio frame has a corresponding area interference pattern. The area interference pattern sequence Ak for the unit area k may be, for example, {(−1,1,0,0), (1,−1,0,0), (−1,1,1,−1), (−1,−1,1,1)}, and the area interference pattern sequence Am for the unit area m may be, for example, {(1,−1,−1,1), (−1,1,0,0), (−1,−1,1,1), (0,0,−1,1)}, and to this end, the area interference pattern sequence correlation between the unit area m and the unit area k within the period of time may be determined according to Equation 1 above.

At operation S630, whether a minimum value L among the area interference pattern sequence correlations between the next unit area adjacent to the pending interference measurement area in the predetermined direction and each unit area in the pending interference measurement area is greater than the first predetermined threshold is determined.

If this minimum value L is greater than the first predetermined threshold, at operation S640, the next unit area adjacent to the pending interference measurement area in the predetermined direction is grouped to the pending interference measurement area, and returning to operation S620 to perform the operation of determining the area interference pattern sequence correlation in operation S620 with respect to the pending interference measurement area at this time.

If this minimum value L is not greater than the first predetermined threshold, at operation S650, the pending interference measurement area may be determined as a final interference measurement area and the next unit area adjacent to the pending interference measurement area in the predetermined direction is determined as a new pending interference measurement area, and then returning to operation S620 to perform the operation of determining the area interference pattern sequence correlation in operation S620 with respect to the pending interference measurement area at this time.

Specifically, for example, as shown in panel (a) of FIG. 5, when the pending interference measurement area includes only the unit area 1, the next unit area adjacent to the pending interference measurement area in the clockwise direction is the unit area 2 among the plurality of unit areas, and if an area interference pattern sequence correlation ρ_2,1between the unit area 2 and the unit area 1 is not greater than the first predetermined threshold at this time, proceeding to operation S650, the pending interference measurement area at this time may be determined as one final interference measurement area, for example, as a first interference measurement area, and the next unit area 3 adjacent to the pending interference measurement area in the clockwise direction is determined as the new pending interference measurement area, and returning to operation S620 to perform the operation of determining the area interference pattern sequence correlation in operation S620 with respect to the pending interference measurement area at this time, to determine an area interference pattern sequence correlation between the next unit area 4 adjacent to the pending interference measurement area at this time in time in the predetermined direction and each unit area (i.e., the unit area 3) in the pending interference measurement area, that is, to determine an area interference pattern sequence correlation ρ_4,3between the unit area 4 and the unit area 3, and then operation S630 is performed for the determining.

If the area interference pattern sequence correlation ρ_2,1between the unit area 2 and the unit area 1 is greater than the first predetermined threshold, then at operation S640, the unit area 2 is grouped to the pending interference measurement area, that is, is merged into the pending interference measurement area, wherein the pending interference measurement area at this time includes the unit area 1 and the unit area 2. Thereafter, returning to operation S620, the operation of determining the area interference pattern sequence correlation in operation S620 is performed with respect to the pending interference measurement area at this time, to determine an area interference pattern sequence correlation between the next unit area 3 adjacent to the pending interference measurement area at this time in the clockwise direction and each unit area in the pending interference measurement area, that is, to determine an area interference pattern sequence correlations ρ_3,2and ρ_3,1between the unit area 3 and each of the unit area 1 and the unit area 2, and then operation S630 is performed, to determine whether min(ρ_3,1, ρ_3,2) is greater than the first predetermined threshold, and if min(ρ_3,1, ρ_3,2) is greater than the first predetermined threshold, proceeding to operation S640 to group the unit area 3 into the pending interference measurement area, that is, the unit area 3 is merged with the unit area 1 and the unit area 2, and then returning to operation S620, the operation for the unit area 4 is continued.

By the process shown in FIG. 6, the plurality of interference measurement areas are determined until the last unit area of the plurality of unit areas is grouped, for example, until the grouping of the unit area W in panel (a) of FIG. 5 is completed. As described above with reference to panel (a) of FIG. 5, in the same way, the plurality of interference measurement areas may be determined in the above manner for the W unit areas shown in panel (b) of FIG. 5.

FIG. 7 is a diagram illustrating a relationship between an interference area division operation and a first period according to an embodiment of the disclosure.

Referring to FIG. 7, the above operation of dividing the current cell into the plurality of interference measurement areas is performed at the beginning of each first period P1. For example, the first period P1 is 1 hour or 1 day. In addition, two consecutive first periods P1 are spaced with random time, thus avoiding simultaneous changes of area interference patterns among individual cells.

The above process may divide the current cell into the plurality of interference measurement areas, and then, the interference information about interference suffered by each interference measurement area may be predicted. Specifically, the predicting the interference information about interference suffered by each interference measurement area may include: predicting an area interference pattern of a current interference measurement area at a current second period according to historical interference information of the current interference measurement area, that is, in the disclosure, the historical interference information of the current interference measurement area may be an interference power sequence collected by the current interference measurement area at the previous second period. In addition, the interference information about interference suffered by the current interference measurement area may include the area interference pattern of the current interference measurement area at the current second period. As described above with reference to operation S420, the area interference pattern may represent the coding combination of the intensity of the intra-frequency interference of adjacent cells suffered by one area at respective time unit within the predetermined time. In the disclosure, an AI method or a rule-based non-AI method may be periodically used to determine the area interference pattern for one area (i.e., a unit area and an interference measurement area) in accordance with the second period. Since the interference measurement area is generated by merging several unit areas and in some cases the interference measurement area is a single unit area, and the process of determining the area interference pattern for the interference measurement area is the same as that of determining the area interference pattern for the unit area, therefore, in the disclosure, the process of determining the area interference pattern is described by taking the interference measurement area as an example. Further, in the disclosure, the second period is typically much smaller than the first period, for example, the first period may be, for example, 1 hour or 1 day, and the second period may be, for example, 200 ms.

The process of predicting the area interference pattern of the current interference measurement area at the current second period according to the historical interference information of the current interference measurement area, by the rule-based non-AI method and the AI method, is described successively below.

FIG. 8 is a flowchart illustrating a process of predicting an area interference pattern of one interference measurement area by a rule-based non-AI method according to an embodiment of the disclosure.

Referring to FIG. 8, at operation S810, one interference power matrix of the plurality of interference measurement areas is determined, according to an interference power of each time unit in the previous second period collected at each interference measurement area. Specifically, Equation 2 below illustrates a specific form of this interference power matrix.

InterPower

T
*
M

=

[

P

0
,
0

⋯

P

0
,

M
-
1

⋮

⋱

⋮

P

T
-
1

,
0

⋯

P

T
-
1

,

M
-
1

]

T
*
M

Equation
⁢

2

Wherein, T represents the number of time units within a predetermined duration.

FIG. 9 is a schematic diagram illustrating a structure of a second period according to an embodiment of the disclosure.

Referring to FIG. 9, there are a plurality of predetermined durations in each second period, and each predetermined duration consists of a plurality of time units, for example, each second period contains a plurality of radio frames, and each radio frame contains 4 time units (e.g., time slots TS0, TS1, TS2, and TS3), and the predetermined duration may be one radio frame or a plurality of radio frames, i.e., J radio frames, wherein J is a positive integer greater than or equal to 1. In this case, T=4×J. M represents the total number of interference measurement areas in the current cell. Thus, Pt,m in the interference power matrix InterPower represents a filtered interference power obtained by filtering an average interference power of all time units TSts within the previous second period at the mth interference measurement area. This will be described in detail below with reference to FIG. 10.

FIG. 10 is a flowchart illustrating a process of determining matrix elements in an interference power matrix (i.e., interference power of each interference measurement area at each time unit) according to an embodiment of the disclosure. Before proceeding with the process of FIG. 10, it is necessary to collect interference power of each time unit (e.g., time slot) in each interference measurement area during the previous second period, and then the operations shown in FIG. 10 is performed with respect to each of the plurality of interference measurement areas.

Referring to FIG. 10, at operation S1010, by averaging interference powers of time units TSts with a same serial number in the previous second period with respect to the current interference measurement area, an average interference power of the current interference measurement at these time units TSts of the previous second period area may be obtained.

