Resource allocation method, resource allocation device, and program
The resource allocation method in heterogeneous networks addresses the challenge of efficiently allocating resources by scoring, normalizing, and variance-based ordering of user requests, ensuring that users with diverse communication needs are satisfied without overwhelming high-performance links, thus maintaining network performance and user satisfaction.
Patent Information
- Application Number
- PCT/JP2023/020545
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-06-01
- Publication Date
- 2024-12-05
AI Technical Summary
In heterogeneous networks with multiple links of varying communication performance, existing methods struggle to efficiently allocate resources to users with different requirements, often leading to high computational loads and potential load concentration on high-performance links, which can result in unsatisfied user demands and communication performance degradation.
A resource allocation method that scores and normalizes user requests for each communication performance type, calculates variance values to determine resource allocation order, and allocates resources to users with the flattest dispersion of requests, preventing load concentration on high-performance links.
This approach effectively satisfies user requests by distributing resources based on communication performance needs, preventing load concentration and ensuring that users with biased requests do not overwhelm high-performance links, thereby maintaining network performance and user satisfaction.
Smart Images

Figure JP2023020545_05122024_PF_FP_ABST
Abstract
Description
Resource allocation method, resource allocation device, and program
[0001] The present disclosure relates to techniques for allocating available resources to multiple users, and more particularly to techniques suitable for use in allocating resources to users in a heterogeneous network.
[0002] In recent years, heterogeneous networks that utilize multiple links with different communication characteristics have become widespread. When multiple users with different requirements coexist in a heterogeneous network, resource allocation that can appropriately satisfy the requirements of each user is required.
[0003] Generally, methods for finding an optimal solution that satisfies the requirements of all users are known to use brute force methods and multivariate optimization. However, these methods require a large amount of calculations, placing a heavy load on the computer. Therefore, there is a need for a method that can allocate appropriate resources to each user with less calculations than brute force methods.
[0004] Furthermore, in resource allocation in a heterogeneous network, if the communication performance of links from the viewpoint of each user is highly similar, it is not easy to determine which link's resources should be allocated to which user. Furthermore, there is a risk that the load will be concentrated on a specific link with excellent communication performance.
[0005] Examples of documents that show the state of the art in the technical field related to the present disclosure include Patent Document 1 and Patent Document 2. However, the technology disclosed in either document does not solve the above-mentioned problem.
[0006] JP 2017-038119 A JP 2020-123848 A
[0007] The present disclosure has been made in view of the above-mentioned problems, and aims to appropriately satisfy the requirements of each user when resources of multiple links with different communication performances are allocated to multiple users with different requirements.
[0008] The present disclosure provides a resource allocation method for achieving the above-mentioned object. The resource allocation method of the present disclosure is a computer-executable method and includes the following first to fifth steps. The first step is a step of acquiring, for each of multiple users, a request value for each of multiple types of communication performance required for a link. The second step is a step of scoring the request values from the multiple users for each of the multiple types of communication performance. The third step is a step of normalizing the scored request values for each of the multiple types of communication performance among the multiple users. The fourth step is a step of calculating a variance value between the scored and normalized request values for each of the multiple types of communication performance for each of the multiple users. The fifth step is a step of determining resource allocation target users in ascending order of variance value among the multiple users, and allocating resources of a link selected from one or more available links to the resource allocation target users.
[0009] The present disclosure provides a resource allocation device for achieving the above-mentioned object. The resource allocation device of the present disclosure includes at least one processor, at least one memory coupled to or incorporated in the at least one processor, and a plurality of instructions executable by the at least one processor stored in the at least one memory. The plurality of instructions are configured to cause the at least one processor to execute the following first to fifth processes. The first process is a process of acquiring, for each of a plurality of users, a request value for each of a plurality of types of communication performance required for a link. The second process is a process of scoring the request values from the plurality of users for each of the plurality of types of communication performance. The third process is a process of normalizing the scored request values for each of the plurality of types of communication performance among the plurality of users. The fourth process is a process of calculating a variance value between the scored and normalized request values for each of the plurality of types of communication performance for each of the plurality of users. The fifth process is a process of determining a resource allocation target user in ascending order of variance value among the plurality of users and allocating resources of a link selected from one or more available links to the resource allocation target user.
