KR20040024784A - Method for composing distributed MAC protocol with service differentiation over IEEE 802.11 Ad-Hoc Wireless LANs - Google Patents

Abstract

Disclosed is a distributed MAC protocol design method for service differentiation in an IEEE 802.11 Ad-hoc WLAN. The distributed MAC protocol configuration method includes: (a) configuring a first or second contention window according to whether a corresponding terminal is a real-time terminal or a non-real-time terminal during channel access; And (b) when the real-time terminal and the non-real-time terminal simultaneously request channel access, include a predetermined offset value in the second contention window of the non-real-time terminal, so that the real-time terminal has priority over the non-real-time terminal. And accessing the channel. According to such a distributed MAC protocol configuration method, it is possible to support real-time traffic with simple addition of complexity, and to completely ensure backward compatibility with existing systems.

Description

Method for composing distributed MAC protocol with service differentiation over IEEE 802.11 Ad-Hoc Wireless LANs}

The present invention relates to a wireless local area network (WLAN) technology, and more particularly, to a distributed MAC protocol configuration method capable of efficiently supporting real-time traffic in an IEEE 802.11 Ad-hoc network.

The IEEE 802.11 WG in the United States laid the groundwork for the proliferation of wireless LANs by completing IEEE 802.11, the first standard for wireless LANs in 1997. Since then, the new physical layer standard has been developed to reflect the market demand for high-speed wireless LANs. IEEE 802.11b and IEEE 802.11a are additionally prepared.

The MAC adopted by IEEE 802.11 is called DFWMAC (Distributed Foundation Wireless MAC). DFWMAC is an infrastructure that includes both ad-hoc configuration and direct access point and distribution system that can communicate directly with each station within the communication range. It is intended to be used for.

DFWMAC Coordination Function (DFWMAC Coordination Function) is largely divided into DCF (Distributed Coordination Function) and PCF (Point Coordination Function). The DCF function is used for asynchronous data communication, and the PCF function is used for time-limited synchronous data transmission. DCF uses carrier sense multiple access with collision avoidance (CSMA / CA) as a basic channel access mechanism for transmitting non-real-time data in a contention period. The CSMA / CA method has an advantage that it is easy to implement because there is no additional control message and the operation process is simple, while support of real-time traffic is impossible.

In addition, IEEE 802.11TGe is progressing a standard for providing a quality of service (QoS) in a wireless LAN section. In order to provide QoS to the existing MAC, it applies a differential method of prioritizing parameters in addition to the existing DCF mechanism. It operates queues according to each traffic category (TC) and uses different parameters. To give priority to real-time and non-real-time traffic. However, the above method has a problem of a high possibility of collision in accessing channels of real-time traffic and non-real-time traffic, and in the case of delay-sensitive real-time traffic, it is impossible to give flexible priorities in fact. In some cases (e.g., when there is a lot of delay in the queue), traffic with a higher delay bound should have a higher priority than traffic with a lower delay bound among the traffic with constraints on the delay. Occurs. Therefore, this type of prioritization scheme is not suitable for guaranteeing QoS for real-time traffic.

An object of the present invention is to provide a distributed MAC protocol capable of differentiating a channel access interval between a real time terminal and a non real time terminal by operating two contention windows for each of a real time terminal and a non real time terminal.

Another object of the present invention is to provide a distributed MAC protocol configuration method capable of guaranteeing QoS for real-time traffic by adjusting a probability distribution function to give priority to real-time traffic.

1 is a flowchart illustrating a method of configuring a distributed MAC protocol according to a preferred embodiment of the present invention.

2 is a view showing an initial channel access procedure according to a preferred embodiment of the present invention.

FIG. 3 is a diagram illustrating an example of a multi-hop based Basic Service Set (BSS) in which several cells overlap.

FIG. 4 is a diagram for describing a non-real-time hidden terminal problem that may occur between the terminals illustrated in FIG. 3.

FIG. 5 is a diagram illustrating an example of a probability distribution function G () in the form of a linear function for giving priority to a residual time.

6 is a diagram showing an example of a probability distribution function G () in the form of a quadratic function for giving priority to the remaining life.

7A and 7B are diagrams showing a result of giving priority to k in the probability distribution function G () of the linear function form shown in FIG. 5.

8A and 8B are diagrams showing a result of giving priority to k in the probability distribution function G () of the quadratic function form shown in FIG. 6.

9A and 9B are diagrams showing changes in the convexity of the probability distribution function G () according to ε in the probability distribution function G () of the quadratic function form shown in FIG. 6.

