JP5720008B2 - Wireless device - Google Patents

Description

  The present invention relates to a wireless device.

  There is known an EDCF (Enhanced Distributed Coordination Function) system in which traffic is classified into access categories having different priorities depending on the type, and priority control is performed by providing parameters such as contention windows that differ for each access category ( Non-patent document 1).

  In addition, in a DCF channel access method that does not perform priority control, a method for maximizing effective use of a channel is known (Non-Patent Document 2).

  Furthermore, a method for maximizing effective use of network channels based on the EDCF method is also known (Non-patent Document 3).

IEEE 802.11 Std, 802.11e-2005, Part 11, Wireless LAN Medium Access Control (MAC) and Physical layer (PHY) Specifications: Amendments 8: Medium Access Control (MAC) Quality of Service Enhancements, 2005. G. Bianchi, “Performance Analysis of the IEEE 802.11 Distributed Coordination Function,” IEEE Journal on Selected Areas in Comm., Vol. 18, no. 3, pp. 535-547, 2000. C. Hu and J. C. Hou, “A Novel Approach to Contention Control in IEEE 802.11e-Oriented WLANs,” IEEE INFOCOM 2007, pp. 1190-1198, 2007.

  However, the method disclosed in Non-Patent Document 1 has a problem that as the number of terminals increases, packet collisions frequently occur and the throughput of the entire network deteriorates. In addition, there is a problem that priority control does not operate properly when there is a bias in the number of terminals that transmit packets between access categories.

  In addition, the method disclosed in Non-Patent Document 2 has a problem that priority control is not performed.

  Furthermore, the method disclosed in Non-Patent Document 3 has a problem that priority control does not operate properly.

  Therefore, the present invention has been made to solve such a problem, and the object of the present invention is to suppress a decrease in throughput even when there is a bias in the number of terminals between access categories, and to accurately perform priority control. It is to provide a wireless device capable of performing the above.

  According to the embodiment of the present invention, the wireless device includes a detection unit, a calculation unit, and a communication unit. The detecting means detects the number of terminals and the channel occupation ratio in each of a plurality of access categories having different priorities. The computing means detects the detected channel more than the first reference value indicating that the first band actually used in one access category is smaller than the second band assigned to one access category. When the occupancy is small, the first contention window, which becomes larger as the number of detected terminals increases and is larger than the value at the previous packet transmission, is calculated, and the detected channel occupancy Is greater than the second reference value indicating that more packets than the number of packets that can be transmitted using the first band are generated, and increases as the number of detected terminals increases, and The first calculation process for calculating the second contention window having a value smaller than the value at the previous packet transmission is executed for all of the plurality of access categories. The communication means transmits and receives packets using the first contention window or the second contention window in each of the plurality of access categories.

  In the radio apparatus according to the embodiment of the present invention, when the channel occupancy is smaller than the first reference value, the contention window of one or more access categories is enlarged and an unused band is allocated to another access category. . Also, when the channel occupancy is equal to or higher than the second reference value, the contention window of one or more access categories is reduced and packets are transmitted according to the correct priority control. The contention window is calculated so as to increase as the number of terminals in each access category increases, and to decrease as the number of terminals in each access category decreases.

  Therefore, even if the number of terminals is uneven between access categories, it is possible to improve throughput while accurately performing priority control.

1 is a schematic diagram of a wireless device according to an embodiment of the present invention. It is a conceptual diagram of the time required for transmission of a single packet. 3 is a flowchart for explaining a communication method in the wireless device shown in FIG. 1. It is a flowchart for demonstrating the detailed operation | movement of step S3 shown in FIG. It is a flowchart for demonstrating other detailed operation | movement of step S3 shown in FIG. It is the schematic of the radio | wireless network in embodiment of this invention. It is a figure which shows the relationship between the total throughput when each access category is saturated, and the number of terminals in each access category. It is a figure which shows the throughput for every access category when the conventional EDCA system is used. It is a figure which shows the throughput for every access category when the system by embodiment of this invention is used. It is a figure which shows the total throughput when arbitrary access categories are non-saturated. It is a figure which shows the throughput of each access category at the time of applying the system by embodiment of this invention, a data generation rate, and bandwidth allocation amount.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals and description thereof will not be repeated.

