KR100623000B1 - Method of Cost-based adaptive resource allocation in OFDMA systems - Google Patents

Abstract

In the present invention, a cost function based adaptive resource allocation method of an orthogonal frequency division multiple access system includes a) arranging subpackets to be transmitted by each user using a cost function, and assigning the sorted subpackets to respective subchannels. Assigning; And b) allocating respective powers according to the state of the subchannels.

Quadrature Frequency, OFDMA, Subchannel, Power

Description

Method of Cost-based Adaptive Resource Allocation in OFDMA Systems             

1 is a structural diagram of an orthogonal frequency division multiple access (OFDMA) system.

2 is a conceptual diagram of a water-filling technique for allocating different amounts of power according to channel quality.

3 shows the data throughput of each subchannel with and without considering quality of service at different guarantee levels.

4 is the probability of exceeding the maximum allowable delay when the quality of service of two delay sensitive real-time traffic is set to two guarantee levels (0.01, 0.03), respectively.

5 is an average delay time when the quality of service of each delay sensitive real-time traffic is set to two guarantee levels (0.01, 0.03), respectively.

6 is a comparison of the target amounts of each data rate sensitive traffic and the results obtained with the proposed resource allocation scheme.

The present invention relates to a technique for guaranteeing quality of service in a lower layer including a media access control and a physical layer of a wireless communication system. More specifically, the quality of service of each user in a lower layer of an orthogonal frequency division multiple access system The present invention relates to a method for allocating subchannels using a cost function that simultaneously reflects and channel quality, and a method for efficiently implementing the system using the same.

Orthogonal Frequency Division Multiplexing (OFDM) can overcome the deterioration of multiple paths of wireless channels, which has made it suitable for wideband high-speed communication systems. Orthogonal Frequency Division Multiplexing (OFDM) has been considered as a transmission technique for a single user environment in wireless, but has been variously studied for a multi-user environment in next-generation systems such as portable Internet and next-generation mobile communication systems. As an (OFDMA) system, the multi-user environment has been considered while taking advantage of Orthogonal Frequency Division Multiplexing (OFDM). As a result, the Orthogonal Frequency Division Multiple Access (OFDMA) system can guarantee a high transmission speed for a large number of users in a multipath wireless channel environment, and the channel state and transmitability that have been studied in the conventional Orthogonal Frequency Division Multiplexing (OFDM) scheme can be guaranteed. Adaptive modulation and coding (AMC) based on power, etc. has been made available, which has contributed to higher transmittable data throughput and higher quality of service based on data rates. In general, resources of an assignable lower layer of an orthogonal frequency division multiple access system are frequency division "subchannels" and "powers" allocated to each subchannel, and the "subchannels" are composed of a single or a plurality of subcarriers. And "power" is defined to include the allocation of data resulting from the allocation of power to a given subchannel.

However, as shown in [Table 1], the typical prior art of the orthogonal frequency division multiple access system solves the problem by using a mathematical optimization method, so that the target of the target quality of service is considered only the transmission speed and ensures the transmission speed. In addition, various kinds of data rate limitations cannot be flexibly supported, and the computational complexity of the optimization solution increases exponentially with the number of subchannels.

Therefore, in order to provide various users with various quality of service such as delay sensitivity, transmission speed sensitivity, and best-effort (BE) to the multiple users using orthogonal frequency division multiple access system, There is a need for a method that can flexibly cope with a small amount of computation and maintain a system-wide transmission throughput above a certain level.

① ② ③ ④ proponent C.Y. Wong et al. W. Rhee et al. J. Jang et al. Z. Shen et al. goal Minimize system usage power while maintaining a consistent system-wide transfer rate Maximizes overall transmission throughput while meeting each user's lowest transmission rate Maximize total transfer throughput (sum of each user's capacity) Maximizes overall transfer throughput while satisfying each user's transfer rate proportionality Solutions Optimized subchannel allocation Optimized subchannel allocation Optimized subchannel allocation Optimize power allocation

An object of the present invention is to solve the above problems, to overcome the limitations and complex optimization process that provides only a single quality of service related to the transmission speed of the prior art, delay sensitivity, transmission speed to multiple users It is to easily provide various quality of service such as sensitive and maximum effort service.

