KR100921492B1 - Method of scheduling packet in wireless data communication system - Google Patents

Abstract

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a packet scheduling method of a wireless data communication system. The present invention relates to a packet scheduling method in a multi-hop relay system for improving the yield and coverage of a system in a broadband mobile Internet access system such as WiBro (Wireless Broadband Internet). The present invention relates to a packet scheduling method capable of ensuring fairness. The packet scheduling method of the wireless data communication system of the present invention comprises the steps of: (a) reflecting the instantaneous data rate of each user in consideration of the multi-hop channel; (b) dynamically adjusting the boundary between the access area and the relay area by scheduling without fixing the size of the interval for the relay and the interval for the user connection in the system; And (c) determining a user priority in consideration of the allocation amount of each relay access area.

Description

Packet scheduling method of a wireless data communication system {Method of scheduling packet in wireless data communication system}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a packet scheduling method of a wireless data communication system. The present invention provides a data throughput in a multi-hop relay system for improving the yield and coverage of a broadband mobile Internet access system such as WiBro (Wireless Broadband Internet). In addition, the present invention relates to a packet scheduling method capable of ensuring fairness.

Recently, the next generation mobile communication system has developed into a broadband mobile Internet access system that can provide users with high speed data such as the Internet anytime, anywhere through a mobile communication terminal.

In order to service high-speed data such as the Internet through a mobile communication terminal, WiBro (WiBro) in Korea and IEEE 802.16e system standards have been developed and are currently in service.

In order to provide better service to users in such a broadband mobile Internet access system, a multi-hop relay system is being considered for the purpose of improving system yield and expanding coverage, and international standardization work is in progress. Multi-hop relay systems can improve coverage by introducing relays into existing cellular networks to provide services to subscribers in areas with poor reception strength from base stations (in cell edges, shadow regions, and large buildings). It is also possible to increase the yield by providing a higher channel environment to the subscribers that are already being served.

As a conventional technique, various scheduling methods for data communication with a plurality of terminals connected to a base station have been proposed. First, in case of a base station and terminals, a method of prioritizing system data throughput by providing a service to a terminal having the best channel state (MAX C / I), and ignoring the channel state of the terminal. It is widely known to give a round-robin communication opportunity to each other (Round Robin), and Proportional Fairness that considers fairness while considering channel conditions. There are several scheduling schemes of QoS (Quality of Service) as a scheduling scheme according to delay, jitter, minimum guaranteed data throughput, and the like.

As a problem of the scheduling method according to the prior art, looking at the multi-hop relay system model, it can be seen that there are scheduling considerations, unlike in the conventional cellular model-type wireless data communication system.

1 illustrates channel characteristics in a cellular model and a multi-hop relay model.

1 is a scheduling scheme between a plurality of terminals connected to one base station. As shown in the left side of FIG. 1, a base station and a terminal are directly (1-hop) in data communication with the terminal. It is based on scheduling in the connected system. On the other hand, as shown in the right side of FIG. 1, a multi-hop relay system may have a plurality of relays between a base station and a terminal. Consequently, in a multi-hop relay system, a base station and a terminal may be directly (1-hop) or multiple relays ( Data communication takes place over 2-hop). Therefore, the channel state of the base station and the terminal used in the prior art does not reflect the channels by the relay in the middle.

Therefore, in the conventional scheduling considering the general single hop on the left side of FIG. 1, although the transmission rate according to the channel state between the base station and the terminal is used as a parameter for determining the priority for resource allocation, the multi (multi) on the right side of FIG. Since the channel state may be different for each hop interval in the hop channel, the performance may vary depending on how the channel state of each hop is applied as a scheduling parameter in the scheduling in the multi-hop relay system. In this case, since a radio resource required to pass each hop may have an allocation area that is orthogonal with the frequency reuse area, the data rate between the base station and the terminal considering the radio resource consumption in view of the system should be considered.

2 shows an uplink / downlink configuration in a multihop relay model, and FIG. 3 shows a frame structure of a multihop relay system.

