CN102932308A - Scheduling method and scheduling system under mixed business scenes in orthogonal frequency division multiplexing (OFDM) relay system - Google Patents

Abstract

The invention discloses a resource scheduling system and method based on user quality experience in a mixed service scenario in an OFDM relay system. Involves OFDM relay system. According to the requirements of real-time service (rtPS) on average delay and non-real-time service (nrtPS) on average rate, the invention establishes a unified user satisfaction model for users to reflect the current satisfaction degree of users. The overall user satisfaction of the system is improved by allocating sub-channels to the users whose satisfaction increases the most on the sub-channels. The method of the present invention can solve the problem of waste of resources caused by only considering the system point of view and ignoring user experience in the current mobile communication system, and realizes a resource scheduling algorithm based on user QoE.

Description

A scheduling method and scheduling system for mixed service scenarios in an OFDM relay system

technical field

The invention relates to the technical field of mobile communication, in particular to a technical solution for resource scheduling of an OFDM relay system to maximize the overall satisfaction of system users.

Background technique

In the next-generation wireless communication system, the goal is to realize high-speed data transmission and seamless global coverage. In the case of scarce and expensive spectrum resources, this goal can only be achieved by improving spectrum utilization through frequency reuse. Orthogonal frequency division multiple access technology (OFDMA) has become the focus of people's attention because of its better ability to resist multipath effects. In addition to better reducing the frequency selective fading caused by multipath effects, the OFDM network divides the frequency band into multiple subcarriers, and each subcarrier is independent of each other, making wireless resource allocation more flexible and improving system spectrum utilization. In order to achieve seamless coverage in traditional networks, the number of base stations (BSs) must be increased, which will result in increased costs. The relay (Relay) network can reduce network costs while expanding network coverage, and has better flexibility. FIG. 1 is a schematic diagram of a topology structure of an OFDMA relay system. In Figure 1: BS is a cell base station, RS is a relay, and UE is a user. BS and RS respectively set up buffer areas for users to store user data.

The wireless resource scheduling algorithm plays a vital role in improving the performance of wireless networks, especially in OFDM relay systems, where multiple subcarriers and two-hop links make resource scheduling more flexible. A reasonable resource scheduling algorithm can greatly improve system capacity and ensure user fairness.

At present, most scheduling algorithms for OFDM relay systems are based on channel state information, queue length or user QoS. These resource scheduling algorithms start from the system point of view and aim at maximizing system performance, without considering the user's subjective experience quality, and cannot perform reasonable radio resource allocation according to the urgency of user needs. In this case, the concept of QoE (Quality of Experience) is introduced. QoE can be understood as a service evaluation method based on user experience or perception. It is affected by services, users and the environment, and directly reflects the user's satisfaction with the obtained resources or services through an approximate quantitative method. That is to say, user satisfaction is not only objectively affected by the amount of resources allocated by the system, but also affected by people's subjective consciousness. resources, the increase in user satisfaction will become smaller and smaller. If traditional resource scheduling algorithms are used, resource allocation will be unfair.

At present, the theory of QoE is still immature, and different researchers try to make quantitative evaluation of QoE from different angles, so many different quantitative methods and evaluation methods have emerged. At the same time, different users have different service requirements. For example, real-time services have higher requirements on delay and relatively lower average rate requirements, while non-real-time services or BE services have lower requirements on delay and higher average rate requirements. The same resources are allocated to different types of users, and the improvement of user satisfaction will not be the same. This also brings great challenges to scheduling and allocating wireless resources in mixed service scenarios.

Contents of the invention

The present invention aims at the above-mentioned technical problems existing in the prior art, and aims at establishing a unified user satisfaction model in mixed service scenarios in an OFDM relay system, and allocating wireless resources according to user satisfaction, thereby improving overall system efficiency.

The technical solution adopted by the present invention to solve the technical problem is: set up a buffer area in the base station (BS) and relay (RS) to store user data, set up a channel state information collection module and a queue state information collection module on the BS side, and collect two The channel state information of each link in the hop link and the queue state information of the BS and RS end buffers, and estimate the user satisfaction. Sub-channels are allocated according to user satisfaction.