[0010] The present disclosure provides a program for achieving the above object. The program of the present disclosure is a program including a plurality of instructions executable by at least one processor. The plurality of instructions included in the program of the present disclosure are configured to cause the at least one processor to execute the first to fifth processes described above. The program of the present disclosure may be stored in a computer-readable non-transitory storage medium or provided via a network.
[0011] According to the technology disclosed herein, link resources are allocated in order from the user with the flattest distribution of requirements for link communication performance, which makes it possible to prevent resources of a link with broadly high communication performance from being monopolized by users with biased requirements. Therefore, according to the technology disclosed herein, when resources of multiple links with different communication performances are allocated to multiple users with different requirements, the requirements of each user can be appropriately met.
[0012] 1 is a diagram illustrating an example of multiple link communication by a heterogeneous terminal. FIG. 1 is a diagram illustrating an example of a link performance table (Table 1) indicating the communication performance of a plurality of links constituting a heterogeneous network, and an example of a required performance table (Table 2) indicating the required performance of a plurality of users accommodated in the heterogeneous network. FIG. 2 is a diagram illustrating scoring of the numerical values of required performance. FIG. 3 is a diagram illustrating normalization of scored required values. FIG. 4 is a diagram illustrating calculation of the variance value of normalized scores. FIG. 5 is a diagram illustrating determination of the ranking of resource allocation. FIG. 6 is a diagram illustrating calculation of a weight index for each required performance of a resource allocation target user. FIG. 7 is a diagram illustrating calculation of a fulfillment rate of the required performance of a resource allocation target user. FIG. 8 is a diagram illustrating satisfaction level of the requirements of a resource allocation target user. FIG. 9 is a diagram illustrating calculation of the degree of excess or deficiency of communication performance with respect to the required performance of a resource allocation target user. FIG. 10 is a diagram illustrating calculation of the deviation degree of communication performance with respect to the required performance of a resource allocation target user. FIG. 11 is a diagram illustrating updating of the link performance table. FIG. 12 is a flow diagram illustrating the flow of a resource allocation method according to an embodiment of the present disclosure. FIG. 13 is a block diagram illustrating the configuration of a resource allocation device according to an embodiment of the present disclosure.
[0013] 1. Application Example of Resource Allocation Method With the recent development of wireless access technology, the improvement of convenience, and the expansion of demand, terminals such as smartphones and network cameras have become widespread. These terminals have multiple wireless interfaces with different characteristics, such as LTE / 5G cellular networks, IEEE 802.11, and IEEE 802.15. Terminals may also have wired interfaces. Furthermore, in the future, support for multi-link communication, which is scheduled to be introduced in Wi-Fi 7, is expected. Today, heterogeneous networks can be built using heterogeneous terminals capable of such diverse access and communication capabilities.
[0014] The expansion of coverage and widespread use of networks due to the development of today's access technologies has led to a diversification of applications and their requirements, creating an environment in which terminals with different communication performance requirements are accommodated in a single network. In such an environment, problems such as increased traffic are a concern. To address this concern, attempts are being made to achieve communication characteristics that could not be achieved with a single link by integrating different types of access and communicating over multiple links. Specifically, attempts are being made to achieve higher bandwidth by aggregating the communication bandwidth of multiple links, and to actively utilize low-latency links through multihoming and multipath TCP communication to reduce communication latency.
[0015] Figure 1 shows an example of a network that accommodates multiple heterogeneous terminals (hereinafter referred to as users) with different requirements for communication performance, where each user performs multi-link communication mainly using a link that has the communication performance that meets their requirements. By utilizing multi-link communication in this way, it becomes possible to effectively use network resources and achieve user satisfaction that could not be achieved with single-link communication.