According to FIG. 2, the uplink (uplink) and downlink (downlink) intervals in the frame structure considered in the multi-hop relay system are the base station (BS) or the relay (RS) to the terminal (MS) or the terminal (MS) Is divided into an access zone, which is a section in which transmission is performed to a base station BS or a relay RS, and a relay zone, which is a section in which transmission is performed between a base station BS and a relay RS.

According to FIG. 3, the BS frame structure transmits the access area from the BS to the MS in the downlink, the BS to the RS in the relay area, the MS to the BS in the access area in the uplink, and the relay area. Transmit from RS to BS Meanwhile, according to FIG. 3, the RS frame structure transmits the access area from the RS to the MS in the downlink, operates in the receiver mode in the relay area, and transmits from the MS to RS in the access area in the uplink and in the relay area. Operate in transmitter mode.

Resource allocation in a multi-hop relay system may be divided into an orthogonal allocation and an overlap allocation scheme according to the presence or absence of frequency reuse in an access link.

Orthogonal allocation is a method in which BSs and RSs do not transmit data to the MS using the same subchannel at the same time, and overlap is a scheme in which BSs and RSs simultaneously transmit data to the MS. It is an allocation method that increases the frequency reuse efficiency by using a subchannel.

If the boundary between the access area and the relay area is fixed, inefficient resource usage occurs according to the channel state of the access link and the relay link. That is, resources that are wasted due to insufficient utilization of resources in the fixed area are increased.

4A and 4B illustrate an example of inefficient resource usage according to scheduling that may occur in an overlapping resource allocation scheme, and illustrates a resource allocation situation over two consecutive frames. In the multi-hop relay system, since the relay must transmit data transmitted from the base station to the terminals through the previous frame, the resources consumed in the relay area and the access area are consumed by the difference between the channel state with the base station and the channel state with the terminal. There is a difference in resources. For example, if the channel condition between the RS and the MS is very poor, even if the same amount of data is transmitted, a larger amount of resources are consumed than if the channel condition is good. 4A illustrates a case in which the channel state between the BS and the RS is very good while the channel state between the RS and the MS is very poor. That is, the relay will receive only as much data from the base station as can be transmitted depending on the channel status of the RS and MS intervals. If all relays require only a small amount of data, as shown here, a large amount of link between the BS and RS Resources can be wasted. On the contrary, in FIG. 4B, a channel condition between the BS and the RS is relatively poor. In other words, the link between the RS and the MS can transmit sufficient data with a small amount of resources in a relatively good channel condition. However, if the channel between the BS and the RS has a relatively poor channel condition, the data transmitted to the RS may be insufficient. As a result, the link resource between the RS and the MS cannot be utilized.

In addition, when the allocation is concentrated in the access area of a specific relay in the overlap allocation method, resource access may occur because the access areas of the remaining relays become areas that can no longer be used due to lack of relay areas. Therefore, when the allocation of access areas between relays is allocated in a uniform amount, the data throughput of the system can be maximized. However, even allocation of round robin will impair fairness between users and therefore require an appropriate tradeoff between the two.

It is an object of the present invention to provide a wireless packet scheduling method that guarantees fairness while increasing system data throughput in consideration of characteristics in a multihop relay system. To this end, the present invention proposes a scheduling method that dynamically establishes a boundary between an access region and a relay region in consideration of a multihop channel and considers an allocation amount of an access region between relays.

In order to achieve the above object, the present invention provides a packet scheduling method of a wireless data communication system,

(a) reflecting the instantaneous data rate of each user in consideration of the multi-hop channel to reflect the channel state of each hop;

(b) dynamically adjusting the boundary between the access area and the relay area by scheduling without fixing the size of the interval for the relay and the interval for the user connection in the system; And

(c) a method of packet scheduling in a wireless data communication system, the method including determining a user priority in consideration of an allocation amount of an access area between relays.