The user communicates with the base station through the relay, and the relay sets a buffer area to store the data sent by the base station, and forwards the data in the next sub-slot. The base station collects channel state information between links and queue state information in the buffer area, estimates user satisfaction, and allocates resources according to user satisfaction. The specific scheduling method is as follows:

，计算用户当前时隙的平均时延

和平均速率

；The channel state information extraction unit and the queue state information extraction unit respectively extract the delay required by the user QoS ,rate and the maximum delay that real-time users can tolerate

, calculate the average delay of the user's current time slot

and the average rate

;

According to the different requirements of real-time services and non-real-time services on delay and average rate, the user satisfaction estimation unit normalizes the average delay or average rate of users, and establishes the user satisfaction equation:

${\mathrm{MOS}}_m\left(t\right)={\log }_2({\∂}_m\&Center\; Dot;a\·\frac{{\mathrm{D.}}_{\max }^m-\stackrel{\‾}{{\mathrm{D.}}_m\left(t\right)}}{{\mathrm{D.}}_{\mathrm{req}}^m}+(1-{\∂}_m)\&Center\; Dot;a\&Center\; Dot;\frac{\stackrel{\‾}{R_m\left(t\right)}}{R_{\mathrm{req}}^m})$ ;

Taking the maximization of user satisfaction as the optimization goal, and taking the lower limit of the queue length of the user buffer area and the upper limit of user satisfaction as constraints, the objective function is established:

Determine current user satisfaction;

The feedback information receiving unit receives the channel state information and the RS terminal queue length information in the two-hop link in the feedback channel;

The scheduling allocation unit allocates each sub-channel according to the current satisfaction degree of the user and the channel state information, and schedules the user resources of the relay link.

Resource scheduling is performed according to the data queue lengths of the base station and the relay end.

The feedback information receiving unit obtains the channel status information in the first hop and the queue status information of the RS terminal from the feedback channel, and obtains the subchannel set as X by initialization; selects the subchannel n with the best channel condition from the available subchannels; selects the queue length of the BS terminal The user m with the largest queue length gap with the RS end; assign sub-channel n to user m, update the data queue information of the user BS end and RS end, and update the sub-channel set to X'=X\n; until the sub-channel set is empty.

Allocate each sub-channel to the user whose satisfaction has been improved the most on this sub-channel, so as to improve the user satisfaction of the system.

计算子信道n分配给该用户后的用户满意度增长幅度，对非实时业务用户，计算用户平均速率，并根据公式：

计算子信道n分配给该用户后的用户满意度增长幅度，其中，MOS_m(t)为用户当前满意度，

为把子信道n分配给该用户后的用户满意度；选择ΔMOS(t)最大的用户m，将子信道n分配给用户m，更新用户m的队列状态信息以及平均速率；直至子信道集合为空。Obtain the channel state information in the second hop from the feedback channel, initialize the subchannel set as X, and the user set as U; choose a subchannel n from the available subchannels, and assign the subchannels to users; for real-time service users, calculate the user The average delay, and according to the formula:

Calculate the user satisfaction growth rate after the sub-channel n is allocated to the user. For non-real-time service users, calculate the user average rate, and according to the formula:

Calculate the increase in user satisfaction after sub-channel n is allocated to the user, where MOS _m (t) is the user's current satisfaction,

In order to satisfy the user satisfaction after assigning sub-channel n to the user; select user m with the largest ΔMOS(t), assign sub-channel n to user m, and update the queue status information and average rate of user m; until the set of sub-channels is null.

、速率

和实时用户能够容忍的最大时延

，计算用户当前时隙的平均时延和平均速率；用户满意度估计单元对用户的平均时延或平均速率进行归一化处理，建立用户满意度方程：The present invention also proposes a resource scheduling system in a mixed service scenario in an OFDM relay system. The system includes: a feedback information receiving unit, a channel state information extraction unit, a queue state information extraction unit, a user satisfaction estimation unit, and a resource scheduling unit. The channel state information extraction unit and the queue state information extraction unit respectively extract the delay required by the user QoS

,rate

and the maximum delay that real-time users can tolerate

, calculate the average delay of the user's current time slot and the average rate ; The user satisfaction estimation unit normalizes the average delay or average rate of the user, and establishes the user satisfaction equation:

；以最大化用户满意度为优化目标，且以用户缓存区队列长度下限和用户满意度上限为限制条件，建立目标函数，确定用户当前满意度；反馈信息接收单元接收反馈信道中两跳链路中的信道状态信息和RS端队列长度信息；资源调度单元根据用户当前满意度以及信道状态信息分配各个子信道，对中继链路用户资源进行调度。

; Taking maximization of user satisfaction as the optimization goal, and taking the lower limit of the queue length of the user buffer area and the upper limit of user satisfaction as constraints, an objective function is established to determine the current satisfaction of the user; the feedback information receiving unit receives the two-hop link in the feedback channel The channel state information and the RS terminal queue length information; the resource scheduling unit allocates each sub-channel according to the user's current satisfaction degree and channel state information, and schedules the relay link user resources.