[0016] However, for example, if link C is a link with broadly high communication performance, there is a risk that the communication performance of the entire network will decline due to the load being concentrated on link C. Also, there is a risk that the communication performance required by other users will not be satisfied if the resources of link C are monopolized by a portion of users with biased requirements. Furthermore, for example, if there is a high similarity in communication performance between link C and link D as seen from each user, it is not easy to determine which link's resources should be allocated to which user.
[0017] In one embodiment of the present disclosure, by applying the resource allocation method of the present disclosure to the heterogeneous network described above, resource allocation that can appropriately satisfy the requirements of each user is realized. The resource allocation method according to one embodiment of the present disclosure will be described below.
[0018] 2. Description of Resource Allocation Method 2-1. Heterogeneous Network Table 1 shown in Figure 2 is an example of a link performance table showing the communication performance of multiple links constituting a heterogeneous network. The communication performance items shown in Table 1 are items to be evaluated among various communication performances that characterize the link, such as communication bandwidth, delay, power consumption, jitter, etc. Table 2 shown in Figure 2 is an example of a required performance table showing the required performance of multiple users accommodated in the heterogeneous network. Since the user required performance is the performance that the user requires of the link, the same items as the communication performance items of the link are set in the user required performance items. In the example shown in Figure 2, communication bandwidth, delay, and power consumption are communication performance items, and required communication bandwidth, allowable delay, and allowable power consumption are user required performance items.
[0019] In the example shown in FIG. 2, the heterogeneous network has ten links L 1 -L 10 It consists of ten users 1 -U 10 As shown in Table 1, each link L 1 -L 10 As shown in Table 2, each user U 1 -U 10have different requirements for communication performance. The resource allocation method according to this embodiment is applied to a heterogeneous network having the characteristics shown in Tables 1 and 2.
[0020] 2-2. Scoring of Required Performance Values As described above, in this embodiment, the communication performance required by a user for a link is communication bandwidth, delay, and power consumption. Evaluating which of these communication performances each user places importance on is important in allocating resources to appropriately satisfy each user's requirements. However, because communication bandwidth, delay, and power consumption are different indicators, it is not possible to uniformly compare their merits. Therefore, in the resource allocation method according to this embodiment, first, the numerical values of each required performance (hereinafter, these numerical values may be referred to as required values) are scored, enabling uniform comparisons between users.
[0021] FIG. 3 is a diagram illustrating scoring of the numerical value of required performance. There are cases where the larger the numerical value, the higher the performance required by the user, and cases where the larger the numerical value, the lower the performance required by the user. The required communication bandwidth is an example of the former. When scoring the numerical value of the required communication bandwidth, the numerical value of the required communication bandwidth is calculated as the bandwidth score as is. On the other hand, the allowable delay and the allowable power consumption are examples of the latter. When scoring the numerical value of the allowable delay, the reciprocal value of the numerical value of the allowable delay is calculated as the delay score. Furthermore, when scoring the numerical value of the allowable power consumption, the reciprocal value of the numerical value of the allowable power consumption is calculated as the power consumption score. By comparing the score values calculated in this way between users, it is possible to uniformly compare the level of required performance between users.
[0022] 2-3. Normalization of Scored Request Values In order to evaluate which of the communication performances each user places importance on, it is necessary to compare the required performances. However, simply scoring the required values does not allow for comparison based on the difference in scale between the required performances. Therefore, in the resource allocation method according to this embodiment, the scored required values are normalized, thereby making it possible to compare the level of the required performance between the required performances.
[0023] FIG. 4 is a diagram illustrating normalization of scored requirement values. Normalization is performed using the highest score value for each required performance. For the required communication bandwidth, the maximum value among all users' bandwidth scores is 900, so all values in the bandwidth score column are divided by 900. For the allowable delay, the maximum value among all users' delay scores is 1 / 15, so all values in the delay score column are divided by 1 / 15. For the allowable power consumption, the maximum value among all users' power consumption scores is 1 / 3, so all values in the power consumption score column are divided by 1 / 3. By comparing the normalized score values calculated in this way between required performances, it is possible to evaluate which of the communication bandwidth, delay, and power consumption each user prioritizes.
