The present invention relates to a scheduling method in wireless communication.
In wireless systems of the related art, a plurality of terminals are connected to a base station, and the base station determines which terminal to communicate with in every time slot. This determination processing is called scheduling. This may be called user scheduling. For scheduling, a method such as proportional fairness is used (NPL 1).
In proportional fairness, radio resources are allocated to terminals in which instantaneous throughputs expected when the radio resources are allocated is greater than an average throughput until that time.
In recent years, areas where a plurality of base stations of different types are installed have increased. In areas where a plurality of base stations of different types are installed, situations in which connection destination base stations of terminals and wireless schemes (communication schemes) are different are assumed.
On the other hand, scheduling according to the technology of the related art has been performed in accordance with wireless communication quality, priority, or the like in a base station. Therefore, in the technology of the related art, a plurality of base stations of different types are installed, and there is a problem that scheduling cannot be appropriately performed in situations in which connection destination base stations of terminals and wireless schemes are different.
The present invention has been devised in view of the foregoing circumstances and an object of the present invention is to provide a technology for appropriately performing scheduling in a wireless system in which a plurality of base stations provide services to users.
According to the disclosed technology, a scheduling control apparatus calculates a weight for scheduling a terminal in a wireless system including a plurality of base stations. The scheduling control apparatus includes: an information acquisition unit configured to acquire information regarding a connection state between each of the plurality of base stations and the terminal;
The weight calculation unit calculates the weight by repeating processing for updating the weight so that an error between expected wireless communication quality obtained when the weight is given to each of the terminals and required wireless communication quality of each of the terminals is small.
According to the disclosed technology, there is provided a technology for appropriately performing scheduling in a wireless system in which a plurality of base stations provides services to users.
Embodiments of the present invention (present embodiment) will be described below with reference to the drawings. Embodiments to be described below is merely exemplary and embodiments to which the present invention is applied are not limited to the following embodiments.
As illustrated in
As one example, the base station 10A is a wireless LAN base station (access point), and the base station 10B is a base station of a cellular network such as 5G. The terminal can be connected to both the base station 10A and the base station 10B.
In the wireless system illustrated in
The scheduling control apparatus 100 calculates a weight for each terminal so that wireless communication quality required by each terminal accommodated in a wireless system having a plurality of base stations is satisfied, and notifies each base station of the calculated weight. Each base station performs scheduling for each terminal in accordance with the weight calculated by the scheduling control apparatus 100. In the following description, a throughput is used as a specific example of wireless communication quality in the terminal.
A “weight” for a certain terminal is the proportion of a communication amount allocated to the terminal in the total communication amount that can be allocated by a base station (a base station connected to the terminal) per unit time. In other words, the “weight” indicates the allocation ratio of communication amounts between one or a plurality of terminals connected to a base station.
The term “communication amount” may be a transmission rate, a line capacity, a traffic amount, the number of packets, a resource amount, the number of symbols, the number of slots, or another amount. Target communication to which a communication amount is allocated may be uplink communication, downlink communication, or both uplink communication and downlink communication.
When w is a “weight” for a certain terminal in a certain base station and r is a transmission rate for all terminals connected to the base station, an expected throughput s between the terminal and the base station is calculated as s=r×w. When a terminal is connected to a plurality of base stations, an expected throughput s is obtained by adding r×w for each base station.
The information acquisition unit 110 acquires information regarding a connection state between a base station and a terminal from, for example, a control device in the wireless system or from each base station. The weight calculation unit 120 calculates a weight. The control unit 130 notifies each base station of information regarding the weight calculated by the weight calculation unit 120.
The data storage unit 140 stores information necessary for the weight calculation unit 120 to calculate a weight. The information necessary to calculating the weight is, for example, a transmission rate of each base station, a required throughput of each terminal, or the like.
Next, an operation example of the scheduling control apparatus 100 will be described in accordance with a procedure of the flowchart illustrated in
The information acquisition unit 110 acquires information on a base station and a terminal in an area to which the technology according to the present embodiment is applied. For example, the information acquisition unit 110 acquires a connection state between a base station and a terminal (which base station is connected to which terminal) from a control device or the like of the wireless system, and acquires a transmission rate (line capacity) of the base station and the required throughput of the terminal from the data storage unit 140.
In S102, the weight calculation unit 110 performs weight calculation processing. The weight calculation processing will now be described with reference to a specific example illustrated in
In the example illustrated in
In the example illustrated in
Each base station has a transmission rate (line capacity) for terminals (all connectable terminals). Here, a transmission rate of the base station #1 is defined as R1, and a transmission rate of the base station #2 is defined as R2.
When weights illustrated in
Here, a sum of the weights for the same base station is 1. That is, w11+w12=1, and w22+w23+ . . . , w2N=1. It is assumed that an initial value of a weight between a base station and each terminal connected to the base station is given in advance in each base station.
For example, when a required throughput (required throughput) of the terminal #1 is defined as Y1 and Y1 (required throughput)>TH1 (expected throughput) is satisfied, the weight calculation unit 110 updates the weight so that the weight (that is, w11) of the terminal #1 increases. That is, the weight calculation unit 110 updates the weight so that the weight of the terminal of which the expected throughput is less than the required throughput increases.
Here, since a sum of the weights for the same base station is 1, as the weight is updated, the other weights are updated. For example, since w11+w12=1 in the base station #1 to which the terminal #1 is connected, w12 decreases when w11 increases.
By repeating the above-described processing of the weight updating and the expected throughput calculation, the weight is determined so that the expected throughput satisfies the required throughput as much as possible in each terminal.
More specifically, the weight calculation unit 120 can update the weight according to the following update formula.