FIG. 10 is a diagram illustrating a collision probability analysis result between terminals for different types of probability distribution functions.

FIG. 11 is a diagram illustrating an average packet loss probability of a real-time terminal when five non-real-time terminals (STA _NRT ) are used, when a distributed MAC protocol according to the present invention is applied and when a general DCF is applied.

FIG. 12 is a diagram illustrating an average MAC delay time of a real-time terminal when five non-real-time terminals (STA _NRT ) are used, when a distributed MAC protocol according to the present invention is applied and when a general DCF is applied.

FIG. 13 is a diagram illustrating an average packet loss probability of a non-real time terminal when 5 real time terminals (STA _RT ) are used, and when a distributed MAC protocol according to the present invention is applied and a general DCF is applied.

FIG. 14 is a diagram illustrating average MAC delay time of a non-real-time terminal when five real time terminals (STA _RT ) are used, and when a distributed MAC protocol according to the present invention is applied and a general DCF is applied.

In order to achieve the above object, a method of configuring a distributed MAC protocol according to the present invention includes: (a) configuring a first or second contention window according to whether a corresponding terminal is a real-time terminal or a non-real-time terminal during channel access; And (b) when the real-time terminal and the non-real-time terminal simultaneously request channel access, include a predetermined offset value in the second contention window of the non-real-time terminal, so that the real-time terminal has priority over the non-real-time terminal. And accessing the channel.

Hereinafter, with reference to the accompanying drawings will be described in detail a preferred embodiment of the present invention.

1 is a flowchart illustrating a method of configuring a distributed MAC protocol according to a preferred embodiment of the present invention. Referring to FIG. 1, in the method for configuring a distributed MAC protocol according to the present invention, two contention windows are configured according to whether a corresponding terminal is a real-time terminal or a non-real-time terminal when a channel is accessed, and when the real-time terminal is accessed, Priority is given between the real-time terminals by using the remaining life of the terminals.

To this end, the method for configuring a distributed MAC protocol according to the present invention first determines whether a corresponding terminal is a real-time terminal (STA _RT ) or a non-real-time terminal (STA _NRT ) during channel access (step 1000).

As a result of the determination in step 1000, if the corresponding terminal is a real time terminal STA _RT , a first contention window CW _i ^(rt ) is configured (step 1100), and priority is given to the real time terminals by using the remaining life of the real time terminals. After granting (step 1300), channel access is performed (step 1400).

If it is determined in step 1000 that the corresponding terminal is a non-real-time terminal STA _NRT , a second contention window CW _i ^(nrt ) is configured (step 1200) and channel access is performed accordingly (step 1400). ).

Here, the method of determining the first and second contention windows will be described in detail with reference to FIGS. 2 to 4 below. In addition, a method for determining priority among real-time terminals will also be described in detail with reference to FIGS. 5 to 9B below.

2 is a view showing an initial channel access procedure according to a preferred embodiment of the present invention. Referring to FIG. 2, the present invention operates two contention windows to differentiate channel access between a real-time terminal and a non-real-time terminal. One is the contention window CW _i ^(rt) for the real-time terminal STA _RT , and the other is the contention window CW _i ^(nrt) for the non-real-time terminal STA _NRT .

If a collision between terminals initially occurs (that is, if i≥1), the contention window CW _i ^(rt) (or CW _i ^(nrt) ) increases exponentially to CW _{RT_max} (or CW _{NRT_max} ). The non-real time terminal STA _NRT selects the back-off time by adding as much as the maximum contention window value CW _{RT_max} of the real-time terminal as an offset value during initial access. The maximum contention window value CW _{RT_max} (ie, the offset value) is a time value for guaranteeing a channel access period for the real time terminal STA _RT .

If a collision occurs during channel access, the backoff time _Bi ^(nrt) of the non-real-time terminal is selected in the interval [0, CW _i ^(nrt) ]. At this time, the non-real-time terminal (STA _NRT ) has an offset value for access differentiation only when attempting the first access after the packet is generated, and does not apply the offset value in the retry process. The reason is to prevent the rapid increase in the average delay time of the non-real-time terminal (STA _NRT ). The backoff time _Bi ^(nrt) for the non-real time terminal STA _NRT is determined by Equation 1 below.