In the cost function-based adaptive resource allocation method of the orthogonal frequency division multiple access system of the present invention, a) using a cost function to arrange subpackets to be transmitted by each user, and assigning the sorted subpackets to each subchannel step; And b) allocating power according to the state of the subchannel.

In the present invention, the step a) may include: a1) updating the channel quality for the subchannel n of the user k for all users, and updating the quality of service state of the lower packet j for all users; a2) calculating, for all subpackets j, the cost for subchannel n of subpackets j owned by user k using a cost function; a3) sorting in large order according to the calculated cost; a4) allocating subchannel n to subpacket j of user k having the highest cost, excluding the allocated subchannel from an assignable subchannel, and excluding the allocated subpacket from a list of subpackets to be allocated ; And a5) repeating step a4) until the assignable subchannel and subpacket are not present.

로 정의되는 것이 바람직하다.In the present invention, the cost function (C _Q ) is

It is preferred to be defined as.

In the present invention, the power allocation is preferably by a water-filling technique.

Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings. Detailed descriptions of well-known functions and configurations that are determined to unnecessarily obscure the subject matter of the present invention will be omitted, and terms and symbols used will be defined as follows.

Lower layer: media access control layer and physical layer

* Subchannel: Subcarrier set consisting of one or more carriers in an orthogonal frequency division multiple access system

* Power: The power allocated to each subchannel can be converted into the amount of transmission data according to the modulation method, and the power allocated to subchannel _n is expressed as P _n .

* Cost function (C _Q ): A function that represents the cost according to the service quality or channel quality to consider, which is defined as in Equation 1, and the higher the value, the higher the priority.

* Cost: A value calculated from the cost function, where cos t _{j (k), n} represents the cost that the lower packet j (packet of user k) has for subchannel n

Delay-sensitive service: A service that has a limit on how long it takes for a packet to be sent to its destination.

* Data rate-sensitive service: A service that has a certain limit on the average amount of packets sent.

* Best-effort service: A service that allocates resources to packets that need to guarantee quality of service first and then provides them through the remaining resources.

* N: number of subchannels in orthogonal frequency division multiple access system

* K: total number of users in one orthogonal frequency division multiple access system network

* Allocation Unit: Minimum amount of data that can be transmitted through one subchannel

* Lower packet: One allocation unit packet that inherits the quality of service granted by the upper layer because the packet generated at the upper layer is divided into allocation units at the lower layer.

* Quality of Service Required: The quality of service granted when a packet is first generated at a higher layer, which should be provided by the system.

* Quality-of-service Status (QS): A state in which the generated packet is evaluated at a specific point in time in consideration of a changed environment such as a service provided or a delayed time, and qs _j represents a quality state of a subpacket j.

* Subchannel quality (CS: Channel Status): means that the size as the state of the sub-channel evaluation at a particular point in time the greater can transmit the same data with less energy, cs _n is the channel quality of the subchannel cs _k, and _n Denotes the channel quality of subchannel n for user k

* Frame: Physical Layer Transmission Symbol in Orthogonal Frequency Division Multiple Access System

* Frame period: It is a cycle in which transmission symbols are repeated, each time resource (sub-channel, power) allocation

1 is a block diagram showing an entire downlink of an orthogonal frequency division multiple access system according to an embodiment of the present invention.

Referring to FIG. 1, a base station transmitter considers a quality of service (QS) of a packet to be transmitted to each user and a subchannel quality (CS) according to each user secured in advance. Sub-packets appropriate for each subchannel are multiplexed and transmitted. Each user demodulates a frame transmitted through an orthogonal frequency division multiplexing system and demodulates its own information again.

The system model assumed in the present invention is as follows. It is assumed that the total number of users connected to one base station is k, each user always has lower packets to transmit, and the total number of subchannels in the orthogonal frequency division multiple access system is n.