In the mixed service scenario, the present invention establishes a unified user satisfaction model for different types of users, and adopts a resource scheduling method based on user satisfaction and aiming at maximizing the overall satisfaction of system users, so as to realize wireless resource scheduling. Reasonable distribution.

Description of drawings

Figure 1 is a schematic diagram of the topology of the cellular mobile communication system of the relay station

Fig. 2 is a schematic diagram of the system structure of the scheduling mechanism in the specific embodiment of the present invention

Fig. 3 is a flow chart of resource scheduling for the first hop of a relay link in a specific embodiment of the present invention

Fig. 4 is a flow chart of the resource scheduling of the second hop of the relay link in the specific embodiment of the present invention

Detailed ways

In order to describe the advantages of the technical solutions of the present invention more clearly, the specific implementation manners of the present invention will be further described in detail below with reference to the accompanying drawings.

Figure 1 is the application scenario of the present invention. As shown in Figure 1, the system includes a base station (BS), a relay (RS) and a user (UE). It is assumed that due to distance reasons, all UEs can only communicate with the BS through the RS and cannot communicate with the BS. direct communication. Both the BS end and the RS end are provided with buffer intervals to store user data. The system adopts the OFDM access method, divides the bandwidth into multiple independent orthogonal sub-channels in the frequency domain, and divides a time slot into two sub-slots in the time domain for the BS-RS link and the RS-UE chain respectively. Data transmission on the road.

Fig. 2 is a schematic diagram of a BS-side information collection module and a scheduling module. It includes: a feedback information receiving unit 201 , a temporary data storage unit 202 , a channel state information extraction unit 203 , a queue state information extraction unit 204 , a user satisfaction estimation unit 205 and a resource scheduling unit 206 .

The feedback information receiving unit is responsible for receiving the information fed back in the feedback channel, including the channel state information in the two-hop link and the queue length information at the RS end.

The temporary data storage unit is used for accessing temporary data, such as the average rate of users, the average length of queues at the BS end and the RS end, etc.

和实时用户能够容忍的最大时延，获取用户当前时隙的平均时延

和平均速率

，

可以通过排队论中的Little定理推出，

可以通过设定时间窗口求得。The channel state information extraction unit and the queue state information extraction unit respectively extract channel state information and queue state information. Including: delay required by user QoS ,rate

and the maximum delay that real-time users can tolerate , to obtain the average delay of the user's current time slot

and the average rate

,

It can be deduced by Little's theorem in queuing theory,

It can be obtained by setting the time window.

The user satisfaction estimation unit is responsible for estimating the current user satisfaction. The specific method is to calculate user satisfaction according to channel state information, queue state information, temporary data, and QoS requirements of different types of users.

The scheduling allocation unit is responsible for allocating each sub-channel according to the user's current satisfaction degree and channel state information.

The user communicates with the base station through the relay, and the relay sets a buffer area to store the data sent by the base station, and forwards the data in the next sub-slot. The base station collects channel state information between links and queue state information in the buffer area, estimates user satisfaction, and allocates resources according to user satisfaction.