Specifically, as shown in FIG. 9, each radio frame contains four time units TS0, TS1, TS2 and TS3. Assuming that each second period contains 10 radio frames, then for the current interference measurement area, an average interference power of 10 TS0s, an average interference power of 10 TS1s, an average interference power of 10 TS2s and an average interference power of 10 TS3s in the previous second period are calculated, respectively.

Then, at operation S1020, the average interference power of the current interference measurement at these time units TSts of the previous second period area is filtered, based on a historical filtered interference power of the current interference measurement area at the time units TSts, to obtain a filtered interference power of the current interference measurement area at the time units TSts of the previous second period. Specifically, this average interference power may be filtered according to the following Equation 3.

P_filter=a×P_filter+(1−a)×P_ave Equation 3

- wherein, P_averepresents an average interference power of one interference measurement area at time units with a same serial number in a second period, and P_filterindicates the filtered interference power of this interference measurement area at these time units of this second period, P_filterrepresents the historical filtered interference power of this measurement interference area at these time units, a represents a filtering coefficient and is settable.

In addition, the historical filtered interference power of the current interference measurement area at the time units TSts is updated as the filtered interference power of the current interference measurement area at the time units TSts of the previous second period, that is, whenever a filtered interference power of one interference measurement area at one time unit is obtained, the obtained filtered interference power of this interference measurement area at this time unit may be used for updating the historical filtered interference power of this interference measurement area at this time unit.

At operation S1030, the filtered interference power of the current interference measurement area at the time units TSts of the previous second period may be set as the interference power of the current interference measurement area at the time units TSts in the interference power matrix. The filtered interference power for each of the time units TS0, TS1, TS2 and TS3 in the previous second period may be determined for the current interference measurement area by each of the steps described above, thus a column of elements in the interference power matrix expressed by the above Equation 2 is determined.

Accordingly, the interference power matrix for the plurality of interference measurement areas may be obtained, by performing the operation shown in FIG. 10 with respect to each of the plurality of interference measurement areas.

Referring back to FIG. 8, at operation S820, one area interference pattern matrix CMatrix_N*Tis selected from an area interference pattern table including at least one area interference pattern matrix according to serial numbers of time units within the current second period.

Specifically, the area interference pattern table may include at least one area interference pattern matrix, each area interference pattern matrix represents all possible area interference patterns within a predetermined duration (e.g., one radio frame or a plurality of radio frames, i.e., J radio frames, wherein J is a positive integer greater than or equal to 1), and this interference pattern table is preset.

FIG. 11 is a diagram illustrating an example of area interference pattern table according to an embodiment of the disclosure.

Referring to FIG. 11, As described above with reference to FIG. 3B, in each area interference pattern, +1 and −1 need to appear in pairs, thus, in two area interference pattern matrices CMatrix1 and CMatrix2 contained in the area interference pattern table shown in FIG. 11, the area interference pattern matrix CMatrix1 contains a total of 6 area interference patterns, and the area interference pattern matrix CMatrix2 contains a total of 18 area interference patterns. Assuming that serial numbers of the time units in the current second period are TS0, TS1, TS2 and TS3, that is, each predetermined duration within the second period contains only 4 time units TS0, TS1, TS2 and TS3, and at this time, according to the serial numbers of the available time units (i.e., TS0, TS1, TS2 and TS3) of an uplink/downlink within the current second period, the area interference pattern matrix CMatrix2 is selected from the area interference pattern table shown in FIG. 11.

At operation S830, the area interference pattern of the current interference measurement area at the current second period may be predicted according to the interference power matrix and the selected area interference pattern matrix.

Specifically, firstly, as shown in Equation 4 below, a result matrix is determined based on a multiplication of an interference power matrix InterPower_T*Mand the area interference pattern matrix CMatrix_N*T.

CMatrix

N
*
T

*

InterPower

T
*
M

=

[

V

0
,
0

⋯

V

0
,

M
-
1

⋮

⋱

⋮

V

N
-
1

,
0

⋯

V

N
-
1

,

M
-
1

]

Equation
⁢

4

- wherein N represents the total number of area interference patterns in the selected area interference pattern matrix CMatrix_N*T, and T represents the serial numbers of the available time units of the uplink/downlink within the second period, that is, the number of time units within the predetermined duration, wherein, as described above, the second period includes the plurality of predetermined durations, each predetermined duration may be one radio frame or a plurality of radio frames, that is, J wireless frames, wherein J is a positive integer greater than or equal to 1. That is, each predetermined duration consists of a plurality of time units with different serial numbers of time units. As shown in FIG. 9, each predetermined duration (e.g., one radio frame) contains 4 time units TS0, TS1, TS2, and TS3, thus, for FIG. 9, T is 4. M represents the total number of interference measurement areas in the current cell.

Then, a maximum value among a plurality of matrix elements corresponding to the current interference measurement area may be determined from among the result matrix. Specifically, the maximum value is selected from a column of matrix elements corresponding to the current interference measurement area in the result matrix.

Finally, an area interference pattern corresponding to this maximum value in the area interference pattern matrix CMatrix_N*Tmay be determined as the area interference pattern of the current interference measurement area at the current second period.

For example, assuming that the interference power of 4 time units in the current interference measurement area is [−100 dBm, −80 dBm, −110 dB, −90 dBm], one column of matrix elements corresponding to the current interference measurement area in the result matrix is one column of data indicated by 1110 in FIG. 11, wherein 40 is the maximum value in this column of matrix elements, and accordingly, an area interference pattern 16 corresponding to 40 is determined as the area interference pattern of the current interference measurement area at the current second period.

By performing the operations of FIG. 8 above with respect to each of the plurality of interference measurement areas, the area interference pattern for each of the interference measurement areas may be determined.

The process described above with reference to FIG. 8 is the process of determining the area interference pattern for each interference measurement area by the rule-based non-AI method, that is, an area interference pattern is predefined firstly and then a predefined area interference pattern that is most relevant to an actual historical interference is found as the predicted area interference pattern. In addition, the disclosure may also determine the area interference pattern of each interference measurement area through an A1 method (i.e., an AI model), which is described in detail below.

FIG. 12 is a schematic diagram illustrating a process of determining the area interference pattern of each interference measurement area through an AI model according to an embodiment of the disclosure, wherein, in the disclosure, the AI model may be, for example, a convolutional neural network (CNN), a recurrent neural network (RNN), etc.

In summary, the following operation may be performed with respect to each of the plurality of interference measurement areas: predicting the area interference pattern of the current interference measurement area at the current second period using the historical interference information of the current interference measurement area through the AI model, wherein the historical interference information of the current interference measurement area may be an interference power sequence collected by the current interference measurement area at the previous second period.

Referring to FIG. 12, H interference power sequences (IS1, IS2, . . . ISH) collected at the previous second period P1 may be input to a trained AI model, to obtain the area interference pattern of the current interference measurement area at the current second period, wherein Inter in the table represents an interference power of a corresponding time unit, e.g. Inter10 represents an interference power of a time unit TS0 in an interference power sequence IS1. An output of the AI model is the area interference pattern (−1, 1, −1, 1) of the current interference measurement area at the current second period.

By a similar operation, the AI model may be used to obtain the area interference pattern of each interference measurement area at the current second period.

In addition, the predicting the area interference pattern of each interference measurement area may include: if an area interference pattern of any of the plurality of interference measurement areas changes, a next operation to determine the area interference pattern of the any interference measurement area is delayed, for example, the next operation to determine the area interference pattern of the any interference measurement area is extended to 2×the second period. This is to maintain stable interference and scheduling patterns, to avoid inaccurate prediction of area interference patterns due to simultaneous modification of area interference patterns and/or area scheduling patterns by a plurality of cells.

Referring back to FIG. 2, at operation S220, scheduling priorities of respective UE at different time units may be determined according to the predicted interference information and locations of the UEs in the current cell. This is described in detail below with reference to FIG. 13.

FIG. 13 is a flowchart illustrating a process of determining scheduling priorities of respective UE at different time units according to a predicted interference information and locations of the UEs in a current cell, according to an embodiment of the disclosure.