As shown in FIG. 1, the present invention proceeds with two separate processes of subchannel and power allocation. This greatly helps to reduce the impossibility of implementing complicated computational calculations each time, which is difficult to quantify, such as delay sensitive services or services with many different characteristics.

First, in the subchannel allocation process, the cost function (C _Q ) of Equation (1) reflecting the quality of service (qs _j ) and the subchannel quality (cs _{k, n} ) of each user's lower packet is in a large order. It allocates the corresponding subchannel of the cost cos t _{j (k), n} from the lower packet j (the lower packet of user k) having the maximum cost.

Here, f (·) means a function and C _Q should include at least two parameters of a quality of service state (qs _j ) and a subchannel quality (cs _{k, n} ) as shown in Equation 1 above.

When one subpacket is allocated to a specific subchannel, the lower packet and the subchannel are excluded from the allocation process, and the allocation process is repeated at the costs of the remaining subpackets. Through the above process, subchannels are allocated in the order of the cost of each user, and all subchannels are allocated.

The subchannel allocation scheme proposed in the present invention can be summarized as follows and repeated every frame.

First, as a preliminary preparation step, for all users, the channel quality (cs _{k, n} ) of subchannel n of user k is updated, and the service quality state (qs _j ) of lower packet j is updated for all users.

Next, as a cost calculation step, for all subpackets j, the cost for subchannel n of subpacket j (owning user k), cos t _{j (k), n} is calculated using the cost function of Equation 1 above. do.

Subsequently, the subchannels are allocated to the lower packet j (user k) having the largest cost cos t _{j (k), n,} and the subchannels are allocated according to the calculated cost. Except in the channel, the allocated subpacket is excluded from the list of subpackets to be allocated, and iterates while allocating subchannels and subpackets are repeated until all subchannels are allocated.

In the power allocation scheme following the subchannel allocation, power is supplied by a water-filling scheme as shown in FIG. 2 to maximize the transmission speed of the entire system under the constraint of a given subchannel allocation and the total amount of power that can be allocated. Assign. The power allocated to subchannel n is denoted by p _n , which represents the power to which the difference between the water-filling level and Γ / cs _n is allocated. Where cs _n is the channel quality state for the subchannel, Γ is a constant according to the modulation method and transmission limit (target bit error rate, etc.), and the better the channel state, the larger the value. do.

Hereinafter, as an example of the proposed scheme, the subchannel allocation using the cost function and the performance of the implementation method will be described in consideration of the case where there are three service groups such as delay sensitivity, transmission rate sensitivity, and maximum effort service. .

<Support Service Group>

There are three groups of delay sensitive, rate sensitive, and maximum effort service. Each delay sensitive and rate sensitive service can have different delay limit and rate limit. The network traffic environment used for the performance evaluation is shown in Table 2. Delay-sensitive services and rate-sensitive services are defined by the following quality of service limitations. The constraint of the delay sensitive service is defined as in Equation 2 below.

Where W _j is the delay of the subpacket j, T _j is the maximum allowable delay, and σ _j is the delay exceeded probability. Therefore, the probability that the delay of the packet j exceeds the predetermined maximum allowable delay must be kept below the delay allowable probability σ _j . In this example, σ _j is considered in the result diagram considering two situations, 0.03 and 0.01.

The constraint of the transmission rate sensitive service is defined as in Equation 3.

R _j ≥dr _j

Here, R _j is the transmission rate of the lower packet _j , dr _j is the target transmission rate. Therefore, the transmission rate of the packet j must be equal to or higher than the predetermined target transmission rate dr _j .

Service group Delay sensitive Baud rate sensitive Maximum effort Generated traffic 64 128 64 128 384 64 Service goals Delay 9 ms Delay 4.5 ms Transmission speed 64 128 baud Baud rate 384 none Number of users 12 20 12 8 4 8

<Cost function>

In the example, we use the M-LWDF method that was used for wireless scheduling of 3G mobile communication. This cost function is given by Equation 4.