The system of the present invention includes real-time users and non-real-time users, and they have different requirements on QoS. Real-time services have higher requirements on time delay, while non-real-time users have higher requirements on average speed. However, from the perspective of users, real-time services and non-real-time services have the same characteristics, that is, after users obtain resources to a certain extent, and then continue to allocate resources to users, the increase in user satisfaction will gradually decrease. According to this characteristic, a unified user satisfaction model is established. The user satisfaction estimation unit normalizes the average delay or average rate of users to establish a unified satisfaction equation:

${\mathrm{<span\; class="notranslate">\; MOS</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; log</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>{<span\; class="notranslate">\; \&PartialD;</span>}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; \&CenterDot;</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>\frac{{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>}}{{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; req</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{<span\; class="notranslate">\; \&PartialD;</span>}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>\frac{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>}}{{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; req</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}}}<span\; class="notranslate">\; )</span>$

代表用户类型，

表示用户为实时用户，

表示用户为非实时用户。Among them, MOS _m (t) represents the satisfaction degree of user m at time t. According to the widely used subjective evaluation standard, the "mean estimated score" (MOS) is between 1 and 4.5, and the MOS is continuous. a is a constant, and the value of a guarantees that the value of MOS _m (t) is any value in the closed interval [1,4.5].

represents the user type,

Indicates that the user is a real-time user,

Indicates that the user is a non-real-time user.

Taking the maximization of system user satisfaction as the optimization goal, and taking the lower limit of the queue length of the user buffer area and the upper limit of user satisfaction as constraints, an objective function is established to estimate the current user satisfaction:

， I(t)_m能够反映用户m获得的资源，在多业务环境下，对应实时业务和非实时业务分别和用户平均延时和用户平均速率有关。

表示用户QoS要求的最小速率或延时，

代表用户类型，当用户为实时业务时

，当用户为非实时用户时。a为常数，能够保证用户满意度的值在一定范围内。当I(t)_m大于或等于时用户处于满意状态，否则用户不满意。

 表示在t时刻用户m在中继链路第二跳上获得的速率，

表示在t时刻中继缓存中用户m的队列长度，MOS_m(t)表示在t时刻用户m的满意度, T₀为时隙长度,M为用户数。限制条件可以保证当RS端用户队列没有数据或者用户满意度达到最高时，不再为用户分配过多资源，以避免分配过多的资源。in

, I(t) _m can reflect the resources obtained by user m. In a multi-service environment, the corresponding real-time services and non-real-time services are related to the average user delay and user average rate respectively.

Indicates the minimum rate or delay required by the user QoS,

Represents the user type, when the user is a real-time business

, when the user is a non-real-time user . a is a constant, which can ensure that the value of user satisfaction is within a certain range. When I(t) _m is greater than or equal to When the user is satisfied, otherwise the user is not satisfied.

Indicates the rate obtained by user m on the second hop of the relay link at time t,

Indicates the queue length of user m in the relay buffer at time t, MOS _m (t) indicates the satisfaction degree of user m at time t, T ₀ is the length of time slot, and M is the number of users. The restrictive conditions can ensure that when the user queue at the RS end has no data or the user satisfaction reaches the highest level, no more resources will be allocated to the user, so as to avoid allocating too many resources.

Since the objective function is a non-convex mixed integer optimization problem, it is difficult to directly find the optimal solution. In this embodiment, a scheduling algorithm based on the improvement degree of satisfaction degree can be adopted. Other methods known to those skilled in the art can also be used to solve.

The feedback information receiving unit receives information fed back in the feedback channel, including channel state information in the two-hop link and queue length information at the RS end.

The scheduling allocation unit allocates each sub-channel according to the current satisfaction degree of the user and the channel state information, and schedules the user resources of the relay link.

In order to ensure the balance between the data in the buffer area of the base station and the data in the buffer area of the relay end, resource scheduling is performed according to the data queue lengths of the base station and the relay end. The specific implementation method is as follows:

As shown in Figure 3, in this embodiment, the first hop user resource scheduling flowchart of the relay link, the scheduling process includes the following steps:

Step 301: The feedback information receiving unit obtains the channel state information in the first hop and the queue state information of the RS terminal from the feedback channel, and obtains the subchannel set as X by initialization;

Step 302: Select the sub-channel n with the best channel condition from the available sub-channels;

确定。其中，

和

分别表示BS端和RS端的数据队列长度，q为常数，M为系统用户数。Step 303: Select the user m with the largest difference between the queue length at the BS end and the queue length at the RS end, and obtain user m according to the formula:

Sure. in,

and

represent the data queue lengths of the BS end and the RS end respectively, q is a constant, and M is the number of system users.

Step 304: assign subchannel n to user m, update the data queue information of user BS and RS, and update the set of subchannels to X'=X\n.

Step 305: Determine whether the sub-channel set is empty, if not, continue to execute step 304 to allocate sub-channels until the sub-channel set is empty.