Referring to FIG. 13, at operation S1310, an area scheduling pattern of each interference measurement area is determined based on the predicted area interference pattern of each interference measurement area, wherein the area scheduling pattern represents a set of weight values for adjusting scheduling priorities of UEs, at different locations within the current interference measurement area, at different time units. A UE at an edge location is far away from a base station, the signal is often poor and cannot use a higher order MCS, thus it is necessary to avoid an impact of high interference as much as possible. A UE at a central location is closer to the base station, the signal is stronger, and its anti-interference capability is better than the UE at the edge location.

FIG. 14A is a diagram illustrating a relationship between UE locations and area scheduling patterns according to an embodiment of the disclosure.

FIG. 14B is a schematic diagram illustrating a UE at a scheduling edge location according to an embodiment of the disclosure.

Referring to FIGS. 14A and 14B, for one interference measurement area, an interference power is lower, a weight value of a scheduling order (i.e., scheduling priority) of the UE at the edge location is higher, and the interference power is higher, a weight value of a scheduling order of the UE at the central location is higher, and thus referring to FIG. 14B, when suffering high interference from a neighbor cell, for the UE at the edge location in cell 0, the UE at the edge location may be scheduled preferentially when the interference is low. This operation S1320 is described in detail below with reference to FIG. 15.

FIG. 15 is a flowchart illustrating a process of determining an area scheduling pattern of each interference measurement area based on a predicted area interference pattern of each interference measurement area according to an embodiment of the disclosure.

Referring to FIG. 15, at operation S1510, an interference mean deviation weight of time units with the same serial number within the current second period is determined, i.e., the interference mean deviation weight of time units with each serial number within the current second period is determined. As shown in FIG. 9, the time units with the same serial number within the current second period represent, for example, TS0, TS1, TS2, and TS3 within the current second period.

Specifically, firstly, an average interference power of the time units with the same serial number (i.e., time units with each serial number) within the current second period and an average value of interference powers of all time units within the current second period may be determined.

FIG. 16 is a schematic diagram illustrating a process of calculating an average interference power per time unit according to an embodiment of the disclosure.

Referring to FIG. 16, an average interference power Pave0 of all time units TS0s, an average interference power Pave1 of all time units TS1s, an average interference power Pave2 of all time units TS2s and an average interference power Pave3 of all time units TS3s within the current second period may be calculated, respectively, and then an average value Pave of interference powers of all time units (i.e., all TS0s, all TS1s, all TS2s and all TS3s) within the current second period may be calculated.

Then, an absolute difference between the average interference power of the time units with the same serial number (i.e., time units with each serial number) and the average value of the interference powers of all time units within the current second period may be calculated.

FIG. 17 illustrates an example of the absolute difference (i.e., mean deviation) between the average interference power of time units with each serial number and the average value of the interference powers of all time units within the current second period according to an embodiment of the disclosure.

Referring to FIG. 17, the interference mean deviation weight of the time units with the same serial number (i.e., time units with each serial number) is determined according to the absolute difference of the time units with the same serial number (i.e., time units with each serial number). Specifically, firstly, an average value of the obtained absolute differences of time units with each serial number may be calculated using the absolute differences, and then a ratio of the absolute difference of time units with each serial number to the average value of the absolute differences is calculated, to obtain the interference mean deviation weight of time units with each serial number (e.g., each of time unit TS0, time unit TS1, time unit TS2, and time unit TS3). For example, assuming that average interference powers [Pave0, Pave1, Pave2, Pave3] of time units with each serial number TS0, TS1, TS2 and TS3 within the second period are [−100 dBm, −85 dBm, −110 dB, −90 dBm], an average value of interference powers of these time units with four serial numbers is Pave=(Pave0+Pave1+Pave2+Pave3)/4=−96.25 dBm, absolute differences between the average interference powers of time units with each serial number and the average value of interference powers of the time units with four serial numbers are [abs(Pave0−Pave), abs(Pave1−Pave), abs(Pave2−Pave), abs(Pave3−Pave)]=[3.75, 11.25, 13.75, 6.25], respectively, and the average value of the absolute differences of these time units with four serial numbers is (3.75+11.25+13.75+6.25)/4=8.75, therefore, the interference mean deviation weights of these time units with four serial numbers are [0.4, 1.3, 1.6, 0.7], respectively.

In the disclosure, the interference mean deviation weight actually characterizes a deviation between the interference powers of each time unit and their average value. The deviations of the interference powers of different time units with respect to the average value are larger, an impact of adjusting the scheduling priorities of the UE at the edge location and the UE at the central location in different time units is greater; if the interference powers on each time unit are equal, there will be no impact when adjusting the scheduling priorities of the UE at the edge location and the UE at the central location in different time units.

At operation S1520, a standard area scheduling pattern of each interference measurement area is determined according to the interference mean deviation weight of the time units with the same serial number (i.e., time units with each serial number) and an area interference pattern of each interference measurement area.

Specifically, the standard area scheduling pattern of each interference measurement area is obtained by multiplying the interference mean deviation weight of time units with each serial number and the area interference pattern of each interference measurement area. For example, as mentioned above, the interference mean deviation weights of the time units with four serial numbers are [0.4, 1.3, 1.6, 0.7], and assuming that the area interference pattern of the current interference measurement area is [−1, 1, −1, 1], the standard area scheduling pattern of the current interference measurement area is obtained by multiplying the corresponding elements among them.

standard area scheduling pattern=interference mean deviation weight*area interference pattern==[0.4,1.3,1.6,0.7]*[−1,1,−1,1]=[−0.4,1.3,−1.6,0.7] Equation 5

The standard area scheduling pattern of each interference measurement area may be determined by the above process.

At operation S1530, area scheduling patterns of UEs at different locations in each interference area may be determined according to the standard area scheduling pattern of each interference measurement area. Wherein, the area scheduling patterns of UEs at different locations may include at least one of: an area scheduling pattern of a UE at the central location, an area scheduling pattern of a UE at the middle location and an area scheduling pattern of a UE at the edge location.

Specifically, in the disclosure, area division may be performed on the interference measurement area, for example, the interference measurement area is divided according to a distance from the base station, for example, divided into a central area closest to the base station, an edge area farthest from the base station, and a middle area between the central area and the edge area. In this case, a UE located at the central area may be referred to as a UE at the central location, a UE located at the middle area may be referred to as a UE at the middle location, and a UE located at the edge area may be referred to as a UE at the edge location. However, the disclosure is not limited to this, for example, the interference measurement area is divided into more areas or fewer areas depending on a distance from the base station, for example, the interference measurement area is divided into only the central area and the edge area without the middle area depending on the distance from the base station. In the following description, the division of the interference measurement area into the central area, the edge area, and the middle area is illustrated as an example.

With respect to the UE at the edge location in the current interference measurement area, the area scheduling pattern of the UE at the edge location is obtained by calculating a negative value of the standard area scheduling pattern of the current interference measurement area. For example, if the standard area scheduling pattern of the current interference measurement area is [−0.4, 1.3, −1.6, 0.7], the scheduling pattern of the UE at the edge location area is [0.4, −1.3, 1.6, −0.7].

Similarly, with respect to the UE at the central location in the current interference measurement area, the area scheduling pattern of the UE at the central location may be obtained by calculating a positive value of the standard area scheduling pattern of the current interference measurement area, that is, the standard area scheduling pattern of the current interference measurement area is directly used as the area scheduling pattern of the UE at the central location. For example, if the standard area scheduling pattern of the current interference measurement area is [−0.4, 1.3, −1.6, 0.7], the area scheduling pattern of the UE at the central location is [−0.4, 1.3, −1.6, 0.7].

With respect to the UE at the middle location in the current interference measurement area, 0 is used as the area scheduling pattern of the UE at the middle location of the current interference measurement area, that is, the scheduling of the UE at the middle location is not actively adjusted.

So far, the area scheduling pattern of each interference measurement area, e.g., the area scheduling pattern of the UE at the central location, the area scheduling pattern of the UE at the middle location and the area scheduling pattern of the UE at the edge location, may be determined.