In the above formula, x represents a transmission rate of the base station. wi represents a weight of terminal #i. Each of x and wi is a vector. y is a required throughput of the terminal #i.
wi·x represents an expected throughput of the terminal #i. For example, when the terminal #i is the terminal #2 in the situation of
In g(w) (which is a square value of a difference between the required throughput and the expected throughput), a change amount (∂g(w)/∂wi) (=inclination) for wi is obtained. The inclination is used to update wi by the third formula so that g(w) decreases (=the realization throughput approaches the required throughput). ρ is a predetermined coefficient. This method is a gradient method (which may be referred to as a gradient descent method).
The wi can be calculated for each terminal similarly to the parameter updating (for example, an error back propagation method) of a neural network so that the following mean square error (which may be called “dispersion”) is minimized. In the following formula, E is a sum for the terminals, and N is the number of terminals.
The weight calculation processing (repeated processing of the weight updating and the expected throughput calculation) in S102 is repeated a predetermined prescribed number of times until the mean square error is equal to or less than a predetermined threshold. When the mean square error is not equal to or less than the predetermined threshold as a result of repeating the weight calculation processing the prescribed number of times, the processing of S102 ends.
Next, S103, S104, and S105 in the flow of
In S103, the weight calculation unit 110 determines whether the mean square error is equal to or less than the threshold.
When a determination result of S103 is Yes (mean square error≤threshold), in S104, the weight calculation unit 110 transfers the weights of all the terminals to the control unit 130. The control unit 130 notifies each base station of the weight(s). The control unit 130 may notify the base station of only the weight(s) related to the base station. For example, in the example illustrated in
Each base station receiving the weight(s) performs scheduling according to the weight(s). As an example, a communication operation of distributing resources in a time-division manner will be described. For example, in the connection state of
When the determination result of S103 is No (mean square error>threshold), the processing proceeds to S105. The weight calculation unit 110 updates the required throughput and performs the weight calculation processing of S102, using the updated required throughput.
Since the required throughput is a throughput required for the terminal, an initially set value of the required throughput may be basically used as long as an initially set value can be used as it is.
However, even if the weight updating processing is repeated, the mean square error is not equal to or less than the threshold in some cases. In these cases, the required throughput is updated, and the weight update processing is performed using the updated required throughput, so that the mean square error becomes equal to or less than the threshold. By using the weight and the required throughput so that the mean square error becomes equal to or less than the threshold, it can be expected that communication in which the radio resources of the wireless system are used most efficiently can be performed.
An initial value of the required throughput is not necessarily set for each terminal. That is, the initial value of the required throughput may be set to a predetermined appropriate value.
Although a method of updating the required throughput is not limited to a specific method, the following updating formula can be used, for example.
As in the foregoing formula, the required throughput y is changed at a rate of change of ∂f(W)/∂y. α is a predetermined coefficient. This method is a gradient method (which may be referred to as a gradient descent method). The larger f(W) (mean square error) is, the larger (an absolute value) of ∂f(W)/∂y is. Therefore, y can be changed greatly.
The scheduling control apparatus 100, the base station, and the terminal can all be implemented, for example, by causing a computer to execute a program. The computer may be a physical computer or a virtual machine on a cloud. The scheduling control apparatus 100, the base station and the terminal are collectively called devices.
That is, the device can be implemented by executing a program corresponding to the processing performed by the device using hardware resources such as a CPU and a memory contained in a computer. The foregoing program can be recorded on a computer-readable recording medium (a portable memory or the like) to be stored or distributed. The foregoing program can also be provided through a network such as the Internet or an e-mail.
A program implementing processing in the computer is provided by, for example, a recording medium 1001 such as a CD-ROM or a memory card. When the recording medium 1001 that stores a program is set in the drive device 1000, the program is installed from the recording medium 1001 to the auxiliary storage device 1002 via the drive device 1000. However, the program need not necessarily be installed from the recording medium 1001 and may be downloaded from another computer via a network. The auxiliary storage device 1002 stores the installed program and also stores necessary files, data, and the like.
The memory device 1003 reads and stores the program from the auxiliary storage device 1002 when an instruction to start the program is given. The CPU 1004 implements functions related to the device in accordance with the program stored in the memory device 1003. The interface device 1005 is used as an interface for connection to a network or the like. The display device 1006 displays a graphical user interface (GUI) or the like by the program. The input device 1007 is configured with a keyboard, a mouse, a button, a touch panel, or the like, and is used to input various operation instructions. The output device 1008 outputs a calculation result.
According to the technology according to the above-described present embodiment, scheduling for providing wireless communication quality (a throughput or the like) required by the terminal can be performed in a situation where the connection destination base station of each terminal and wireless communication quality are different.
The following supplements will be disclosed according to the above-described embodiments.
A scheduling control apparatus that calculates a weight for scheduling a terminal in a wireless system including a plurality of base stations, the scheduling control apparatus comprising:
The scheduling control apparatus according to Supplement 1, wherein the processor calculates the weight so that a mean square error between the required wireless communication quality and the expected wireless communication quality is equal to or less than a threshold.
The scheduling control apparatus according to Supplement 2, wherein the processor updates the required wireless communication quality when the mean square error is not equal to or less than the threshold after the processing is repeated a prescribed number of times, and calculates the weight using the updated required wireless communication quality.
A scheduling control system comprising:
A scheduling control method executed by a computer that calculates a weight for scheduling a terminal in a wireless system including a plurality of base stations, the method comprising:
A non-transitory storage medium storing a program that can be executed by a computer to perform weight calculation processing for calculating a weight for scheduling a terminal in a wireless system including a plurality of base stations, the weight calculation processing comprising:
Although the embodiment has been described above, the present invention is not limited to the specific embodiment, and various modifications and changes can be made within the scope of the gist of the present invention described in the claims.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2022/006258 | 2/16/2022 | WO |