[0024] 2-4. Calculation of Variance of Normalized Score By scoring the required value of each required performance as described above and then normalizing it, it is possible to check the bias in the required performance of each user. In the resource allocation method according to this embodiment, the bias in the required performance of each user is quantified by calculating the variance between the normalized scores of the required performance.
[0025] 5 is a diagram illustrating the calculation of the variance of the normalized score. The variance is calculated for each user using the normalized bandwidth score, normalized delay score, and normalized power consumption score as samples. A user with a large variance value means that the user has a large deviation in required performance, while a user with a small variance value means that the user has a small deviation in required performance, i.e., the user has a flat distribution of requirements for the communication performance of the link.
[0026] 2-5. Determining the Resource Allocation Order The variance value of the normalized scores calculated for each user is used to determine the resource allocation order among users. In the resource allocation method according to this embodiment, users to whom resources are allocated (hereinafter referred to as resource allocation target users) are determined in ascending order of variance value among users.
[0027] 6 is a diagram illustrating how resource allocation ranking is determined. In the example shown in FIG. 6, user U9 has the smallest variance value. Therefore, resources are allocated to user U9 first. Next, resources are allocated to user U7, which has the second smallest variance value, and so on in ascending order of variance value. Finally, resources are allocated to user U5, which has the largest variance value.
[0028] In this embodiment, link resources are allocated in order starting with users with the flattest distribution of requirements, rather than starting with users with biased requirements for link communication performance. This makes it possible to prevent a decrease in overall user satisfaction due to competition for resources on the same link, which can occur when the similarity in communication performance between links is high from the perspective of each user.
[0029] 2-6. Calculation of weight index for each required performance of resource allocation target users Next, we will explain how to determine resources to be allocated to resource allocation target users determined in the above order. In the following explanation, user U9 with the smallest variance value is considered to be the resource allocation target user.
[0030] First, according to the resource allocation method of this embodiment, a weight index representing the importance of each required performance for a user to whom resources are to be allocated is calculated using a normalized score of each required performance.
[0031] FIG. 7 is a diagram illustrating the calculation of a weight index for each required performance of a resource allocation target user. In calculating the weight index, first, the average value between the normalized scores is calculated for the resource allocation target user. Then, the ratio of each normalized score to the average value is calculated as the weight index. Specifically, the bandwidth weight index is obtained by dividing the normalized bandwidth score by the average value, the delay weight index is obtained by dividing the normalized delay score by the average value, and the power consumption weight index is obtained by dividing the normalized power consumption score by the average value. In the example shown in FIG. 7, the power consumption weight index has the largest value and the bandwidth weight index has the smallest value. This indicates that for user U9, who is a resource allocation target user, the importance of power consumption requirements is relatively high and the importance of communication bandwidth requirements is relatively low.
[0032] 2-7. Calculation of the fulfillment rate of the required performance of the resource allocation target user When determining the resources to be allocated to the resource allocation target user, the extent to which the communication performance of each link fulfills the required performance of the resource allocation target user is taken into consideration. According to the resource allocation method of this embodiment, the fulfillment rate of the required performance of the resource allocation target user is calculated for each link and for each communication performance.
[0033] 8 is a diagram illustrating the calculation of the fulfillment rate of the required performance of a user to which resources are allocated. In calculating the fulfillment rate, for elements such as communication bandwidth, where a larger numerical value indicates higher performance, the ratio of the index value to the required value is calculated as the fulfillment rate. For example, in the case of link L1, the communication bandwidth is 500 Mbps and the required communication bandwidth is 5 Mbps, so the communication bandwidth fulfillment rate is 500 / 5. On the other hand, for elements such as delay and power consumption, where a smaller numerical value indicates higher performance, the ratio of the required value to the index value is calculated as the fulfillment rate. For example, in the case of link L1, the power consumption is 75 mWh and the allowable power consumption is 10 mWh, so the power consumption fulfillment rate is 10 / 75.