Referring back to FIG. 13, at operation S1320, scheduling adjustment weight values of each UE are determined based on the location of each UE and the area scheduling pattern of an interference measurement area to which the UE belongs. Specifically, before performing operation S1320, the location of each UE and the interference measurement area to which it belongs need to be determined.

FIG. 18 is a schematic diagram illustrating a process of determining a location of each UE in a current cell and an interference measurement area to which the UE belongs, according to an embodiment of the disclosure.

Referring to FIG. 18, firstly, a path loss, Reference Signal Received Power (RSRP) or Signal to Interference & Noise Ratio (SINR) of each UE in the current cell may be acquired. Then, all UEs in the current cell may be divided into UEs at different locations, for example, UEs at the edge location of the cell, UEs at the middle location of the cell, and UEs at the central location of the cell, based on the acquired path loss, RSRP or SINR of each UE. Since values of the RSRP and path loss may reflect a distance between a UE and the current base station, that is, if the RSRP is smaller and the path loss is larger, the UE is farther away from the current base station, and if the RSRP is larger and the path loss is smaller, the UE is closer to the current base station. Therefore, all UEs in the current cell may be divided into UEs at the edge location of the cell, UEs at the middle location of the cell and UEs at the center location of the cell, accordingly, wherein, the number of the UEs at the edge location of the cell and the number of the UEs at the central location of the cell may be approximately the same. However, the disclosure is not limited to this, and all UEs within the current cell may be divided into UEs at more locations or fewer locations, for example, all UEs within the current cell may be only divided into UEs at the center location of the cell and UEs at the edge location of the cell. Finally, the interference measurement area to which each UE belongs is determined based on a horizontal arrival angle of signal of each UE.

Operation S1320 is described in detail below with reference to FIG. 19.

FIG. 19 is a flowchart illustrating a process of determining scheduling adjustment weight values of each UE based on a location of each UE and an area scheduling pattern of the interference measurement area to which the UE belongs according to an embodiment of the disclosure.

Referring to FIG. 19, at operation S1910, an area scheduling pattern corresponding to a location of the current UE may be selected from among the area scheduling pattern of the interference measurement area to which the current UE belongs, according to the location of the current UE.

Specifically, for example, assuming that the current UE is the UE at the edge location of the cell, since the area scheduling pattern of each interference measurement area has been previously determined according to operation S1310 (e.g., the area scheduling pattern of the UE at the central location, the area scheduling pattern of the UE at the middle location and the area scheduling pattern of the UE at the edge location of each interference measurement area), the area scheduling pattern corresponding to the location of the current UE may be selected from among the area scheduling pattern of the interference measurement area to which the current UE belongs, according to the location of the current UE, that is, selecting the area scheduling pattern of the UE at the edge location, for example, the area scheduling pattern of the UE at the edge location [0.4, −1.3, 1.6, −0.7] mentioned above in operation S1530.

At operation S1920, the scheduling adjustment weight values of the current UE may be determined according to an impact factor Q and the selected area scheduling pattern.

The impact factor Q may be used to determine a magnitude of the impact of the scheme proposed in the disclosure on an actual system. When an interference power of different time units (e.g., different time slots) of a communication network varies widely, an effect of adjusting the scheduling priorities of the UE at the edge location and the UE at the central location in different time units are better, and the impact factor Q may be set larger at this time, in order to obtain a more appropriate impact factor Q to obtain the highest throughput. Therefore, in the disclosure, an average throughput of the current base station may be counted according to a third period P3, and then the impact factor Q is adjusted in positive correlation with this average throughput.

FIG. 20 is a schematic diagram illustrating a process of adjusting an impact factor according to an embodiment of the disclosure.

Referring to FIG. 20, firstly, an initial value of the impact factor Q is set, the impact factor Q is increased, and then average throughput in one third period P3 is counted. If the average throughput in this third period P3 increases, the impact factor Q is increased at the beginning of the next third period P3. If the average throughput within that third period P3 decreases, the impact factor Q is adjusted in reverse until an optimal impact factor Q is obtained. Wherein the third period P3 is typically larger than the first period P1, which may be, for example, one week, two weeks, etc.

In the disclosure, a scheduling adjustment weight value of one UE may be obtained by multiplying the area scheduling pattern of this UE with the current impact factor Q. For example, if the area scheduling pattern selected for this UE is the area scheduling pattern of the UE at the edge location (e.g., [0.4, −1.3, 1.6, −0.7]), the scheduling adjustment weight value of this UE is [0.4, −1.3, 1.6, −0.7]*Q=[0.4Q, −1.3Q, 1.6Q, −0.7Q], and if the area scheduling pattern selected for this UE is the area scheduling pattern of the UE at the central location (e.g., [0, 0, 0, 0]), the scheduling adjustment weight value of this UE is [0, 0, 0, 0]*Q=[0, 0, 0, 0], and if the area scheduling pattern selected for this UE is the area scheduling pattern of the UE at the central location (e.g. [−0.4, 1.3, −1.6, 0.7]), the scheduling adjustment weight of this UE is [−0.4, 1.3, −1.6, 0.7]*Q=[−0.4Q, 1.3Q, −1.6Q, 0.7Q].

Referring back to FIG. 13, at operation S1330, an initial scheduling priority of each UE at each time unit may be adjusted according to the scheduling adjustment weight values of each UE, to determine the scheduling priorities of each UE at different time units, thereby avoiding interference.

Specifically, firstly, an initial scheduling priority of one UE waiting to be scheduled at each time unit may be calculated for this UE according to a general method, and then the scheduling adjustment weight values of this UE are used to adjust the initial scheduling priority of this UE at each time unit, for example, the scheduling adjustment weight values of this UE are added to the initial scheduling priorities of this UE at the corresponding time units to obtain the scheduling priorities of this UE on each time unit, i.e., the enhanced scheduling priorities.

FIG. 21A illustrates a process of determining the scheduling priorities of the UE at the edge location, the UE at the middle location, and the UE at the central location in an interference measurement area according to an embodiment of the disclosure.

Referring to FIG. 21A, in this interference measurement area, the scheduling adjustment weight values of the UE at the edge location are [0.4Q, −1.3Q, 1.6Q, −0.7Q], the scheduling adjustment weight values of the UE at the central location are [−0.4Q, 1.3Q, −1.6Q, 0.7Q], and the scheduling adjustment weight values of the UE at the middle location are [0, 0, 0, 0], the initial scheduling priorities of the UE at the edge location at four time units are [PEdge0, PEdge1, PEdge2, PEdge3], the initial scheduling priorities of the UE at the middle location at the four time units are [PMe0, PMe1, PMe2, PMe3], and the initial scheduling priorities of the UE at the central location at the four time units are [PCe0, PCe1, PCe2, PCe3], and the scheduling adjustment weight values of this UE are added to the initial scheduling priorities of this UE at the corresponding time units, thereby obtaining the scheduling priorities of this UE on each time unit shown at the bottom of FIG. 21A, i.e., the enhanced scheduling priorities. Further, in the disclosure, the initial scheduling priority may be, for example, a proportional fair (PF) priority, however, the disclosure is not limited thereto.

Through the process of FIG. 13, the enhanced scheduling priorities of each UE within the current cell at each time unit may be determined. The disclosure obtains the enhanced scheduling priorities of the UE by adjusting the initial scheduling priorities of the UE based on the scheduling adjustment weight values of the UE, thus achieving a goal of using a higher priority to schedule the UE at the edge location in a low-interference time unit and to schedule the UE at the central location in a high-interference time unit.

In order to better understand the process of determining the area interference pattern, the area scheduling pattern, the scheduling adjustment weight value, and the scheduling priority, a summary is described below with reference to FIG. 21B.

FIG. 21B is a schematic diagram illustrating the process of determining an area interference pattern, an area scheduling pattern, a scheduling adjustment weight value, and a scheduling priority according to an embodiment of the disclosure. A total of three cells are illustrated in FIG. 21B.