(여기서,

은 사용자 n의 멀티홉 릴레이 시스템에서의 순시 전송률을 나타내며,

는 사용자 n을 서비스하는 릴레이 k와 기지국사이의 순시 전송률을 의미한다.) 형태로 적용함으로써, 릴레이의 기지국과의 순시전송률과 사용자와 릴레이간의 순시전송률을 자원소모의 관점에서 통합한다. Preferably, in the step (a), the instantaneous data rate of each user used in the scheduler in the multi-hop relay system reflects the channel state of each hop,

(here,

Denotes instantaneous data rate in multi-hop relay system of user n,

Is the instantaneous transmission rate between the relay k serving the user n and the base station.) By integrating the instantaneous transmission rate between the relay base station and the user and the relay from the viewpoint of resource consumption.

delete

Preferably, the step of dynamically adjusting by scheduling in the multi-hop relay system using the time division scheme in step (b) is

가 정해진다. 모두 N개의 RS를 가정하고, k는 k번째 RS를 나타내는 첨자임; 즉,

). 그러나 (

,

)가 주어졌을 때, 시스템 전체적으로는 액세스 구간과 릴레이 구간의 경계는 하나로 설정된다. 하향링크 프레임 구간의 크기를

이라고 한다.) 에서, 각 릴레이에서 사용하는 자원중 최대치(Left Marker)와 릴레이로의 전송을 위해 사용되는 자원(Right Marker)의 합이 프레임의 자원량과 같아질 때까지 증가시켜가면서 동적으로 경계를 결정한다. (Where BS or each RS is the size of different segments

Is determined. Assuming all N RSs, k is the subscript representing the k-th RS; In other words,

). But (

,

), The boundary between the access section and the relay section is set to one. Downlink frame size

In this case, the boundary is dynamically increased while the sum of the left marker used for each relay and the right marker used for transmission to the relay is equal to the resource amount of the frame. Decide

Preferably in the step (c), for the efficiency of the resource

으로 나타내며, MS n의 장기 (long-term) 평균 전송률을

이고, MS n에 대한 종단간 유효 전송 속도를

이고,

는 사용자를 의미함)과 같은 형태로 사용자 우선순위를 정한다. Where the RAG value of each MS n

The long-term average transmission rate of MS n

End-to-end effective transmission rate for MS n

ego,

Is the user's priority).

을 아래와 같은 방법으로 결정하는 방식으로, Preferably the above

Is determined in the following way,

로 나타내고, t번째 프레임에서 MS n을 서비스하는 RS에서 사용되는 부채널의 수와 BS가 MS n을 서비스할 때 사용되는 부채널의 수를 각각

와

로 나타내고, MS n을 서비스하는 RS 또는 BS에서 사용된 총 부채널의 수와 그 보다 작은 수의 부채널을 사용하는 RS에서 사용된 총 부채널의 수를 뺀 값들의 합이 MS n이 선택되었을 때 낭비되는 자원에 해당하며, 이를 자원 할당 갭(resource allocation gap; RAG)이라고 정의하고, 각 MS n의 RAG 값을

으로 나타낸다.),(Where MS groups serviced via BS and MS groups serviced via RS, respectively) Wow

In the t-th frame, the number of subchannels used by the RS serving MS n and the number of subchannels used when the BS services MS n are respectively represented.

Wow

The sum of the total number of subchannels used in the RS or BS serving the MS n and the total number of subchannels used in the RS using the smaller number of subchannels is subtracted. It corresponds to resources that are wasted when it is defined as a resource allocation gap (RAG), and the RAG value of each MS n is defined.

),

　프레임 이전에 할당을 통해 각 릴레이에서 할당된 자원의 최대치로 임시 결정된 경계를 넘지 않을 경우

을 0로 정하고, 이를 넘어서면 현재 기지국에서 할당한 자원의 크기에서 각 릴레이에서 할당된 자원의 크기를 뺀 값들을 RAG(Resource Allocation Gap)으로 정의하고, The user is receiving service from a base station

The maximum number of resources allocated by each relay through allocation before the frame does not cross the temporarily determined boundary.