FIG. 4 is a flowchart of user resource scheduling in the second hop of the relay link in this embodiment. According to the optimization objective, a sub-optimal scheduling algorithm is used to assign each sub-channel to the user with the highest satisfaction improvement on the sub-channel, so as to improve the system user satisfaction. The specific implementation method is as follows:

Step 401: Obtain the channel state information in the second hop from the feedback channel, initialize the subchannel set to X, and the user set to U;

Step 402: Select a sub-channel n from the available sub-channels;

判断其用户类型，如为实时业务用户执行步骤404，否则执行步骤405；Step 403: For each user in the user set, according to

Judging its user type, as performing step 404 for a real-time service user, otherwise performing step 405;

。根据公式：

计算子信道n分配给该用户后的用户满意度增长幅度。其中MOS_m(t)为用户当前满意度；Step 404: Calculate the average user delay after the sub-channel n is allocated to the user (can be calculated according to Little's theorem), and calculate the corresponding user satisfaction

. According to the formula:

Calculate the increase in user satisfaction after sub-channel n is allocated to the user. Among them, MOS _m (t) is the user's current satisfaction;

计算子信道n分配给该用户后的用户满意度增长幅度。其中MOS_m(t)为用户当前满意度；Step 405: Calculate the average rate of the user after the sub-channel n is allocated to the user, and calculate the corresponding user satisfaction . According to the formula:

Calculate the increase in user satisfaction after sub-channel n is allocated to the user. Among them, MOS _m (t) is the user's current satisfaction;

Step 406, select user m with the largest ΔMOS(t), assign subchannel n to user m, update queue status information and average rate of user m;

Step 407: Determine whether the satisfaction degree of user m is 4.5 or the queue at the RS end is empty. If satisfied, execute step 408, otherwise execute 409;

Step 408: Update the user set so that U'=U\m;

Step 409: Determine whether the sub-channel set is empty, if not, continue to allocate sub-channels according to the above method, otherwise end.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention.

Claims (9)

1. A method for resource scheduling under a hybrid service scenario in an OFDM relay system, characterized in that it comprises the steps of: a channel state information extracting unit and a queue state information extracting unit respectively extract the time delay required by user QoS

,rate

and the maximum delay that real-time users can tolerate

, calculate the average delay of the user's current time slot

and the average rate

; The user satisfaction estimation unit normalizes the average delay or average rate of the user, and establishes the user satisfaction equation:

; Taking maximization of user satisfaction as the optimization goal, and taking the lower limit of the queue length of the user buffer area and the upper limit of user satisfaction as constraints, an objective function is established to determine the current satisfaction of the user; the feedback information receiving unit receives the two-hop link in the feedback channel The channel state information in and the queue length information at the RS end; the scheduling allocation unit allocates each sub-channel according to the user’s current satisfaction degree and channel state information, and schedules the user resources of the relay link, where MOS _m (t) means that the user m is in Satisfaction degree at time t, a is a constant,

Represents the user type, when the user is a real-time business

, when the user is a non-real-time user

. 2. The method for resource scheduling in a mixed service scenario according to claim 1, wherein the objective function established is:

, where I(t) _m represents the resources obtained by user m, Indicates the minimum rate or delay required by the user QoS,

Indicates the rate obtained by user m on the second hop of the relay link at time t,

Indicates the queue length of user m in the relay buffer at time t, MOSm(t) indicates the satisfaction degree of user m at time t, and T ₀ is the length of time slot. 3. The resource scheduling method in a mixed service scenario according to claim 1, wherein the scheduling of the relay link user resources is specifically: performing resource scheduling according to the length of the data queue of the base station and the relay end, and each A sub-channel is assigned to the user whose satisfaction has been improved the most on that sub-channel. 4. The resource scheduling method in a mixed service scenario according to claim 3, wherein the resource scheduling is performed according to the data queue lengths of the base station and the relay end, specifically comprising: the feedback information receiving unit acquires the channel in the first hop from the feedback channel State information and RS-side queue state information, the sub-channel set obtained by initialization is X; the sub-channel n with the best channel condition is selected from the available sub-channels; the user m with the largest difference between the BS-side queue length and the RS-side queue length is selected; Channel n is allocated to user m, and the data queue information of user BS and RS is updated, and the sub-channel set is updated to X'=X\n; until the sub-channel set is empty. 5. The resource scheduling method in a mixed service scenario according to claim 3, wherein assigning each sub-channel to the user whose satisfaction degree is most improved on the sub-channel specifically comprises: obtaining the second hop channel from the feedback channel State information, initialize the set of sub-channels as X, and the set of users as U; choose a sub-channel n from the available sub-channels, and assign the sub-channels to users; for real-time service users, calculate the average delay of users, and for non-real-time service users , calculate the user average rate, and according to the formula: Calculate the increase in user satisfaction after sub-channel n is allocated to the user, where MOS _m (t) is the user's current satisfaction,