[0034] The meaning of the numerical values of the fulfillment rates shown in Fig. 8 can be explained using Fig. 9. A fulfillment rate value of 1 for a certain communication performance means that the requirement for that communication performance of the resource-allocated user is 100% met, i.e., there is no excess or deficiency in satisfaction. Furthermore, a fulfillment rate value greater than 1 for a certain communication performance means that the requirement for that communication performance of the resource-allocated user is met more than 100%, i.e., the satisfaction level is excessive. Furthermore, a fulfillment rate value less than 1 for a certain communication performance means that the requirement for that communication performance of the resource-allocated user is not met 100%, i.e., the satisfaction level is insufficient.
[0035] 2-8. Calculation of Excess / Shortage of Communication Performance with Respect to Required Performance of Resource Allocation Target User Next, according to the resource allocation method of this embodiment, the excess / shortage of communication performance with respect to required performance of resource allocation target user is calculated based on the fulfillment rate calculated by the above method. The excess / shortage is calculated for each link and for each communication performance.
[0036] Fig. 10 is a diagram illustrating the calculation of the degree of excess or deficiency of communication performance relative to the required performance of a user to which resources are allocated. In Fig. 10, the table on the left shows the fulfillment rate calculated for each link and for each communication performance, and the table on the right shows the degree of excess or deficiency calculated for each link and for each communication performance. As can be seen from these tables, the degree of excess or deficiency is calculated by subtracting the fulfillment rate from 1, which means that there is no excess or deficiency in user satisfaction. Here, a value greater than 0 indicates that the communication performance exceeds the required performance, and a value greater than 0 indicates that the communication performance is insufficient compared to the required performance. Furthermore, the magnitude of each absolute value indicates the degree of excess or deficiency.
[0037] 2-11. Calculation of deviation of communication performance from required performance of resource allocation target user Next, according to the resource allocation method of this embodiment, the deviation of communication performance from required performance of resource allocation target user is calculated for each link. Then, the link with the smallest deviation, i.e., the link having communication performance that matches the required performance of the resource allocation target user, is allocated to the resource allocation target user.
[0038] 11 is a diagram illustrating the calculation of the deviation of link performance from the required performance of a user to which resources are allocated. The deviation is calculated using the excess / deficiency for each link and each communication performance, and a weight index for each required performance of the user to which resources are allocated. The deviation is then corrected for each communication performance using the weight index. Specifically, if the excess / deficiency is a positive value, the excess / deficiency is corrected by dividing the excess / deficiency by the weight index. On the other hand, if the excess / deficiency is a negative value, the excess / deficiency is corrected by multiplying the excess / deficiency by the weight index. The deviation is then calculated as the sum of the absolute values of the corrected excess / deficiency.
[0039] As described above, by changing the method of correction using the weight index depending on whether the excess or deficiency is a positive value or a negative value, it is possible to evaluate the degree of deviation according to the weight of the requirement for each communication performance. Specifically, when the weight index is greater than 1, a positive excess or deficiency is corrected to a smaller value, and a negative excess or deficiency is corrected to a larger value. This means that when the weight of a requirement is high, a shortage of communication performance relative to the required performance is given more importance than an excess. In other words, for important requirements, emphasis is placed on eliminating the shortage even if it means turning a blind eye to a slight excess. On the other hand, when the weight index is less than 1, a positive excess or deficiency is corrected to a larger value, and a negative excess or deficiency is corrected to a smaller value. This means that when the weight of a requirement is low, an excess of communication performance relative to the required performance is given more importance than an insufficiency. In other words, emphasis is placed on avoiding an excess for less important requirements.