,

로 정의된다. 여기서,

는 평균 품질 상태이며 β는 임의의 상수이다. 또한,

는 현재 채널의 품질상태로써, k의 부채널 n에 대한 채널 품질 상태를 나타낸다.

,

Is defined as here,

Is the average quality state and β is an arbitrary constant. Also,

Is the quality state of the current channel, and represents the channel quality state for subchannel n of k.

Delay W _j shown in Equation 4 is a delay time, but a speed-sensitive service can also be easily expressed as a delay time using the amount of remaining subpackets and the ratio of packets to be transmitted.

<Orthogonal Frequency Division Multiple Access System Lower Layer Model and Radio Channel Model>

In this example, the media access control and the physical layer of IEEE 802.16a, a powerful lower layer model of the portable Internet, are used as an example model of an orthogonal frequency division multiple access system.

* Carrier frequency: 2.3

* Total carriers: 2048

* Total number of subchannels (N): 8

* Modulation method: QPSK, 16QAM, 64QAM

Wireless Channel Model: Modified SUI model

* Target Bit error rate: 10-6

Power Limit: Average

<Performance Evaluation Results>

3 shows transmission throughput results (1 Mbps) obtained by assigning power only to users who exhibit the best channel performance in each subchannel without considering quality of service, and quality of service with two probabilities (δ _j = 0.03, 0.01). The transmission throughput was considered when compared. It can be seen that the transmission throughput is reduced by about 27% compared with not considering the quality of service of the user at all, but this is generally a sacrifice in consideration of quality of service.

In FIG. 4, it can be seen that two delay sensitive services (64 Kbps and 128 Kbps) maximum allowable delay time exceeding probability satisfy a preset allowable probability (δ _j = 0.03, 0.01). In addition, in FIG. 5, it can be seen that the average delay time according to the user satisfies the maximum allowable delay time and is evenly distributed.

In FIG. 6, the change in the actual transmission rate according to the target transmission rate dr _j is set according to the number of repetitions in the continuous simulation, and it can be seen that the target transmission rate is maintained regardless of the number of attempts.

As described above, it has been described with reference to the preferred embodiment of the present invention, but those skilled in the art various modifications and changes of the present invention without departing from the spirit and scope of the present invention described in the claims below I can understand that you can.

As described above, the present invention provides a relatively simple service by allocating a subchannel using a cost function to a service such as a delay sensitive service that has been difficult to guarantee until now and a mathematically difficult approach in the presence of various services at the same time. Quality can be satisfied.

In addition, the importance is amplified when the orthogonal frequency division multiple access system is considered as a major foundation technology such as portable Internet and next generation mobile communication system.

Claims (4)

In the Cost Function Based Adaptive Resource Allocation Method for Orthogonal Frequency Division Multiple Access System, a) arranging subpackets to be transmitted by each user using a cost function, and assigning the sorted subpackets to each subchannel; And b) A cost function based adaptive resource allocation method of an orthogonal frequency division multiple access system, comprising: assigning each power according to the state of the subchannel. The method of claim 1, wherein step a) a1) updating channel quality for subchannel n of user k for all users, and updating the quality of service state of subpacket j for all users; a2) calculating, for all subpackets j, the cost for subchannel n of subpackets j owned by user k using a cost function; a3) sorting in large order according to the calculated cost; a4) allocating subchannel n to subpacket j of user k having the highest cost, excluding the allocated subchannel from an assignable subchannel, and excluding the allocated subpacket from a list of subpackets to be allocated ; And a5) repeating the step a4) until the assignable subchannel and the subpacket are not present; and a cost function based adaptive resource allocation method of an orthogonal frequency division multiple access system. The method of claim 1, wherein the cost function C _Q is

Cost function-based adaptive resource allocation method of orthogonal frequency division multiple access system, characterized in that defined as. However, cs _{k, n} represents the quality of the subchannel, and qs _j represents the quality of the service. The method of claim 1, wherein the power allocation is performed by a water-filling technique.

Images