Referring to FIG. 21B, for cell 0, a second period (e.g., 1 hour or 1 day) may be randomly started to divide cell 0 into a plurality of interference measurement areas. In a first second period P2, firstly, interference information (e.g., area interference pattern) about interference suffered by each interference measurement area is determined, and then, an area scheduling pattern of each interference measurement area is determined (e.g., an area scheduling pattern of a UE at a central location, an area scheduling pattern of a UE at a middle location and an area scheduling pattern of a UE at an edge location of each interference measurement area). When an area interference pattern of a certain interference measurement area changes, in order to maintain stable interference and scheduling patterns, to avoid inaccurate prediction of area interference patterns due to simultaneous modification of area interference patterns and/or area scheduling patterns by a plurality of cells, a next operation to determine the area interference pattern of this interference measurement area needs to be extended to 2×second period P2. Therefore, for this interference measurement area, within this 2×second period P2, scheduling adjustment weight values of each UE (e.g., the respective scheduling weight values of the UE at the edge location, the UE at the central location, and the UE at the middle location) are firstly determined based on the area scheduling pattern selected for each UE, and thereafter, initial scheduling priorities of each UE at each time unit (e.g., initial PF priorities) are adjusted based on the scheduling adjustment weight values of each UE, to determine scheduling priorities of each UE at each time unit (i.e., enhanced scheduling priorities). Finally, the determined scheduling priority for each UE at each time unit may be applied to the scheduling process.

For cells 1 and 2, processes are similar to that of cell 0, with a difference that the area interference pattern of each interference measurement area is not changed, and therefore the next operation to determine the area interference pattern/scheduling pattern of this interference measurement area is not extended to 2×second period P2.

Referring back to FIG. 2, at operation S230, UEs in the current cell are scheduled according to the determined scheduling priorities.

The operation S230 may include: determining available resources for a current cell at a current time unit; allocating resources to UEs in the current cell in a descending order of the scheduling priorities of all UEs in the current cell under the current time unit.

Specifically, step (1): remaining available resources for the current cell at the current time unit (e.g., the current time slot) may be queried. Wherein, a starting PRB location for allocating resources may be 0, or the starting PRB location for allocating resources may be determined according to a PRB randomization scheme. If there are the remaining available resources, the following step (2) is performed.

Step (2): an amount of data to be transmitted/received may be determined. Specifically, the amount of data to be transmitted/received of the UE with the highest scheduling priority among UEs that have not yet been allocated resources under the current time unit may be determined, in the descending order of the scheduling priorities of UEs within the current cell under the current time unit.

Step (3): resources required for the UE with the highest scheduling priority may be calculated based on the amount of data to be transmitted/received of this UE and a modulation coding scheme available by this UE.

Step (4): maximum required resources may be allocated for this UE within the remaining available resources of the current cell at the current time unit, from the previously determined starting PRB location.

Step (5): If there still are the remaining available resources for the current cell under the current time unit, the performing may start again from step (2) with respect to a next UE in descending order of scheduling priorities of all UEs in the current cell under the current time unit.

If it is found that there are no remaining available resources for the current cell at the current time unit in step (1) or step (5) above, the process of allocating resources to UEs with respect to the current time unit ends, and then reallocating resources to the UE in the current cell starts again from step (1) with respect to a next time unit.

In the above description, the utilized AI model is trained model. The training process of the AI model will be described in detail below with reference to FIG. 22.

FIG. 22 is a flowchart illustrating a process of training an AI model according to an embodiment of the disclosure.

Referring to FIG. 22, at operation S2210, an interference power sequence may be collected by performing interference management on a current base station through a plurality of predetermined area interference patterns, with respect to each of a plurality of time periods.

Specifically, a 5G system is at millisecond level, and resource scheduling is also at millisecond level. However, user service is at tens of seconds level. According to actual network statistics, an average talk time for voice is 88.31 seconds and an average duration of data service is 21.25 seconds. Therefore, it may be assumed that interference caused by the service does not change quickly and may maintain stable for a short period of time, compared to millisecond level of resource scheduling, i.e., an area interference pattern is assumed to maintain stable for one time period (e.g., 18 seconds). For ease of description, one time period of 18 seconds is used as an example in the following description.

In order to make the trained AI model sufficiently accurate, the interference power sequence need to be collected over a longer data collection time (e.g., 1 day or 1 week) as training data. The collected interference power sequence may include both uplink interference data and downlink interference data. The uplink interference data may be uplink interference data measured by a base station using different receiving beams corresponding to each interference measurement area to receive signals. The downlink interference data may be calculated by the base station from a RSRP and SINR reported by a UE. In addition, the base station determines an interference measurement area where the UE is located, by detecting a horizontal arrival angle of a signal of the UE.

Assuming that there are four time units (e.g., time slots) in each radio frame, there are a total of 18 predefined area interference patterns (pattern 0~pattern 17) as shown in FIG. 11. Therefore, one different predefined area interference pattern is used for each second of the 18 seconds, respectively, to perform manage interference on the current base station, and the interference power sequence in this predefined area interference pattern is collected in the meantime.

At operation S2220, for each time period, the interference power sequence collected within a current time period is marked, according to one predetermined area interference pattern in which a maximum average throughput is obtained within the current time period during the interference management.

Specifically, for example, within an 18-second period, if the average throughput is greatest when the interference management is performed on an actual network using the pattern 16 as shown in FIG. 11, the pattern 16 is selected as the optimal area interference pattern within this 18-second period, and then all interference power sequences collected during this 18-second period is marked using the pattern 16, that is, all interference power sequences collected during this 18-second period are labeled. According to a similar manner, a plurality of marked interference power sequences may be obtained by collecting the interference power sequences over a longer data collection time period (e.g., 1 day or 1 week) in one time period of 18 seconds and marking the collected interference power sequences using the corresponding optimal area interference pattern, wherein each marked interference power sequence contains an interference power sequence and a corresponding label (i.e., the corresponding optimal area interference pattern).

At operation S2230, the AI model may be trained by using the plurality of labeled interference power sequences.

Specifically, an interference power sequence among the plurality of labeled interference power sequences and its corresponding label (i.e., the corresponding optimal area interference pattern) are input to the AI model to complete the training of the AI model.

In addition, since each base station is in a different interference environment, the AI model used for each base station needs to be trained separately. In addition, the training of the AI model may be performed on the network node performing the present method, or the AI model may be trained outside of that network node and then the trained AI model may be transmitted to this network node, to save the resources of this network node. The AI model of which the training is completed may be used to predict the area interference pattern of the interference measurement area and the area interference pattern of the unit area described above. However, since it takes a long time to train the AI model before application, in order to enable the network node, when being started, to apply the method of the disclosure as soon as possible, the rule-based non-AI method described above may be used to predict the area interference pattern of each interference measurement area and the area interference pattern of the unit area before obtaining the training data for the AI model.

In order to make those skilled in the art more clearly understand the process of the method proposed in the disclosure, the process of the method will be described in general with reference to FIG. 23 below.

FIG. 23 is an example flowchart illustrating a method performed by a network node according to an embodiment of the disclosure.

First, at operation S2310, a starting PRB location may be determined, wherein the starting PRB location may be 0, or the starting PRB location for allocating resources may be determined based on a PRB randomization scheme.

At operation S2320, for a UE waiting to be scheduled, initial scheduling priorities of this UE at each time unit may be calculated according to a general method.

At operation S2330, an interference power sequence of a previous second may be is collected with respect to each interference measurement area, and an area interference pattern of each interference measurement area at a current second period is predicted based on the AI model or the rule-based non-AI method described above, according to interference power sequence of the previous second period collected for each interference measurement area.

At operation S2340, an area scheduling pattern of each interference measurement area at the current second period may be determined based on the predicted area interference pattern of each interference measurement area at the current second period.

At operation S2350, an impact factor Q is determined, wherein the impact factor Q may be adjusted in positive correlation with an average throughput calculated according to a third period, which has been described in detail above at operation S1820 with reference to FIG. 20 and will not be repeated here.

At operation S2360, scheduling adjustment weight values of each UE may be determined based on the impact factor Q and the area scheduling pattern of each UE.

At operation S2370, the initial scheduling priorities of each UE on each time unit may be adjusted based on the scheduling adjustment weight values of each UE, to determine a scheduling priority of each UE on each time unit.