[0040] After the deviations for all links are calculated, the magnitudes of the deviations between the links are compared. The resource of the link with the smallest deviation is then determined as the resource to be allocated to the target user for resource allocation. In the example shown in Figure 11, the deviation of link L9 is the smallest among all links, so the resource of link L9 is allocated to user U9, who is the target user for resource allocation.
[0041] As described above, the resource allocation method according to this embodiment selects links that enable resource allocation that prioritizes insufficient performance for requests with high importance and prioritizes excess performance for requests with low importance. This prevents links with excessive performance from being allocated to unimportant requests, while enabling the selection of appropriate links that satisfy important requests as much as possible. As a result, it is possible to appropriately satisfy each user's request while suppressing the concentration of load on a specific link. Furthermore, the resource allocation method according to this embodiment can allocate such resources with a smaller amount of calculation than the brute force method.
[0042] 2-12. Updating the Link Performance Table When resources are allocated to a resource allocation target user, the link performance table (Table 1 shown in FIG. 2) is updated based on the results. FIG. 12 is a diagram explaining the updating of the link performance table. The requested communication bandwidth of user U9, who is the resource allocation target user, is 5 Mbps, while the communication bandwidth of link L9 allocated to user U9, i.e., the resource, is 5 Mbps. Therefore, the entire communication bandwidth of link L9 is allocated to user U9, and the remaining communication bandwidth of link L9 after allocation to user U9 becomes zero.
[0043] The next user to be allocated resources is user U according to the order shown in FIG. 10 For user U9, resource allocation is determined based on the updated link performance table shown on the right side of Fig. 12. Note that in the example shown in Fig. 12, the communication bandwidth required by user U9 and the communication bandwidth of link L9 coincidentally match, but the communication bandwidth as a resource may be greater or less than the required communication bandwidth. In the former case, the remaining communication bandwidth can be allocated as a resource to another user. In the latter case, for example, if the communication bandwidth of link L9 is insufficient for the communication bandwidth required by user U9, the communication bandwidth of link L8, which has the next smallest deviation degree after link L9, is allocated to user U9.
[0044] 2-13. How to deal with resource shortages Using the method described above, resources are allocated sequentially to users, starting with those with the smallest variance values. However, depending on the relationship between the number of links and communication performance and the number of users and their required communication performance, it may happen that resources cannot be allocated to some users due to a resource shortage.
[0045] In the resource allocation method according to this embodiment, when there is a user to which resources have not been allocated (hereinafter referred to as a "resource-unallocated user"), information about the resource shortage is output. Specifically, information about the communication performance required for the link by the resource-unallocated user is output. By receiving such information, the network provider can quantitatively evaluate the resource shortage elements and, based on the evaluation results, design the network to satisfy the requirements of all users.
[0046] Furthermore, the resource allocation method according to this embodiment also outputs information on the weight index for each performance requirement of users to whom resources have not yet been allocated. By receiving such information, the network provider can identify users who should be given priority or resources that should be provided with priority, and reflect the results of this identification in improving the network.
[0047] 2-14. Calculation of deviation of communication performance from required performance of resource allocation target user The resource allocation method according to this embodiment described with reference to Figures 2 to 12 can be expressed as a generalized flow diagram. Figure 13 is a flow diagram showing the flow of the resource allocation method according to this embodiment.
[0048] The flowchart shows a pre-processing in step S100, a repeating process in step S110, and a post-processing in step S120. Of these, the pre-processing in step S100 is a process that is optionally executed prior to the implementation of the resource allocation method according to this embodiment. In the pre-processing, resources of a predetermined link are allocated to a predetermined user.
[0049] The repetitive process of step S110 and the post-processing of step S120 are processes executed by implementing the resource allocation method according to this embodiment. The repetitive process of step S110 includes the processes of steps S111 to S116. The processes of steps S111 to S116 are repeatedly executed until resource allocation to all users is completed, which is the end condition of the repetition, or until resources are depleted.