  FIG. 1 is a schematic diagram of a radio apparatus according to an embodiment of the present invention. Referring to FIG. 1, a radio apparatus 10 according to an embodiment of the present invention includes an antenna 1, a communication unit 2, a detection unit 3, a control unit 4, buffers 5 to 8, a classification unit 9, and an application. 11 and admission control means 12.

The communication unit 2 takes out packets belonging to the access categories AC _{0 to} AC ₃ having different priorities from the buffers 5 to 8 according to the priorities. The communication unit 2 receives the contention window CW _i in each access category AC _i (i = 0, 1, 2, 3) from the control unit 4. Then, the communication unit 2 accesses the channel in each access category AC _i using the contention window CW _i and transmits the packet extracted from the buffers 5 to 8 via the antenna 1.

  The communication unit 2 receives a packet via the antenna 1 and outputs the received packet to the application 11.

The detecting means 3 detects the number of terminals N _i that transmit and receive packets in each access category AC _i based on the packets that the communication means 2 transmits and receives. Since the packet includes a symbol indicating each access category AC _i and a transmission source address, the detection unit 3 detects the symbol indicating each access category AC _i of the packet transmitted and received by the communication unit 2 and the transmission source address. by, it is possible to detect the number of terminals _{N i.}

Further, the detection unit 3 detects the number of packets transmitted / received per unit time by the communication unit 2 as the use band UB _i in each access category AC _i .

Furthermore, the detection means 3 detects the average payload length P _i in each access category AC _{i and} the average payload length P of the entire network based on the packet transmitted and received by the communication means 2.

Further, the detecting means 3 measures the time for the communication means 2 to access the channel, thereby obtaining a channel occupancy ratio (ATR: Air Time Ratio) indicating a time ratio that the channel is busy or in a sensing state in unit time. To detect. In this case, the time for accessing the channel differs depending on each access category AC _{0 to} AC ₃ . Accordingly, the access categories AC _{0 to} AC ₃ are access categories having different priorities.

Then, the detection unit 3 outputs the number of terminals N _i , the used band UB _i , the average payload length P _i , the average payload length P, and the channel occupation rate ATR to the control unit 4.

The control unit 4 receives the number of terminals N _i , the used band UB _i , the average payload length P _i , the average payload length P, and the channel occupation rate ATR from the detection unit 3. Then, the control means 4 determines the communication band in each access category AC _i by a method described later based on the number of terminals N _i , the used band UB _i , the average payload length P _i , the average payload length P and the channel occupancy ATR. effectively utilize and calculates a contention window CW _i in each access category AC _i to the priority control is performed accurately, and outputs the calculated contention window CW _i to the communication means 2.

Further, in response to a request from the admission control unit 12, the control unit 4 calculates the allocatable bandwidth RSVB _i in each access category AC _i by a method to be described later, and performs the admission control on the calculated allocatable bandwidth RSVB _i. Output to means 12.

Buffers 5 to 8 are provided corresponding to access categories AC _{0 to} AC ₃ , respectively. The access category AC ₀ is, for example, a background access category, the access category AC ₁ is, for example, a best-effort access category, and the access category AC ₂ is, for example, a video access category, and the access category AC ₃ is, for example, an audio access category. Access category AC ₃ has the highest priority, access category AC ₂ has the second highest priority, access category AC ₁ has the third highest priority, and access category AC ₀ has the lowest priority.

Buffers 5 to 8 receive packets belonging to access categories AC _{0 to} AC ₃ from classification means 9 and temporarily store the received packets. Then, the buffers 5 to 8 output the stored packet to the communication unit 2 in response to a request from the communication unit 2.

The classification means 9 receives a packet from the application 11 and classifies the received packet into access categories AC _{0 to} AC ₃ . Then, the classification means 9 stores the packets belonging to the access categories AC _{0 to} AC ₃ in the buffers 5 to 8, respectively.

  The application 11 generates a packet. When the application 11 is permitted to transmit a packet in accordance with self-led admission control or admission control driven by the admission control means 12, the application 11 outputs the generated packet to the classification means 9. The application 11 receives a packet from the communication unit 2.