At operation S2380, required resources may be allocated to UEs, within a full bandwidth from the starting PRB location, according to the scheduling priority of each UE in the current cell at each time unit.

Furthermore, although the operations of FIG. 23 are described in a certain order, some of the operations in FIG. 23 may be performed simultaneously, and the order of some operations may be changed, for example, operation S2310 may be performed at any moment before operations S2380 and S2320 may be performed at any moment before operation S2370.

FIG. 24 is a diagram illustrating a deployment scenario of a method according to an embodiment of the disclosure. The base station according to embodiments of the disclosure may be implemented with a distributed deployment according to a centralized unit (CU) configured to perform a function of upper layers (e.g., packet data convergence protocol (PDCP), radio resource control (RRC)) of an access network, and a distributed unit (DU) configured to perform a function of a lower layer. Between the core (e.g., 5G core (5GC) or next generation core (NGC)) network and the radio network (RAN), the deployment of the base station may be implemented in the order of CU, DU, and a radio unit (RU). The interface between the CU and DU may be referred to as an F1 interface.

The CU may be connected to one or more DUs to act as a higher layer than the DU. For example, the CU may be responsible for the functions of radio resource control (RRC) and packet data convergence protocol (PDCP) layers, and the DU and the RU may be responsible for the functions of lower layers. The DU may perform some functions (high PHY) of the radio link control (RLC), the media access control (MAC), and the physical (PHY) layers, and the RU may be responsible for the remaining functions (low PHY) of the PHY layer. Further, as an example, the digital unit (DU) may be included in the distributed unit (DU) according to implementation of a distributed arrangement of the base station. Hereinafter, unless otherwise defined, the operations of the DU will be described, but it is to be noted that various embodiments of the disclosure may be applied to both a base station deployment including the CU or a deployment in which the DU is directly connected to a core network, that is, being incorporated into a base station (e.g., an NG-RAN node) where the CU and the DU are one entity.

Referring to FIG. 24, when a network node performing the method of the disclosure is a base station, the method of the disclosure may be implemented and deployed in a Medium Access Control (MAC) module of a Distribution Unit (DU) of the base station, and in addition, the method of the disclosure may be divided into a non-real-time part of the method and a real-time part of the method, which are implemented in the MAC module of the DU as follows

The non-real-time part of the method is responsible for the following operations:

- (1) At the beginning of the method of the disclosure, an initial impact factor Q is set to the real-time part of the method.
- (2) Throughput data is acquired from the real-time part of the method, and the impact factor Q is adjusted in a third period P3 (e.g., 1 week), and the adjusted impact factor Q is updated to the real-time part of the method. Finally, an optimal impact factor Q is obtained by continuously adjusting the impact factor Q based on the acquired throughput data, and this optimal impact factor Q is updated to the real-time part of the method.

The real-time part of the method is responsible for the following operations:

- (1) A cell is divided into a plurality of interference measurement areas, and an area interference pattern of each interference measurement area at a current second period is predicted by a rule-based non-AI method based on an interference power sequence of each interference measurement area at different time units in a previous second period (e.g., 200 ms).
- (2) An area scheduling pattern of each interference measurement area is determined based on the predicted interference information (e.g., area interference pattern) of each interference measurement area.
- (3) Scheduling adjustment weight values of each UE used to adjust initial scheduling priorities is determined according to the impact factor Q, a location of each UE, and the area scheduling pattern of the interference measurement area to which each UE belongs.
- (4) The initial scheduling priorities of each UE on each time unit are adjusted based on the scheduling adjustment weight values of each UE, to determine the scheduling priorities of each UE on each time unit (i.e., enhanced scheduling priorities).
- (5) Resources are allocated to UEs in the current cell according to the scheduling priorities of each UE on each time unit (i.e., enhanced scheduling priorities).

In addition, if the method of the disclosure further applies the AI model described above, the method further comprises an AI model part that may be deployed in the MAC module of the DU, and this AI model part is responsible for: intelligently predicting the area interference pattern of each interference measurement area at the current second period by using historical interference information of each interference measurement area (e.g., interference power sequence collected for each interference measurement area at the previous second period (e.g., 200 ms)).

FIG. 25 is a block diagram illustrating a network node according to an embodiment of the disclosure.

Referring to FIG. 25, the network node 2500 includes a transceiver 2510 and a processor 2520, wherein the processor 2520 is coupled with the transceiver 2510 and is configured to perform the method performed by the network node described above with reference to FIGS. 2, 3A, 3B, 4 to 13, 14A, 14B, 15 to 20, 21A, 21B, 22, and 23. Details of the operations of the methods performed by the network node described above may be found described with reference to FIGS. 2, 3A, 3B, 4 to 13, 14A, 14B, 15 to 20, 21A, 21B, 22, and 23, all of which are not repeated herein.

The technical solution provided by the embodiments of the disclosure dynamically predicts high, medium and low interference intensity of each interference measurement area at different time units in a next period based on historical interference information, so at to adjust the scheduling priorities of UEs at different locations (e.g., UE at the edge location and UE at the central location), so that time units with relatively low interference may be assigned to UEs at the edge location sensitive to interference, and time units with relatively high interference may be assigned to the UE at the central locations with stronger interference resistance, thereby avoiding interference as actively as possible and reducing the impact of interference even under high loads.

Specifically, when scheduling data in a time unit with low interference, the UEs at the edge location may be scheduled preferentially by increasing scheduling priorities of the UEs at the edge location, and since channel qualities of the UEs at the edge location are poor, avoiding interference may improve a signal quality and modulation coding scheme, thus increasing the throughput. Meanwhile, when scheduling data in a time unit with high interference, the UEs at the central location are scheduled preferentially by increasing scheduling priorities of the UEs at the central location, and since the channel qualities of the UEs at the central location are good and the UEs at the central location are far from the interference of adjacent cells, the UEs have a better anti-interference capability and the throughput is unaffected or less affected. In addition, resources may be allocated for UEs in a full frequency band by using the method of the disclosure.

The above describes a method performed by a network node and the corresponding network node of the disclosure, respectively, and the following is a comparison with the related art in terms of effect when applying the method.

FIGS. 26A and 26B are schematic diagrams showing a comparison between the method of the disclosure and an original PRB interference random optimization scheme (hereinafter, becoming an original scheme) according to various embodiments of the disclosure.

Referring to FIG. 26A, compared to the original scheme, the method of the disclosure may improve an average throughput of a cell by about 19% in a case of a PRB utilization of 100%, the method of the disclosure may improve an average physical layer (IP) throughput of the cell by about 19% in a case of a PRB utilization of 50%, and the method of the disclosure may improve the average physical layer (IP) throughput of the cell by about 6% in a case of a PRB utilization of 30%.

Referring to FIG. 26B, with respect to the original scheme, the method of the disclosure may improve the average throughput of a UE at the cell edge by about 27% in a case of a PRB utilization of 100%, the method of the disclosure may improve the average IP throughput of the UE at the cell edge by about 17% in a case of a PRB utilization of 50%, and the method of the disclosure may improve the average IP throughput of the UE at the cell edge by about 9% in a case of a PRB utilization of 30%.

As seen, the method of the disclosure improves the throughput of the UE at the edge location through intelligent network interference management, and also improves the overall throughput of the network and enhances spectrum efficiency.

Further, according to embodiments of the disclosure, there may be provided an electronic apparatus including: at least one processor; and at least one memory storing computer executable instructions, wherein the computer-executable instructions, when being executed by the at least one processor, cause the at least one processor to perform the method as described above performed by the network node.

According to embodiments, a method performed by a network node, comprises predicting interference information about interference suffered by a current cell, determining scheduling priorities of respective user equipment (UE) at different time units according to the predicted interference information and locations of the UEs in the current cell, and scheduling UEs in the current cell according to the determined scheduling priorities.

According to an embodiment, the predicting the interference information about the interference suffered by the current cell comprises dividing the current cell into a plurality of interference measurement areas. The predicting the interference information about the interference suffered by the current cell comprises predicting interference information about interference suffered by each interference measurement area.