[0050] Step S111 is a step of scoring and normalizing user requests. In this step, the request values from each user for each communication performance of the link are scored. Furthermore, in this step, the scored request values are normalized between users to obtain normalized scores.
[0051] Step S112 is a step for calculating the variance value of user requests and selecting users to whom links are to be allocated. In this step, the variance value between the normalized scores of each communication performance is calculated for each user. In addition, in this step, users to whom links are to be allocated are selected in accordance with an allocation order determined in ascending order of variance value.
[0052] In step S113, a weight index of the user request is calculated. In this step, the average value of the normalized score of each communication performance is calculated for the selected users to whom a link is to be allocated, and the ratio of the normalized score to the average value for each communication performance is calculated as a weight index.
[0053] Step S114 is a step of calculating the degree of excess or deficiency with respect to the user request for each link. In this step, the degree of excess or deficiency of the index value of the communication performance with respect to the request value from the resource allocation target user is calculated for each link and for each communication performance.
[0054] Step S115 is a step of allocating resources. In this step, the degree of deviation of communication performance from the required performance of the resource allocation target user is calculated based on the weighting coefficient calculated in step S113 and the degree of excess or shortage calculated in step S114. Then, in this step, resources of the links selected in ascending order of deviation are allocated to the resource allocation target user.
[0055] Step S116 is a step for updating the link performance table, which is updated based on the resource allocation results performed in step S115.
[0056] Step S120 is a post-processing step that is executed when the condition of step S110 is satisfied. In this step, when a resource shortage occurs, information about the resource shortage and information about the weight index for each required performance of a user to whom resources have not been allocated are output.
[0057] 3. Configuration of the Resource Allocation Device The resource allocation method described above is executed by a resource allocation device. Fig. 10 is a block diagram showing the configuration of a resource allocation device according to this embodiment. The resource allocation device 10 shown in Fig. 14 is connected to a network 40. A large number of wireless terminals 20 are connected to the network 40 via a large number of wireless base stations 30. The wireless terminals 20 include heterogeneous terminals, and the wireless base stations 30 include base stations of different wireless systems.
[0058] The resource allocation device 10 is a computer or computer system including a communication interface 11, a processor 12, a program memory 13, and an information memory 15. There may be a plurality of processors 12, a plurality of program memories 13, and a plurality of information memories 15. The communication interface 11 is an interface for connecting to a network 40. The communication interface 11 may be a wireless interface or a wired interface.
[0059] Processor 12 may be a CPU, RISC, DSP, FPGA, ASIC, PLD, or other processing unit, or may be a combination of two or more thereof, or may be a dedicated processor operating in accordance with the resource allocation methods described above. Processor 12 may also be communicatively coupled to or may include program memory 13 and information memory 15.
[0060] The program memory 13 stores a plurality of instructions 14 executable by the processor 12. A program consisting of the plurality of instructions 14 may be stored in a computer-readable non-transitory storage medium or provided via a network. At least one of the plurality of instructions 14 is configured to cause the processor 12 to execute the resource allocation method described above. The information memory 15 stores various information such as a user request database 16 and a link performance database 17. In the user request database 16, information related to the performance requirements of each user is managed, for example, in the form of a required performance table. In the link performance database 17, information related to the communication performance of each link is managed, for example, in the form of a link performance table.
[0061] 4. Others The resource allocation method of the present disclosure is not limited to the above-described embodiment. For example, scoring and normalization of numerical values may be performed on the index values of each communication performance listed in the link performance table. By scoring and normalizing the index values, it becomes possible to compare and uniformly evaluate the communication performance of each index value between different links or between the same link.