  The application 11-led admission control and the admission control means 12-led admission control will be described later.

The admission control means 12 is provided in the MAC layer and performs self-led admission control to be described later. In this case, the admission control means 12 inquires the control means 4 about the assignable bandwidth RSVB _i in each access category AC _i and receives the assignable bandwidth RSVB _i from the control means 4.

[Control of contention window]
It describes a method of controlling the contention window CW _i in each access category AC _i.

The control means 4 calculates the contention window CW _i in each access category AC _i so that the total throughput is maximized and the priority control between the access categories AC _i is performed simultaneously.

The priority control is performed based on the bandwidth ratio F _i determined in each access category AC _i .

The bandwidth ratio F _i is determined by a fixed value method or a variable value method. When the bandwidth ratio F _i is determined by the fixed value method, the control unit 4 holds the bandwidth ratio F _i in each access category AC _i in advance.

For example, to send and receive packets belonging to the access categories _AC 1 to Ac _3, the band ratio _{F 1} in the access category AC ₁ is previously determined to be 1/7, the band ratio _{F 2} in the access category AC ₂ is 2 / 7 and the bandwidth ratio F ₃ in the access category AC ₃ is predetermined as 4/7.

Thus, the band ratios F _{1 to} F ₃ are determined in advance according to the priorities of the access categories AC _{1 to} AC ₃ . That is, the band ratios F _{1 to} F ₃ are determined in advance so as to increase as the priority increases in the order of the access category AC ₁ , the access category AC ₂ , and the access category AC ₃ .

When the bandwidth ratio F _i is determined by the variable value method, the control unit 4 classifies the access categories AC _{0 to} AC ₃ into an access category RAC that can secure the bandwidth and an access category NAC that cannot secure the bandwidth. . Then, the control means 4 calculates the band ratio F _i according to the following equation at the start of the application.

Control means 4, when calculating the percentage of bandwidth _{F i} at the access category AC _i classified into access categories RAC, receives the requested bandwidth RB from the application 11, the requested bandwidth RB thereof received, the entire network and bandwidth Capacity The bandwidth ratio F _i is calculated by substituting into the upper equation of equation (1). The control means 4 holds the bandwidth capacity of the entire network in advance.

Further, the control means 4 calculates the bandwidth ratio F _i in the access category AC _i classified into the access category NAC by the lower expression of the expression (1). That is, the control unit 4 calculates the band ratio F _i in the access category AC _i classified into the access category NAC by allocating the remaining band not allocated to the access category RAC. Note that FNAC _i represents a band ratio in the access category NAC and is determined in advance. Therefore, the control means 4 holds FNAC _i in advance.

Thus, the control means 4 determines the band ratio F _i in each access category AC _i using a fixed value method or a variable value method.

Control means 4, when determining the bandwidth ratio F _i at each access category AC _i, a band ratio F _i that the determined and the number of terminals N _i in each access category AC _i is substituted into the following equation access category AC _i The relative channel access frequency ρ _i for a single terminal is calculated.

In Equation (2), F _k is the bandwidth ratio in each access category AC _k (k = 0, 1, 2, 3), and N _k is the number of terminals in each access category AC _k . In equation (2), MIN (F _k / N _k ) is used as the denominator to represent the channel access frequency for the narrowest band per terminal (= the smallest number of channel accesses per terminal). That is, among the access categories AC _{0 to} AC ₃ , MIN (F _k / N _{k is used} to represent the channel access count per single terminal in each access category AC _i with the smallest channel access count per terminal as a reference value. ) Was used as the denominator of equation (2).

Furthermore, α _i in equation (2) is a parameter controlled by the channel usage status.

When the control unit 4 calculates the relative channel access frequency ρ _i of a single terminal using the equation (2), the control unit 4 calculates the calculated relative channel access frequency ρ _i and the number of terminals N _{i for} each access category AC _i as follows. The virtual terminal number VN is calculated by substituting it into the equation.

In equation (3), m represents the number of access categories AC _i , and m = 4 in the embodiment of the present invention.