Here, Random [0, CW _i ^(nrt) ] is a pseudo-random integer value extracted by the uniformdistribution function in the interval [0, CW _i ^(nrt) ], and Slot_time is the first of the preamble. From the time the bit is transmitted, it means the time it takes to stop the transmission due to the detection of the collision, and i means the number of retries. Ratio to live terminal (STA _NRT) contention window ^(CW _i ^(nrt)) for a minimum contention window value (CW _{NRT_min)} to, 2 ⁱ CW _{NRT_min} maximum contention window value (CW _{NRT_max)} of value when a conflict occurs, starting at Increases exponentially

The backoff time _Bi ^(rt) for the real time terminal STA _RT is determined by Equation 2 below.

Here, Random_G [0, CW _i ^(rt) ] is an integer value of similarity extracted by the new distribution function G () in the interval [0, CW _i ^(rt) ], and Slot_time is the time when the first bit of the preamble is transmitted. Is the time taken to stop the transmission by detecting the collision, and i means the number of retries. Contention window ^(CW _i ^(rt)) is exponential to the minimum contention window value starting from the (CW _{RT_min),} when a conflict arises 2 ⁱ CW _{RT_min} maximum contention window value of the value (CW _{RT_max)} for real-time station (STA _RT) Increase by enemy. The distribution function G () used in [Equation 2] is a probability distribution function used for giving priority between real-time terminals, and a description thereof will be described below with reference to FIGS. 5 to 9b.

According to Equations 1 and 2, the backoff times B _i ^(rt) and B _i ^(nrt) are determined according to the real time terminal STA _RT and the non-real time terminal STA _NRT , respectively. It is determined differently. The backoff time _Bi ^(rt) and _Bi ^(nrt) determined in this way ensure that the real-time terminal STA _RT accesses the channel prior to the non-real-time terminal STA _NRT .

In FIG. 2, the offset value for separating the channel access interval is 8, the minimum contention window values CW _{RT_min} and CW _{NRT_min of the} real time terminal STA _RT and the non-real time terminal STA _NRT are 4, respectively. Assume that the initial backoff time B ₀ ^(rt ) of STA _RT ) is 4 and the initial backoff time B ₀ ^{(nrt) of the} non-real time terminal STA _NRT is selected to 12. Here, the initial backoff time B ₀ ^{(nrt) of the} non-real time terminal STA _NRT means an offset value (= 8) plus an arbitrary value (= 4) selected in the interval [0, CW _{NRT_min} ]. do.

As can be seen in [Equation 1] and [Equation 2], the initial backoff time (B ₀ ^(nrt) ) of the non-real-time terminal STA _NRT selects any value in the interval [0, CW _{NRT_min} ]. Even if (eg, select a value of ₀ ⁾ , it can be seen that the offset value always has a larger value than the initial backoff time B ₀ ^(rt) of the real time terminal STA _RT . Accordingly, non-real-time live terminal station (STA _RT) having a back-off time than (STA _NRT) is able to preferentially access the channel than the non-real-time station (STA _NRT).

As described above, when real-time traffic and non-real-time traffic coexist, the channel access method according to the present invention is basically based on the CSMA / CA scheme like the DCF scheme. Accordingly, the channel access interval of the real time terminal STA _RT and the non-real time terminal STA _NRT may be effectively mapped by using an appropriate offset value in the process of determining the backoff time _Bi ^(nrt) . Can be separated.

An important consideration in separating the channel access interval between the real time terminal (STA _RT ) and the non-real time terminal (STA _NRT ) is how to set the offset value. There are two types of ways to put an offset value. One of them is to set the offset value adaptively according to the load of the real time terminal STA _RT , and the other is to set the offset value to a fixed value regardless of the load of the real time terminal STA _RT . .

3 is a diagram illustrating an example of a multi-hop based BSS (Basic ServiceSet) in which several cells are overlapped, and FIG. 4 is a hidden terminal that may be generated between the terminals illustrated in FIG. 3. -real-time hidden terminal)

First, referring to FIG. 3, since terminals 110-130 located in one BSS can communicate with each other, it is possible to recognize each collision and update its contention window value. That is, the channel access interval of the real time terminals can be guaranteed by recognizing a collision between the non-real time terminal STA _NRT 120 and the real time terminal STA _RT 110 and updating the offset value. In this case, CW _i ^(rt) is used instead of CW _{RT_max in} [Equation 1] to specify an offset value using a fluid method.