[0062] 10 Resource allocation device 12 Processor 13 Program memory 14 Instructions 15 Information memory 16 User request database 17 Link performance database
Claims
1. A resource allocation method executed by a computer, comprising: obtaining, for each of a plurality of users, a request value for each of a plurality of types of communication performance requested for a link; scoring the request values from the plurality of users for each of the plurality of types of communication performance; normalizing the scored request value for each of the plurality of types of communication performance among the plurality of users; calculating a variance value between the scored and normalized request values for each of the plurality of types of communication performance among the plurality of users; determining a resource allocation target user among the plurality of users in order of smallest variance value, and allocating resources of a link selected from one or more available links to the resource allocation target user.
2. A resource allocation method as described in claim 1, further comprising: acquiring index values for each of the multiple types of communication performance for the one or more available links; calculating, for each of the one or more available links, the degree of excess or deficiency of the index value for each of the multiple types of communication performance relative to the requested value from the resource allocation target user for the one or more available links; calculating, for each of the one or more available links, the degree of deviation from the communication performance requested by the resource allocation target user for the link based on the degree of excess or deficiency of each of the multiple types of communication performance; and selecting, from the one or more available links, a link for allocating resources to the resource allocation target user in order of the smallest degree of deviation.
3. A resource allocation method as claimed in claim 2, further comprising: calculating, for the user to which resource is to be allocated, an average value of the scored and normalized required values for each of the multiple types of communication performance; calculating, for the user to which resource is to be allocated, a ratio of the scored and normalized required values for each of the multiple types of communication performance to the average value as a weighting index for each of the multiple types of communication performance; and correcting the degree of surplus or deficiency of each of the multiple types of communication performance used in calculating the deviation by dividing the degree of surplus or deficiency by the weighting index when the degree of surplus or deficiency is a positive value.
4. A resource allocation method as claimed in claim 2, further comprising: calculating, for the resource allocation target user, an average value of the scored and normalized requirement values for each of the multiple types of communication performance; calculating, for the resource allocation target user, a ratio of the scored and normalized requirement value for each of the multiple types of communication performance to the average value as a weighting index of the requirement for each of the multiple types of communication performance; and correcting the degree of surplus or deficiency of each of the multiple types of communication performance used in calculating the deviation by multiplying the degree of surplus or deficiency by the weighting index when the degree of surplus or deficiency is a negative value.
5. A resource allocation method according to any one of claims 1 to 4, further comprising: when resources cannot be allocated to some of the plurality of users due to a shortage of resources, outputting information relating to the shortage of resources.
6. A resource allocation method according to claim 3 or 4, further comprising: when resources cannot be allocated to some of the plurality of users due to a shortage of resources, outputting information on the shortage of resources and information on the weighting index of the requirements for each of the plurality of types of communication performance of the users to whom resources have not been allocated.
7. A resource allocation device comprising: at least one processor; at least one memory coupled to the at least one processor or built into the at least one processor; and a plurality of instructions stored in the at least one memory and executable by the at least one processor, wherein the plurality of instructions are configured to cause the at least one processor to execute the following: obtain, for each of a plurality of users, a requested value for each of a plurality of types of communication performance requested for a link; score the requested values from the plurality of users for each of the plurality of types of communication performance; normalize the scored requested value for each of the plurality of types of communication performance among the plurality of users; calculate a variance value between the scored and normalized requested values for each of the plurality of types of communication performance among the plurality of users; determine a resource allocation target user according to the smallest variance value among the plurality of users; and allocate resources of a link selected from one or more available links to the resource allocation target user.
8. A program including a plurality of instructions executable by at least one processor, the plurality of instructions being configured to cause the at least one processor to execute the following: obtain, for each of a plurality of users, a requested value for each of a plurality of types of communication performance requested for a link; score the requested values from the plurality of users for each of the plurality of types of communication performance; normalize the scored requested value for each of the plurality of types of communication performance among the plurality of users; calculate a variance value between the scored and normalized requested values for each of the plurality of types of communication performance among the plurality of users; determine a resource allocation target user in order of smallest variance value among the plurality of users; and allocate resources of a link selected from one or more available links to the resource allocation target user.
Citation Information
Patent Citations
Transmission band assigning device
WO2004098226A1