This number VN of virtual terminals represents the number of terminals accessing all channels in all access categories AC _i .

When calculating the number VN of virtual terminals using equation (3), the control means 4 substitutes the calculated number VN of virtual terminals and the average time T required for transmission of a single packet into the following equation. calculating a reference value CWref of CW _i.

  In Equation (4), σ is a slot length and is determined in advance. Therefore, the control means 4 holds the slot length σ in advance.

  FIG. 2 is a conceptual diagram of the time required to transmit a single packet. Referring to FIG. 2, average time T is the sum of the transmission times of data packets and ACK packets and times AIFS and SIFS defined by MAC. When RTS / CTS handshaking is performed, TTS and CTS transmission times are also included in T.

  The data frame includes a preamble, a MAC header, a packet, and an FCS. The preamble (Preamble) is transmitted at a fixed rate, and the other parts are transmitted at an arbitrary data rate. In a system that performs adaptive rate control, the data rate is a rate determined according to link quality or the like.

  On the other hand, the ACK frame is also composed of a preamble and an ACK body (Addr etc. and FCS). The preamble (Preamble) is transmitted at a fixed rate, and the ACK body is transmitted at an arbitrary data rate.

  When the transmission rate of a preamble or the like is TxPhysRate, the transmission rate of the data frame is TxDataRate, and the transmission rate of the ACK frame is TxACKRate, the time T is expressed by the following equation.

In Expression (5), L _H is the length of the MAC header, and P is the length of the packet.

In _{general, TxPhyRate, TxACKRate, PrLength, L} H, FCS and ACK takes a fixed value. Accordingly, the wireless device 10 can easily calculate the average time T by monitoring the data length (P) and TxDataRate at the time of frame transmission / reception.

  The control means 4 calculates the average time T using the equation (5), and calculates the reference value CWref by substituting the calculated average time T and the virtual terminal number VN into the equation (4).

Then, the control means 4 calculates the contention window CW _i in each access category AC _i by substituting the reference value CWref and the relative channel access frequency ρ _i calculated using Expression (2) into the following expression.

In Equation (6), _nk is the number of terminals in each access category AC _i .

Here, the parameter α _i in Expression (2) will be described. In all the access categories _AC 0 to Ac _3, when the packet to be transmitted is always present, i.e., when all the access categories _AC 0 to Ac ₃ is saturated, alpha _i of formula (2) is, alpha _i = Set to 1. As a result, the channel is most effectively used, and priority control between the access categories AC _i is realized.

On the other hand, when an arbitrary access category AC _i is in a non-saturated state (that is, when the traffic generation amount is equal to or less than the bandwidth ratio F _i ), the channel may not be used most effectively. This is because even if the other access category AC _j is saturated, a part of the allocated band is vacated due to the non-saturated access category AC _i . In such a case, the channel is maximized by giving only the necessary bandwidth to the non-saturated access category AC _i and letting the remaining bandwidth be used for other access categories AC _j (= saturated access category). It is used effectively.

Therefore, in the embodiment of the present invention, the channel is effectively used to the maximum extent by controlling α _i in the equation (2).

Here, before describing a specific control method of α _i , a calculation method of the band OB _i allocated to each access category AC _i will be described.

The band OB _i is calculated by the following expression using the channel capacity Capacity and the band ratio F _i .

In Expression (7), K is a constant that satisfies K = (T / (2σ)) ^1/2 .

The control means 4 holds a constant K in advance. Then, the control means 4 receives the average payload length P received from the detection means 3, the slot length σ and constant K held in advance, the band ratio F _i calculated using the expression (1), the expression (5) ) Is used to calculate the allocated bandwidth OB _i by substituting the average time T calculated using) into equation (7).

A specific method for controlling α _i will be described. Control means 4 determines the alpha _i for each access category AC _i based on the terminal number N _i and the channel occupancy ATR periodic or each access category AC _i with Packet.

Control means 4, if the number of terminals _{N i} of any access category AC _i is varied (vertically) when determining the alpha _i, sets the alpha _i for each access category AC _i to α _i = 1. This is because the channel state is considered to change due to the change in the network configuration (number of terminals), and the contention window CW _i is determined based only on the number of terminals N _i .