Is set to 0, and beyond this, the value of the resource allocated by the base station minus the size of the resource allocated by each relay is defined as a resource allocation gap (RAG),

Resource Allocation Gap is a combination of the method of determining the degree of resource waste and the difference between the size of the relay and the relay that serves itself and the size of the resource allocated to the relay that is less than the size of the allocated resource. ) And determine the degree of resource waste.

The present invention includes the steps of (a) reflecting the instantaneous data rate of each user considering the multi-hop channel to reflect the channel state of each hop; (b) dynamically adjusting the boundary between the access area and the relay area by scheduling without fixing the size of the interval for the relay and the interval for the user connection in the system; And (c) determining a user priority in consideration of a quota of access areas between relays, the method comprising: a broadband mobile Internet access system such as WiBro (Wireless Broadband Internet); In the multi-hop relay system for improving the yield and coverage of the system, the data throughput is increased while ensuring fairness.

Hereinafter, a packet scheduling method of a wireless data communication system according to the present invention will be described in detail with reference to the accompanying drawings.

First, in order to achieve the above object of the present invention, in the scheduling in a multi-hop relay system, 1) considering a multi-hop channel, 2) dynamically setting a boundary between an access area and a relay area, and 3 A scheduling method that considers the allocation of the access area between each relay.

로 나타내고, 이를 다음의 수학식 1과 같이 정의한다.First, in order to consider a multi-hop channel in the present invention, when transmitted by two hops through a relay link and an access link, since the instantaneous transmission speed may be different for each link, it is intended to apply opportunistic scheduling considering end-to-end channel state. It should be possible to reflect these transmission rates simultaneously. To do this, the effective end-to-end transmission rate for MS n serviced by the k th RS is determined.

It is defined as shown in Equation 1 below.

은 RS k와 MS n간의 순시 전송 속도이며,

는 BS와 RS k간의 순시 전송 속도를 나타낸다. 본 발명에서는 네트워크 효용성(network utilization)을 극대화하기 위해 각 사용자별 효용성의 합을 최대화하는 문제를 고려한다. here

Is the instantaneous transmission rate between RS k and MS n,

Denotes the instantaneous transmission rate between BS and RS k. In the present invention, in order to maximize network utilization, the problem of maximizing the sum of utility for each user is considered.

Given the traffic distribution and the channel condition, the boundary between the relay section and the access section in the frame is variably determined by the resource allocation result by the packet scheduler. If the boundary is fixed, resource shortage or resource waste occurs in the access section or the relay section depending on the traffic distribution. Therefore, for all RS coverages, an appropriate boundary should be dynamically set in consideration of the amount of resources allocated to the access intervals of the base station and the relay and the amount of resources allocated between the base station and the relay at the same time.

5 shows a data structure for explaining the frame structure and boundary setting of the multi-hop system of the present invention.

라고 나타낸다. 마찬가지로, 우측 표시부(right marker)에 의해 릴레이 구간의 크기가 결정되고, 그 크기를

이라고 나타낸다. 릴레이 구간의 경우에는 모든 릴레이들이 하나의 구간을 직교 분할하여 자원을 할당 받기 때문에 　

크기의 한 개 구간만이 주어진다. 반면, 액세스 영역의 경우에는 각 노드(BS 또는 RS)마다 재사용되는 자원이므로 이들은 서로 다른 좌측 표시부(left marker)를 갖게 되며, BS 또는 각 RS는 각기 다른 구간의 크기

가 정해진다(여기서 모두 N개의 RS를 가정하고, k는 k번째 RS를 나타내는 첨자임; 즉,

). In FIG. 5, a scheme for dynamically setting a section boundary in the downlink is illustrated. Here, the left marker indicates the boundary of the access section of the base station and each relay, and indicates the size of the access section determined by the node k in the frame t.

It is indicated. Similarly, the size of the relay section is determined by the right marker, and the size of the relay section is determined.