In order to satisfy the user satisfaction after assigning sub-channel n to the user; select user m with the largest ΔMOS(t), assign sub-channel n to user m, and update the queue status information and average rate of user m; until the set of sub-channels is null. 6. A resource scheduling system under a mixed service scenario in an OFDM relay system, characterized in that the system includes: a feedback information receiving unit, a channel state information extraction unit, a queue state information extraction unit, a user satisfaction estimation unit, and resource scheduling unit, the channel state information extraction unit and the queue state information extraction unit respectively extract the time delay required by the user QoS

,rate

and the maximum delay that real-time users can tolerate , calculate the average delay of the user's current time slot and the average rate

; The user satisfaction estimation unit normalizes the average delay or average rate of the user, and establishes the user satisfaction equation: ${\mathrm{MOS}}_m\left(t\right)={\log }_2({\&PartialD;}_m\&CenterDot;a\&CenterDot;\frac{{\mathrm{D.}}_{\max }^m-\stackrel{\&OverBar;}{{\mathrm{D.}}_m\left(t\right)}}{{\mathrm{D.}}_{\mathrm{req}}^m}+(1-{\&PartialD;}_m)\&Center\; Dot;a\&Center\; Dot;\frac{\stackrel{\&OverBar;}{R_m\left(t\right)}}{R_{\mathrm{req}}^m})$ ; Taking maximization of user satisfaction as the optimization goal, and taking the lower limit of the queue length of the user buffer area and the upper limit of user satisfaction as constraints, an objective function is established to determine the current satisfaction of the user; the feedback information receiving unit receives the two-hop link in the feedback channel The channel state information in and the queue length information at the RS end; the resource scheduling unit allocates each sub-channel according to the user’s current satisfaction degree and channel state information, and schedules the user resources of the relay link, where MOS _m (t) means that user m is in Satisfaction degree at time t, a is a constant,

Represents the user type, when the user is a real-time business

, when the user is a non-real-time user

. 7. system according to claim 6, is characterized in that, the objective function of described establishment is:

, where I(t) _m represents the resources obtained by user m,

Indicates the minimum rate or delay required by the user QoS,

Indicates the rate obtained by user m on the second hop of the relay link at time t,

Indicates the queue length of user m in the relay buffer at time t, MOS _m (t) indicates the satisfaction degree of user m at time t, and T ₀ is the time slot length. 8. The system according to claim 6, wherein the scheduling of the relay link user resources specifically comprises that the feedback information receiving unit obtains the channel state information in the first hop and the RS terminal queue state information from the feedback channel , the set of sub-channels obtained by initialization is X; select the sub-channel n with the best channel condition from the available sub-channels; select the user m with the largest difference between the queue length of the BS end and the queue length of the RS end; assign the sub-channel n to user m, and update The data queue information of the user BS end and the RS end, the sub-channel set is updated to X'=X\n; until the sub-channel set is empty. 9. The system according to claim 6, wherein the scheduling of user resources on the relay link specifically includes obtaining channel state information in the second hop from the feedback channel, initializing the subchannel set as X, and the user set is U; select a sub-channel n from the available sub-channels, and assign the sub-channels to users; for real-time service users, calculate the user average delay, and for non-real-time service users, calculate the user average rate, and according to the formula:

Calculate the increase in user satisfaction after sub-channel n is allocated to the user, where MOS _m (t) is the user's current satisfaction,

In order to satisfy the user satisfaction after assigning sub-channel n to the user; select user m with the largest ΔMOS(t), assign sub-channel n to user m, and update the queue status information and average rate of user m; until the set of sub-channels is null.

Images