On the other hand, if at least one of the number of terminals _N 0 to N ₃ in the access category _AC 0 to Ac ₃ is not changed, the control unit 4 determines the alpha _i based on the channel occupancy ATR. More specifically, when the channel occupancy ATR is smaller than the reference value ATRmin, the control means 4 determines that the channel is not being used effectively, and the band OB allocated by the actually used band UB _i for small access category AC _i than _i, by setting the composed alpha _i from a value smaller than "1", to reduce the bandwidth to be allocated. That is, the control unit 4 determines the alpha _i by _{α i = MIN (UB i /} OB i, 1). In this case, since the band UB _i is smaller than the band OB _i , α _i has a value smaller than “1”.

The reference value ATRmin is a reference value indicating that the band UB _i actually used in one access category is smaller than the band OB _i allocated to one access category. The reference value ATRmin is, for example, 80%.

On the other hand, when the channel occupancy ATR is larger than the reference value ATRmax (ATRmax> ATRmin), the control unit 4 increases the value of α _i for the access category AC _i where UB _i <OB _i . That is, the control unit 4 determines the alpha _i by _{α i = MIN (C × UB} i / OB i, 1). The constant C is a constant larger than “1”. α _i is determined by α _i = MIN (C × UB _i / OB _i , 1) because the traffic volume of the access category AC _i that was not saturated at the previous setting of α _i increased. This is so that correct priority control is performed correspondingly.

ATRmax is a reference value indicating that more packets than the number of packets that can be transmitted using the band UB _i actually used are generated, and is set to 88%, for example.

[Admission Control]
The bandwidth allocated to one wireless device is OB _i / N _i . Therefore, as the number of terminals N _i increases, the bandwidth used by a single wireless device decreases. Therefore, when the access category AC _i is real-time traffic with strict quality requirements such as voice and moving images, it is expected that the total traffic cannot satisfy the quality requirement when the number of terminals N _i increases.

Accordingly, in embodiments of the present invention, by controlling the number of terminals N _i, to ensure the quality of the application. More specifically, when P-length data packets are generated at I time intervals, a P / I bandwidth is required. Therefore, in order to satisfy the quality requirements of the application that requires the P / I bandwidth and to prevent the quality of other applications already communicating in the same access category from degrading, the allocatable bandwidth RSVB _i is the required bandwidth P / I. The application needs to be started only when it is larger than (RSVB _i > P / I).

When the bandwidth allocation F _i is a fixed value, the control means 4 calculates the allocatable bandwidth RSVB _{i according} to equation (8). When the bandwidth allocation F _i is a variation value, the control means 4 assigns the allocatable bandwidth RSVB according to equation (9). _i is calculated.

(I) Admission control driven by the MAC layer When the admission control is performed by the MAC layer, the application 11 calculates the requested bandwidth P / I of the access category AC _i and adds the calculated requested bandwidth P / I. Output to the mission control means 12. When the admission control unit 12 receives the request band P / I from the application 11, the admission control unit 12 outputs an inquiry signal of the allocatable band RSVB _i to the control unit 4. The control means 4 calculates the allocatable bandwidth RSVB _i using the formula (8) or formula (9) according to the inquiry signal, and outputs the calculated allocatable bandwidth RSVB _i to the admission control means 12. .

Upon receiving the allocatable bandwidth RSVB _i from the control unit 4, the admission control unit 12 determines whether or not RSVB _i > P / I. When the admission control unit 12 determines that RSVB _i > P / I, the admission control unit 12 generates a permission signal and outputs the permission signal to the application 11. On the other hand, when it is determined that RSVB _i > P / I, the admission control means 12 generates a rejection signal and outputs it to the application 11.

When the application 11 receives the permission signal from the admission control unit 12, the application 11 generates a packet belonging to the access category AC _i and outputs it to the classification unit 9.

On the other hand, when the application 11 receives the rejection signal from the admission control means 12, the application 11 stops communication of packets belonging to the access category AC _i or starts transmission in the access category NAC that does not require admission control.