When multiple cells are extended based on overlapping multi-hop, a non-real-time hidden terminal (STA _{NRT_hidden} ) 130 should be considered. In addition, the terminal 130 _{notifies the} non-real time hidden terminal (STA _{NRT_hidden} ; 130) of the collision information of the real time terminal (STA _RT ; 110) in the BSS in order not to disturb the channel access of the real time terminal. A real time hidden terminal (STA _{NRT_hidden} ; 130) needs to update the offset value. That is, when a collision occurs, the real-time terminal (STA _RT ; 110) increases the backoff value, but all terminals belonging to one cell can recognize it and increase their backoff size, but the non-real-time concealment terminal (STA) _{NRT_hidden} ; 130) does not recognize this.

When the offset value is flexibly placed in FIG. 4, the offset value is determined by the initial contention window value CW ₀ ^(rt) of the real-time terminal STA _RT . After the collision occurs, the non-real-time terminal (STA _NRT ; 120) in the BSS checks whether there is a collision between the real-time terminals (STA _RT ; 110) ^, and updates its offset value to CW ₁ ^(rt) by real-time terminal. The channel access interval of (STA _RT ; 110) is guaranteed. However, the non-real time concealment terminal (STA _{NRT_hidden} ; 130) does not recognize whether or not there is a collision, and thus has an offset value. At this time, non-real-time aiding terminal; because (STA _{NRT_hidden} 130) is not the offset value is updated, real-time terminal; a channel access period of the (STA _RT 110) non-real-time aiding terminal; be subject affected by (STA _{NRT_hidden} 130) As a result, the channel access interval of the real time terminal STA _RT cannot be guaranteed. Accordingly, in the present invention, in order to solve such a problem, the offset value is fixed to the maximum contention window value CW _{RT_max} of the real-time terminal when the initial backoff time B ₀ ^(nrt ) is obtained as shown in [Equation 1]. Let's do it.

In addition to such multi-hop-based BSSs, single-hop-based BSSs defined in IEEE 802.11 Ad-hoc wireless LANs require that information be exchanged between terminals whenever the network topology changes in order to flexibly take offset values. . At this time, if the network topology changes frequently, the exchange of control information between the terminals can greatly reduce the throughput of the network, and the complexity of implementation is also greatly increased. Therefore, it may be reasonable to fix the offset value to the maximum contention window value CW _{RT_max} of the real-time terminal from the implementation point of view.

Here, the maximum contention window value (CW _{RT_max} ) of the real-time terminal should be able to select a suitable value according to the load of traffic. If the value CW _{RT_max} used as the offset value is increased, a problem occurs in the delay time of the non-real-time terminal. Therefore, it is important to select the maximum contention window value CW _{RT_max} that maximizes the performance of the real-time terminal while minimizing the performance degradation due to the delay of the non-real-time terminal.

In the distributed MAC protocol according to the present invention, in order to guarantee the QoS for the real-time traffic, in addition to the channel access differentiation method between the real-time terminal and the non-real-time terminal as described above, the priority scheme between the real-time traffic using the probability distribution function G () is proposed. Doing. Priority provision between real-time traffic according to the present invention is as follows.

In general, a delay bound value is mainly used as a criterion for giving priority to real-time traffic. However, even if the delay bound is large, a situation in which a packet of the corresponding terminal waits in the queue for a long time may require a higher priority than the terminal having a small delay bound. Therefore, in the present invention, priority is given by using the probability distribution function G () using the remaining lifetime of each traffic. The probability distribution function G () is used to appropriately set the priority required according to the remaining lifetime of each traffic.

FIG. 5 is a diagram illustrating a probability distribution function G () in the form of a linear function for giving priority to a residual time, and FIG. 6 is a probability in the form of a quadratic function for giving priority to a remaining life. Figure showing the distribution function G ().

An example of the probability distribution function G () in the form of a linear function shown in FIG. 5 is represented by Equation 3, and an example of the probability distribution function G () in the form of a quadratic function shown in FIG. If it is represented by Equation 4, Equation 4 is the same.

In [Equation 3] and [Equation 4], And k (laxity factor) is determined as shown in [Equation 5].

When the delay bound of the real-time traffic is D _max in the functions derived by Equations 3 and 4, k is the remaining lifetime ( ) And D _max are determined as follows.

7A and 7B are diagrams showing a result of giving priority to k in the probability distribution function G () of the linear function form shown in FIG. 5, and FIGS. 8A and 8B are probabilities of the quadratic function form shown in FIG. 6. It is a figure which shows the result of giving priority to k in distribution function G ().