According to an embodiment, the dividing the current cell into the plurality of interference measurement areas comprises dividing the current cell into a plurality of unit areas with a current base station as the center. The dividing the current cell into the plurality of interference measurement areas comprises determining an area interference pattern for each of the plurality of unit areas. The dividing the current cell into the plurality of interference measurement areas comprises grouping the plurality of unit areas, according to area interference pattern sequence correlations between unit areas of the plurality of unit areas, to determine the plurality of interference measurement areas.

According to an embodiment, the determining the plurality of interference measurement areas comprises determining a first unit area as a pending interference measurement area. The determining the plurality of interference measurement areas comprises, with respect to the pending interference measurement area, performing an operation of determining the area interference pattern sequence correlation. The operation of determining the area interference pattern sequence correlation is used to determine the area interference pattern sequence correlation between each unit area in the pending interference measurement area and a next unit area adjacent to the pending interference measurement area in a predetermined direction. The determining the plurality of interference measurement areas comprises, if a minimum value among the area interference pattern sequence correlations between the next unit area and each unit area in the pending interference measurement area is greater than a first predetermined threshold, grouping the next unit area to the pending interference measurement area, and performing the operation of determining the area interference pattern sequence correlation with respect to the pending interference measurement area at this time. The determining the plurality of interference measurement areas comprises, if the minimum value among the area interference pattern sequence correlations between the next unit area and each unit area in the pending interference measurement area is not greater than the first predetermined threshold, determining the pending interference measurement area as a final interference measurement area, determining the next unit area as a new pending interference measurement area, and performing the operation of determining the area interference pattern sequence correlation with respect to the pending interference measurement area at this time.

According to an embodiment, the dividing the current cell into the plurality of interference measurement areas is performed at a beginning of each first period. Two consecutive first periods are spaced with separate random time.

According to an embodiment, the predicting the interference information about interference suffered by each interference measurement area comprises predicting an area interference pattern of a current interference measurement area at a current second period according to historical interference information of the current interference measurement area, wherein the area interference pattern represents a coding combination of an intensity of an intra-frequency interference of adjacent cells suffered by one area at respective time unit within a predetermined time.

According to an embodiment, the predicting the area interference pattern of the current interference measurement area at the current second period according to the historical interference information of the current interference measurement area comprises predicting the area interference pattern of the current interference measurement area at the current second period using the historical interference information of the current interference measurement area through an artificial intelligence (AI) model.

According to an embodiment, the predicting the area interference pattern of the current interference measurement area at the current second period according to the historical interference information of the current interference measurement area comprises determining one interference power matrix of the plurality of interference measurement areas, according to an interference power of each time unit in a previous second period collected at each interference measurement area. The predicting the area interference pattern of the current interference measurement area at the current second period according to the historical interference information of the current interference measurement area comprises selecting one area interference pattern matrix from an area interference pattern table including at least one area interference pattern matrix according to serial numbers of time units within the current second period. The predicting the area interference pattern of the current interference measurement area at the current second period according to the historical interference information of the current interference measurement area comprises predicting the area interference pattern of the current interference measurement area at the current second period according to the interference power matrix and the area interference pattern matrix.

According to an embodiment, the determining the one interference power matrix of the plurality of interference measurement areas, according to the interference power of each time unit in the previous second period collected at each interference measurement area comprises: performing the following operations with respect to each interference measurement area by averaging interference powers of time units with a same serial number in a previous second period with respect to the current interference measurement area, obtaining an average interference power of the current interference measurement area at the time units of the previous second period, filtering the average interference power, based on a historical filtered interference power of the current interference measurement area at the time units, to obtain a filtered interference power of the current interference measurement area at the time units of the previous second period, and setting the filtered interference power as the interference power of the current interference measurement area at the time units in the interference power matrix.

According to an embodiment, the historical filtered interference power of the current interference measurement area at the time units is updated as the filtered interference power of the current interference measurement area at the time units of the previous second period.

According to an embodiment, the predicting the area interference pattern of the current interference measurement area at the current second period according to the interference power matrix and the area interference pattern matrix comprises determining a result matrix based on the interference power matrix and the area interference pattern matrix. The predicting the area interference pattern of the current interference measurement area at the current second period according to the interference power matrix and the area interference pattern matrix comprises determining a maximum value among a plurality of matrix elements corresponding to the current interference measurement area from among the result matrix. The predicting the area interference pattern of the current interference measurement area at the current second period according to the interference power matrix and the area interference pattern matrix comprises determining an area interference pattern corresponding to the maximum value in the area interference pattern matrix as the area interference pattern of the current interference measurement area at the current second period.

According to an embodiment, the determining the scheduling priorities of the respective UE at different time units based on the predicted interference information and the locations of the UEs in the current cell comprises determining an area scheduling pattern of each interference measurement area based on the predicted interference information of each interference measurement area, wherein the area scheduling pattern represents a set of weight values for adjusting scheduling priorities of UEs, at different locations within the current interference measurement area, at different time units. The determining the scheduling priorities of the respective UE at different time units based on the predicted interference information and the locations of the UEs in the current cell comprises determining scheduling adjustment weight values of each UE based on a location of each UE and the area scheduling pattern of an interference measurement area to which the UE belongs. The determining the scheduling priorities of the respective UE at different time units based on the predicted interference information and the locations of the UEs in the current cell comprises adjusting an initial scheduling priority of each UE at each time unit according to the scheduling adjustment weight values of each UE, to determine the scheduling priorities of each UE at different time units.

According to an embodiment, the determining the area scheduling pattern of each interference measurement area based on the predicted interference information of each interference measurement area comprises determining an interference mean deviation weight of time units with the same serial number within a current second period The determining the area scheduling pattern of each interference measurement area based on the predicted interference information of each interference measurement area comprises determining a standard area scheduling pattern of each interference measurement area according to the interference mean deviation weight of the time units with the same serial number and an area interference pattern of each interference measurement area. The determining the area scheduling pattern of each interference measurement area based on the predicted interference information of each interference measurement area comprises determining area scheduling patterns of UEs at different locations in each interference area according to the standard area scheduling pattern of each interference measurement area.

According to an embodiment, the area scheduling patterns of UEs at different locations comprises at least one of an area scheduling pattern of a UE at a central location, an area scheduling pattern of a UE at a middle location, or an area scheduling pattern of a UE at an edge location.

According to an embodiment, the determining the interference mean deviation weight of time units with the same serial number within the current second period comprises determining an average interference power of the time units with the same serial number within the current second period and an average value of interference powers of all time units within the current second period. determining the interference mean deviation weight of time units with the same serial number within the current second period comprises calculating an absolute difference between the average interference power of the time units with the same serial number and the average value. The determining the interference mean deviation weight of time units with the same serial number within the current second period comprises determining the interference mean deviation weight of the time units with the same serial number according to the absolute difference of the time units with the same serial number.

According to an embodiment, the determining the scheduling adjustment weight values of each UE based on the location of each UE and the area scheduling pattern of the interference measurement area to which the UE belongs comprises selecting an area scheduling pattern corresponding to a location of the current UE from among the area scheduling pattern of the interference measurement area to which the current UE belongs, according to the location of the current UE. The determining the scheduling adjustment weight values of each UE based on the location of each UE and the area scheduling pattern of the interference measurement area to which the UE belongs comprises determining the scheduling adjustment weight values of the current UE according to an impact factor and the selected area scheduling pattern.

According to an embodiment, the method further comprises counting an average throughput of a current base station according to a third period. The method further comprises adjusting the impact factor, in positive correlation with the average throughput.

According to an embodiment, the method further comprises dynamically predicting high, medium and low interference intensity of each interference measurement area at different time units in a next period based on historical interference information.

According to an embodiment, when scheduling data in a time unit with low interference, the UEs at the edge location are scheduled preferentially by increasing scheduling priorities of the UEs at the edge location.

According to an embodiment, when the scheduling data in the time unit with high interference, the UEs at the central location are scheduled preferentially by increasing scheduling priorities of the UEs at the central location.

According to embodiments, comprises a transceiver, and at least one processor, coupled to the transceiver. The at least one processor configured to predict interference information about interference suffered by a current cell. The at least one processor configured to determine scheduling priorities of respective user equipment (UE) at different time units according to the predicted interference information and locations of the UEs in the current cell. The at least one processor configured to schedule UEs in the current cell according to the determined scheduling priorities.