It is indicated. In the case of relay section, all relays are allocated resources by orthogonally dividing one section.

Only one interval of magnitude is given. On the other hand, in the case of the access area, since resources are reused for each node (BS or RS), they have different left markers, and the BS or each RS has different sizes of sections.

(Where assuming all RSs, k is a subscript representing the kth RS;

).

,

)가 주어졌을 때, 시스템 전체적으로는 액세스 구간과 릴레이 구간의 경계는 하나로 설정된다. 하향링크 프레임 구간의 크기를

이라고 한다면, 다음 수학식 2의 관계를 만족할 때까지 좌측 표시부(left marker)와 우측 표시부(right marker)를 각각 오른쪽과 왼쪽으로 이동하고 이들 표시부(marker)가 만나는 지점을 경계로 설정할 수 있다. But (

,

), The boundary between the access section and the relay section is set to one. Downlink frame size

In this case, the left marker and the right marker may be moved to the right and the left, respectively, until the relation of Equation 2 is satisfied, and the point where these markers meet may be set as the boundary.

　

　

That is, such a boundary is dynamically determined according to whether a resource is allocated to a BS or a terminal connected to a certain RS by packet scheduling. Meanwhile, the amount of resources that can be used by each link varies according to the set boundary, and resources that can be allocated to each user are limited. If the traffic load is not uniformly distributed, resource waste may occur in the access interval of the BS or a specific RS. Therefore, since boundary setting and packet scheduling for load balancing are interlocked mechanisms, an implementation considering both boundary setting and packet scheduling for load balancing is necessary.

6 illustrates a data structure exemplarily illustrating resource allocation in a t-th frame according to the present invention.

)번째 프레임의 RS-MS 구간을 결정하게 된다. 따라서 t번째 프레임에서 MS n이 BS에 의해 서비스 중일 때 낭비되는 자원은 t번째 프레임의 액세스 구간에서 구해지며, 이미 각 RS-MS 구간에서 사용되는 자원의 수는 결정되어 있다. 이때 각 RS-MS구간 중에서 가장 큰 값으로 액세스 구간이 잠정적으로 결정된다. 따라서 MS n이 BS에 의해 서비스 중일 경우에 이미 결정되어 있는 임시의 액세스 구간을 넘지 않으면 낭비되는 자원이 없다고 봐야 한다. 한편, t번째 프레임에서 MS n이 RS에 의해 서비스 중이면 (t+

)번째 프레임에서의 낭비되는 자원을 구하여야 한다. 따라서, MS n이 선택되었을 때 낭비되는 자원은 이 두 가지 경우로 구분하여 고려해야 한다. As shown in FIG. 6, it should be considered that in a real system, a packet serviced through a BS or an RS is allocated with resources in different frames. Referring to FIG. 6, the BS-MS section and the relay section in the t-th frame are determined through a scheduling algorithm, and the allocation result for the relay section thus determined is (t +

The RS-MS section of the) th frame is determined. Therefore, resources that are wasted when MS n is being serviced by the BS in the t th frame are obtained in the access period of the t th frame, and the number of resources used in each RS-MS period is already determined. At this time, the access section is tentatively determined as the largest value among the respective RS-MS sections. Therefore, it should be regarded that there is no wasted resource if the MS n does not cross the temporary access interval that is already determined when the BS is in service. On the other hand, if MS n is being serviced by the RS in the t th frame (t +

The wasted resources in frame 1 must be obtained. Therefore, resources that are wasted when MS n is selected should be considered in two cases.

와

로 나타내자. 그리고, t번째 프레임에서 MS n을 서비스하는 RS에서 사용되는 부채널의 수와 BS가 MS n을 서비스할 때 사용되는 부채널의 수를 각각

와

로 나타내자. MS n을 서비스하는 RS 또는 BS에서 사용된 총 부채널의 수와 그 보다 작은 수의 부채널을 사용하는 RS에서 사용된 총 부채널 의 수를 뺀 값들의 합이 MS n이 선택되었을 때 낭비되는 자원에 해당하며, 이를 자원 할당 갭(resource allocation gap; RAG)이라고 정의한다. 각 MS n의 RAG 값을

으로 나타내며, 각 경우에 따른 RAG은 다음 수학식 3과 같다. MS group served through BS and MS group served through RS, respectively

Wow

Let's represent it. In addition, the number of subchannels used in the RS serving the MS n in the t-th frame and the number of the subchannels used when the BS serves the MS n are respectively.