Referring to FIGS. 7A to 8B, as can be seen from Equation 5, the probability distribution function G _k (x) has a lower priority when the remaining lifespan remains large (that is, when the k value is large), and the residuals It can be seen that the priority is high when the life span is low (that is, when the value of k is small).

In giving priority among real-time traffic, the consideration in the probability distribution function G () of the quadratic function is to select which value of ε to minimize the collision when several terminals have arbitrary remaining lifetime. Is the point. That is, in the case of a quadratic function, the convexity of the function varies depending on the value of ε.

9A and 9B are diagrams showing changes in the convexity of the probability distribution function G () according to ε in the probability distribution function G () of the quadratic function form shown in FIG. 6.

9A and 9B, it can be seen that as the value of ε increases, the convexity of the probability distribution function G () decreases, and as the value of ε decreases, the convexity of the probability distribution function G () increases. However, the epsilon value at this time is always smaller than 1 / CW _RT .

In the above, as an embodiment of the present invention, the probability distribution function G () of the first and second functions is specifically illustrated as a probability distribution function for prioritizing real-time traffic, but in addition, the priority distribution for real-time traffic is provided. Various types of probability distribution functions can be used as the probability distribution function.

FIG. 10 is a diagram illustrating a collision probability analysis result between terminals for different types of probability distribution functions. FIG. 10 shows a collision probability analysis result between terminals for a linear probability distribution function and probability distribution functions of quadratic functions having various values of ε.

Referring to FIG. 10, the contention window (CW ₀ ^(RT) ) value of the real-time terminal is 10, and the remaining lifetime ( ) Is arbitrarily selected in the range of [0, D _max ], it can be seen that the probability distribution function when the ε value is 1/32 is an optimal probability distribution function for prioritizing real-time traffic.

Parameters for simulating the performance of the distributed MAC protocol according to the present invention are shown in Table 1 below.

Simulation parameters value Channel data rate 1 Mbps D _max 25㎳ Slot time 20㎲ Short interframe time (SIFS) 10㎲ PCF interframe time (PIFS) 30㎲ DCF interframe time (DIFS) 50㎲ CW _{RT_min} 31 CW _{RT_max} 255 CW _{NRT_min} 31 CW _{NRT_max} 1023 L 0

11 and 12 illustrate the average packet loss probability of the real-time terminal and the non-real-time terminal when 5 non-real-time terminals (STA _NRT ) are used, when a distributed MAC protocol according to the present invention is applied and when a general DCF is applied. A diagram showing average MAC delay times.

First, referring to FIG. 11, it can be seen that the packet loss probability when the distributed MAC protocol according to the present invention is applied is significantly smaller than other cases. In particular, when the distributed MAC protocol, while the present invention is processed to 12 live terminal (STA _RT) when applied in accordance with, DCF scheme is applied, it can be seen that the treated only six live terminal (STA _RT). Therefore, in the distributed MAC protocol method according to the present invention, the amount of real-time terminals (STA _RT ) to be processed is increased by about 2 times compared to the existing method.

12, it can be seen that when the distributed MAC protocol according to the present invention is applied, the average MAC delay time of the non-real time terminal is slightly delayed compared to the case where the general DCF is applied. This is derived from the offset value for the non-real-time terminal, the offset value affects the non-real-time terminal, but does not affect the real-time terminal. This average MAC delay of the non-real time terminal has little effect on the overall performance of the wireless LAN system.

13 and 14 illustrate the average packet loss probability of the non-real time terminal and the non-real time terminal when 5 real time terminals (STA _RT ) are used, when the distributed MAC protocol according to the present invention is applied and when a general DCF is applied. A diagram showing average MAC delay times.

13 and 14, it can be seen that when the distributed MAC protocol according to the present invention is applied, similar results to those shown in FIGS. 11 and 12 are shown for the non-real time terminal.

Such performance of the distributed MAC protocol according to the present invention is described according to the maximum contention window value CW _{RT_max} of the real-time terminal.

CW _{RT_max} Average packet loss probability Average MAC Latency 63 0.03099 0.40024 127 0.01890 0.42523 255 0.00940 0.46982 511 0.00529 0.51405

The performance comparison result of the present invention and the DCF method shown in [Table 2] is an example in which five real-time terminals (STA _RT ) and 16 non-real-time terminals (STA _NRT ) are used. 255.

As described above, in the DCF scheme of the existing 802.11 Ad hoc MAC, the present invention guarantees QoS for real-time traffic by configuring the contention window part and the probability distribution function part differently. Therefore, the present invention can effectively support real-time traffic while maintaining coexistence with the existing Ad-hoc based WLAN system.