According to embodiments, an electronic apparatus, comprises at least one processor, and at least one memory storing computer executable instructions. The computer executable instructions, when being executed by the at least one processor, cause the at least one processor to perform a method of predicting interference information about interference suffered by a current cell, determining scheduling priorities of respective user equipment (UE) at different time units according to the predicted interference information and locations of the UEs in the current cell, and scheduling UEs in the current cell according to the determined scheduling priorities.

In embodiments, a method performed by a network node is provided. The method comprises obtaining interference information for a current cell, the interference information for indicating interference levels at time units of a next period in each interference measurement area of a plurality of interference measurement areas of the current cell. The method comprises obtaining scheduling priorities of user equipments (UEs) at the time units of the next period according to the interference information and locations of the UEs in the current cell. The method comprises performing a scheduling of the UEs in the current cell according to the scheduling priorities.

According to an embodiment, each of the interference levels indicates one of a high interference intensity, a medium interference intensity, or a low interference intensity. The interference levels at the time units of the next period are obtained based on an artificial intelligence (AI) model using interference levels at least one previous period in each interference measurement area of the plurality of interference measurement areas of the current cell.

According to an embodiment, the locations of the UEs are divided into a plurality of levels including a cell-edge level and a cell-center level. When scheduling data in a time unit with the low interference intensity, a scheduling priority of a UE at the cell-edge level increases. When scheduling data in a time unit with the high interference intensity, a scheduling priority of a UE at the cell-center level increases.

According to an embodiment, the obtaining of the interference information comprises dividing the current cell into a plurality of unit areas with a current base station as a center. The obtaining of the interference information comprises determining an area interference pattern for each of the plurality of unit areas. The obtaining of the interference information comprises grouping the plurality of unit areas, according to area interference pattern sequence correlations between unit areas of the plurality of unit areas, to determine the plurality of interference measurement areas.

According to an embodiment, the grouping of the plurality of unit areas comprises determining a first unit area as a pending interference measurement area. The grouping of the plurality of unit areas comprises, with respect to the pending interference measurement area, performing an operation of determining the area interference pattern sequence correlation, the operation of determining the area interference pattern sequence correlation being used to determine the area interference pattern sequence correlation between each unit area in the pending interference measurement area and a next unit area adjacent to the pending interference measurement area in a predetermined direction. The grouping of the plurality of unit areas comprises, if a minimum value among the area interference pattern sequence correlations between the next unit area and each unit area in the pending interference measurement area is greater than a first predetermined threshold, grouping the next unit area to the pending interference measurement area, and performing the operation of determining the area interference pattern sequence correlation with respect to the pending interference measurement area at this time. The grouping of the plurality of unit areas comprises, if the minimum value among the area interference pattern sequence correlations between the next unit area and each unit area in the pending interference measurement area is not greater than the first predetermined threshold, determining the pending interference measurement area as a final interference measurement area, determining the next unit area as a new pending interference measurement area, and performing the operation of determining the area interference pattern sequence correlation with respect to the pending interference measurement area at this time.

According to an embodiment, the obtaining of the interference information comprises predicting an area interference pattern of a current interference measurement area at a current second period according to historical interference information of the current interference measurement area, wherein the area interference pattern represents a coding combination of an intensity of an intra-frequency interference of adjacent cells suffered by one area at respective time unit within a predetermined time.

According to an embodiment, the predicting the area interference pattern of the current interference measurement area at the current second period according to the historical interference information of the current interference measurement area comprises determining one interference power matrix of the plurality of interference measurement areas, according to an interference power of each time unit in a previous second period collected at each interference measurement area. The predicting the area interference pattern of the current interference measurement area at the current second period according to the historical interference information of the current interference measurement area comprises selecting one area interference pattern matrix from an area interference pattern table including at least one area interference pattern matrix according to serial numbers of time units within the current second period. The predicting the area interference pattern of the current interference measurement area at the current second period according to the historical interference information of the current interference measurement area comprises predicting the area interference pattern of the current interference measurement area at the current second period according to the interference power matrix and the area interference pattern matrix.

According to an embodiment, the determining the one interference power matrix of the plurality of interference measurement areas, according to the interference power of each time unit in the previous second period collected at each interference measurement area comprises performing the following operations with respect to each interference measurement area by averaging interference powers of time units with a same serial number in a previous second period with respect to the current interference measurement area, obtaining an average interference power of the current interference measurement area at the time units of the previous second period, filtering the average interference power, based on a historical filtered interference power of the current interference measurement area at the time units, to obtain a filtered interference power of the current interference measurement area at the time units of the previous second period, and setting the filtered interference power as the interference power of the current interference measurement area at the time units in the interference power matrix.

According to an embodiment, the historical filtered interference power of the current interference measurement area at the time units is updated as the filtered interference power of the current interference measurement area at the time units of the previous second period.

According to an embodiment, the obtaining of the area interference pattern of the current interference measurement area at the current second period according to the interference power matrix and the area interference pattern matrix comprises determining a result matrix based on the interference power matrix and the area interference pattern matrix. The obtaining of the area interference pattern of the current interference measurement area at the current second period according to the interference power matrix and the area interference pattern matrix comprises determining a maximum value among a plurality of matrix elements corresponding to the current interference measurement area from among the result matrix. The obtaining of the area interference pattern of the current interference measurement area at the current second period according to the interference power matrix and the area interference pattern matrix comprises determining an area interference pattern corresponding to the maximum value in the area interference pattern matrix as the area interference pattern of the current interference measurement area at the current second period.

According to an embodiment, the obtaining of the scheduling priorities comprises determining an area scheduling pattern of each interference measurement area based on the predicted interference information of each interference measurement area, wherein the area scheduling pattern represents a set of weight values for adjusting scheduling priorities of the UEs, at different locations within the current interference measurement area, at different time units. The obtaining of the scheduling priorities comprises determining scheduling adjustment weight values of each UE based on a location of each UE and the area scheduling pattern of an interference measurement area to which the UE belongs. The obtaining of the scheduling priorities comprises adjusting an initial scheduling priority of each UE at each time unit according to the scheduling adjustment weight values of each UE, to obtain the scheduling priorities of the UEs at different time units.

According to an embodiment, the determining the area scheduling pattern of each interference measurement area based on the predicted interference information of each interference measurement area comprises determining an interference mean deviation weight of time units with the same serial number within a current second period, The determining the area scheduling pattern of each interference measurement area based on the predicted interference information of each interference measurement area comprises determining a standard area scheduling pattern of each interference measurement area according to the interference mean deviation weight of the time units with the same serial number and an area interference pattern of each interference measurement area. The determining the area scheduling pattern of each interference measurement area based on the predicted interference information of each interference measurement area comprises determining area scheduling patterns of UEs at different locations in each interference area according to the standard area scheduling pattern of each interference measurement area.

According to an embodiment, the area scheduling patterns of UEs at different locations comprises at least one of an area scheduling pattern of a UE at a central location, an area scheduling pattern of a UE at a middle location, or an area scheduling pattern of a UE at an edge location.

According to an embodiment, the determining the interference mean deviation weight of time units with the same serial number within the current second period comprises determining an average interference power of the time units with the same serial number within the current second period and an average value of interference powers of all time units within the current second period. The determining the interference mean deviation weight of time units with the same serial number within the current second period comprises calculating an absolute difference between the average interference power of the time units with the same serial number and the average value. The determining the interference mean deviation weight of time units with the same serial number within the current second period comprises determining the interference mean deviation weight of the time units with the same serial number according to the absolute difference of the time units with the same serial number.

According to an embodiment, the determining the scheduling adjustment weight values of each UE based on the location of each UE and the area scheduling pattern of the interference measurement area to which the UE belongs comprises selecting an area scheduling pattern corresponding to a location of a current UE from among the area scheduling pattern of the interference measurement area to which the current UE belongs, according to the location of the current UE. The determining the scheduling adjustment weight values of each UE comprises determining the scheduling adjustment weight values of the current UE according to an impact factor and the selected area scheduling pattern.