Wow

Let's represent it. The sum of the total number of subchannels used by the RS or BS serving the MS n minus the total number of subchannels used by the RS using the smaller number of subchannels is wasted when MS n is selected. It corresponds to a resource and is defined as a resource allocation gap (RAG). RAG value of each MS n

RAG in each case is represented by the following Equation 3.

이라고 할 때 이에 따른 효용성 함수(utility function)를

로 나타내자. 한편, 각 MS에 대한 자원 할당 시에 해당 RS에서의 자원 낭비에 따라 효용성이 감소하게 되므로, 다음의 수학식 4와 같은 목적 함수를 극대화 하는 자원 할당을 고려한다. The long-term average transfer rate of MS n

When we call this utility function

Let's represent it. On the other hand, when the resource allocation for each MS, the efficiency is reduced according to the waste of resources in the corresponding RS, consider resource allocation to maximize the objective function as shown in Equation (4).

목적함수=　

Objective =

로 설정하게 되면, 상기 수학식 4를 극대화하기 위한 최적의 스케줄링은 다음과 같이 매 채널 할당 단위로 다음 수학식 5를 만족하는 사용자

에게 채널을 할당하는 것이다. That is, the objective function reflects the inefficiency according to the RAG as a weight to the utility of each user. In other words, the smaller the user's RAG, the more it can contribute to overall utility. On the other hand, in order to provide proportional fairness among users,

If set to, the optimal scheduling for maximizing Equation 4 is a user satisfying the following Equation 5 in every channel allocation unit as follows:

Is to assign channels to

7 is a flowchart for resource allocation in the present invention. The resource allocation scheduler of FIG. 7 implies that the priority varies according to the RAG value of each MS at every scheduling time point, and intuitively, the allocation priority is given to an MS having a small RAG value. As such, a method of determining an appropriate boundary by reducing resources wasted while allocating resources according to channels is called load balancing opportunistic (LoBO) scheduling.

In FIG. 7, the resource allocation scheduling of the present invention first obtains the instantaneous data rate considering the multi-hop for each user by applying Equation 1 (step 700).

A resource allocation gap (RAG) value for each user is obtained by applying Equation 3 (step 720).

The highest-priority user selection is calculated by applying Equation 5 through priority calculation through RAGs, instantaneous rates, and average rates for each user (step 740).

Equation 2 is applied to update each region boundary (step 760).

The average transmission rate of each user is updated through Equation 6 below (step 780).

R _n (t + 1) = (T-1) R _n (t) / T + r _{k , n} / T for selected

R _n (t + 1) = (T-1) R _n (t) / T for others

In order to verify the effect of the present invention, the simulation results of the system level simulator (System Level Simulator) are as follows.

In the simulation, we increased the number of users in the multi-hop relay model and verified the yield, resource efficiency, and fairness for each case where the user traffic is full buffer and the actual Ethernet traffic is considered. The performance comparison target is when the boundary between the access section and the relay section is fixed and the PF (Proportional Fair) scheduling algorithm is applied. In this case, the section boundary is applied by finding and fixing a ratio that maximizes the average yield. On the other hand, in order to compare the fairness compared to the yield between the user defined a scale such as the following equation (7).

8 is a graph comparing yields as the number of users increases. First, comparing the average yield performance of the system in Figure 8, it can be seen that the yield performance of up to about 31% and 21%, respectively, in the case of full buffer and Ethernet traffic. In other words, it is a result of securing resource efficiency by reducing resource waste according to dynamic boundary setting for load balancing.