As described above, optimal embodiments have been disclosed in the drawings and the specification. Although specific terms have been used herein, they are used only for the purpose of describing the present invention and are not intended to limit the scope of the invention as defined in the claims or the claims. Therefore, those skilled in the art will understand that various modifications and equivalent other embodiments are possible from this. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

As described above, according to the distributed MAC protocol according to the present invention, two contention windows are operated separately for the real time terminal and the non-real time terminal. Therefore, the channel access interval between the real time terminal and the non-real time terminal can be differentiated.

In addition, according to the distributed MAC protocol according to the present invention, by using the probability distribution function using the remaining lifetime of each traffic instead of the uniform probability distribution function, priority is given to the real-time traffic, thereby ensuring QoS for real-time traffic. Can be provided.

Claims (18)

(a) configuring a first or second contention window according to whether the corresponding terminal is a real-time terminal or a non-real-time terminal during channel access; And (b) When the real-time terminal and the non-real-time terminal simultaneously request channel access, the real-time terminal is given priority over the non-real-time terminal by including a predetermined offset value in the second contention window of the non-real-time terminal. And accessing the channel. The method of claim 1, The first backoff period corresponding to the first contention window is similar to Random_G [0, CW _i ^(rt) ] extracted by a predetermined probability distribution function G () in the interval [0, CW _i ^(rt) ]. When it represents an arbitrary integer value and Slot_time represents the time taken to stop transmission after collision detection, Distributed MAC protocol configuration method characterized in that it has a value. The method of claim 2, And wherein the first contention window increases exponentially within a predetermined range from a first minimum contention window value (CW _{RT_min} ) to a first maximum contention window value (CW _{RT_max} ). The method of claim 1, The second backoff period corresponding to the second contention window represents an integer value of Random [0, CW _i ^(nrt) ] similarly extracted by the uniform distribution function in the interval [0, CW _i ^(nrt) ]. When Slot_time represents the time taken to stop transmission after collision detection, Distributed MAC protocol configuration method characterized in that it has a value. The method of claim 4, wherein And the second contention window increases exponentially within a predetermined range from a second minimum contention window value (CW _{NRT_min} ) to a second maximum contention window value (CW _{NRT_max} ). The method of claim 1, The non-real-time terminal uses the offset value for access differentiation when the first attempt after the packet is generated, and does not use the offset value in the retry process. The method of claim 1, The initial backoff time of the non-real-time terminal, the distributed MAC protocol configuration method, characterized in that always has a value larger than the initial backoff time of the real-time terminal. The method of claim 1, The offset value may have the offset value in accordance with the load of the real-time terminal, and may have a fixed arbitrary value irrespective of the load of the real-time terminal. The method of claim 8, If the offset value has a flexible value, the offset value is set to the distributed MAC protocol, characterized in that the initial backoff time of the real-time terminal is set. The method according to claim 3 or 8, And when the offset value has a fixed value, the offset value is set to the first maximum contention window value (CW _{RT_max} ) of the real-time terminal. The method of claim 1, The distributed MAC protocol configuration method may further include (c) assigning priority to the real time terminals by using the remaining life of the real time terminals when the real time terminal is accessed. The method of claim 11, The step (c) of the distributed MAC protocol configuration method characterized in that to give priority to the real-time terminals by using the probability distribution function G () using the remaining life of each traffic. The method of claim 12, The probability distribution function G () in the form of a linear function When CW _RT represents the first contention window and k represents a relaxation coefficient proportional to the latency of each traffic, ego, Distributed MAC protocol configuration method characterized in that. The method of claim 12, The probability distribution function G () in quadratic form When CW _RT represents the first contention window and k represents a relaxation coefficient proportional to the latency of each traffic, ego, Distributed MAC protocol configuration method characterized in that. The method according to claim 13 or 14, The relaxation coefficient k is D _max is the delay bound of the real-time traffic, Is the remaining life of the traffic, Distributed MAC protocol configuration method characterized in that. The method of claim 13, The probability distribution function G () in the form of the linear function indicates a low priority when the remaining lifespan remains large and a high priority when the remaining lifespan remains low. The method of claim 14, The probability distribution function G () of the quadratic function form is characterized in that the collision probability of the terminals varies according to the value of ε when the terminal has a certain remaining lifetime. The method of claim 17, The epsilon value is always smaller than 1 / CW _RT value when CW _RT is the first contention window value.