FIG. 9 is a graph comparing resource efficiency of each link, and the resource efficiency described above becomes more apparent when comparing resource efficiency of each link in a frequency reuse environment as shown in FIG. 9.

On the other hand, Figure 10 is a comparison chart of the fairness according to the increase in the number of users, and compares the performance measures defined by the equation (7), the proposed algorithm improves the system yield while the fairness (fairness) performance degradation compared to the yield performance improvement Shows not big. Therefore, it can be seen that the proposed algorithm can improve the system yield without significantly degrading fairness performance.

1 illustrates channel characteristics in a cellular model and a multi-hop relay model.

2 shows an uplink / downlink configuration in a multi-hop relay model.

3 shows a frame structure of a multi-hop relay system.

4A and 4B illustrate examples of inefficient resource usage according to scheduling that may occur in an overlapping resource allocation scheme.

5 shows a data structure for explaining the frame structure and boundary setting of the multi-hop system of the present invention.

6 illustrates a data structure exemplarily illustrating resource allocation in a t-th frame according to the present invention.

7 is a flowchart for resource allocation in the present invention.

8 is a graph comparing yields as the number of users increases.

9 is a graph comparing resource efficiency for each link.

10 is a graph comparing fairness as the number of users increases.

Claims (5)

In the packet scheduling method of a wireless data communication system, (a) reflecting the instantaneous data rate of each user in consideration of the multi-hop channel to reflect the channel state of each hop; (b) dynamically adjusting the boundary between the access area and the relay area by scheduling without fixing the size of the interval for the relay and the interval for the user connection in the system; And (c) determining a user priority in consideration of the allocation of the access area between relays. The method according to claim 1, wherein the instantaneous data rate of each user used in the scheduler in the multi-hop relay system in step (a) reflects the channel state of each hop.

(here,

Denotes instantaneous data rate in multi-hop relay system of user n,

Is the instantaneous data transfer rate between the relay k serving the user n and the base station.) The wireless data communication system integrates the instantaneous data transfer rate between the relay base station and the user and the relay from the viewpoint of resource consumption. Packet scheduling method. The method of claim 1, wherein the step of dynamically adjusting by scheduling in the multi-hop relay system using the time division scheme in step (b) is

(Where BS or each RS is the size of different segments

Is determined. Assuming all N RSs, k is the subscript representing the k-th RS; In other words,

). But (

,

), The boundary between the access section and the relay section is set to one. Downlink frame size

In this case, the boundary is dynamically increased while the sum of the left marker used for each relay and the right marker used for transmission to the relay is equal to the resource amount of the frame. Packet scheduling method of a wireless data communication system to determine. The method of claim 1, wherein in step (c),

Where the RAG value of each MS n

The long-term average transmission rate of MS n

End-to-end effective transmission rate for MS n

ego,

Is a user; and a packet scheduling method of a wireless data communication system for determining user priorities. The method of claim 4, wherein

Is determined in the following way,

(Where MS groups serviced via BS and MS groups serviced via RS, respectively)

Wow

In the t-th frame, the number of subchannels used by the RS serving MS n and the number of subchannels used when the BS services MS n are respectively represented.

Wow

The sum of the total number of subchannels used in the RS or BS serving the MS n and the total number of subchannels used in the RS using the smaller number of subchannels is subtracted. It corresponds to resources that are wasted when it is defined as a resource allocation gap (RAG), and the RAG value of each MS n is defined.

), The user is receiving service from a base station

The maximum number of resources allocated by each relay through allocation before the frame does not cross the temporarily determined boundary.

Is set to 0, and beyond this, the value of the resource allocated by the base station minus the size of the resource allocated by each relay is defined as a resource allocation gap (RAG), Resource Allocation Gap is a combination of the method of determining the degree of resource waste and the difference between the size of the relay and the relay that serves itself and the size of the resource allocated to the relay that is less than the size of the allocated resource. Packet scheduling method of a wireless data communication system, which is defined as

Images