(II) Application-driven admission control When the application-driven admission control is performed, the application 11 outputs an inquiry signal of the allocatable bandwidth RSVB _i to the admission control means 12. When the admission control means 12 receives the inquiry signal of the assignable band RSVB _i from the application 11, the admission control means 12 receives the assignable band RSVB _i from the control means 4 by the method described above. Then, the admission control means 12 outputs the received allocatable bandwidth RSVB _i to the application 11.

Thereafter, when the application 11 receives the allocatable bandwidth RSVB _i from the admission control means 12, the application 11 calculates the requested bandwidth P / I of the access category AC _i and determines whether RSVB _i > P / I. When the application 11 determines that RSVB _i > P / I, the application 11 generates a packet belonging to the access category AC _i and outputs the packet to the classification unit 9.

On the other hand, when it is determined that RSVB _i > P / I, the application 11 stops communication of packets belonging to the access category AC _i or starts transmission in the access category NAC that does not require admission control.

In both of the MAC-driven admission control and the application-driven admission control described above, when it is determined that RSVB _i > P / I, the application 11 generates a packet and outputs the packet to the classification unit 9. communication means 2 accesses the channel using a contention window CW _i received from the control unit 4, and transmits the taken out packet from the buffer 5-8. That is, when it is determined that RSVB _i > P / I, the communication unit 2 uses the contention window CW _i received from the control unit 4 to access the channel, and extracts and transmits the packet from the buffers 5 to 8. To do.

Accordingly, the communication means 2 generally performs communication processing for transmitting and receiving packets using the contention window CW _i when the allocatable bandwidth RSVB _i is larger than the traffic request bandwidth P / I. Run for everything.

  FIG. 3 is a flowchart for explaining a communication method in radio apparatus 10 shown in FIG.

Referring to FIG. 3, when the communication operation in radio apparatus 10 is started, assignable band RSVB _i and requested band P / I of access category AC _i are calculated (step S1).

Then, it is determined whether RSVB _i > P / I (step S2). When it is determined in step S2 that RSVB _i > P / I is not satisfied, the series of operations ends.

On the other hand, when it is determined in step S2 that RSVB _i > P / I, the control unit 4 of the wireless device 10 determines all α _{0 to} α ₃ of all access categories AC _{0 to} AC ₃ ( Step S3).

Then, the control unit 4 calculates the bandwidth ratio _F 0 to F ₃ access categories _AC 0 to Ac ₃ by the above-described method. Thereafter, the control means 4 calculates the relative channel access frequencies ρ _{0 to} ρ ₃ by substituting the bandwidth ratios F _{0 to} F ₃ , α _{0 to} α ₃ , and the number of terminals N _{0 to} N ₃ into equation (2). .

Then, the control means 4 substitutes the calculated relative channel access frequencies ρ _{0 to} ρ ₃ , average time T, slot length σ, and number of terminals _nk into the equation (6), and sets the access categories AC _{0 to} AC ₃ . The contention windows CW _{0 to} CW ₃ are calculated. That is, the control unit 4 calculates the contention window _CW 0 _{~CW 3} access categories _AC 0 to Ac ₃ with alpha ₀ to? _3. (Step S4).

Then, the control means 4 outputs the calculated contention windows CW _{0 to} CW ₃ to the communication means 2, and the communication means 2 uses the contention windows CW _{0 to} CW ₃ received from the control means 4 to channel. Access is performed, and packets belonging to the access categories AC _{0 to} AC ₃ are extracted from the buffers 5 to 8 and transmitted (step S5). As a result, the series of operations ends.

  Steps S1 and S2 are executed by either the above-described MAC layer-driven admission control or application-driven admission control.

Also, the flowchart shown in FIG. 3 is executed periodically or at the time of packet transmission for all of the access categories AC _{0 to} AC ₃ .

  FIG. 4 is a flowchart for explaining the detailed operation of step S3 shown in FIG.

Referring to FIG. 4, if it is determined in step S2 shown in FIG. 3 that RSVB _i > P / I, control means 4 has a change in the number N _i of terminals in any access category AC _i . It is determined whether or not (step S31).

When it is determined in step S31 that the number of terminals N _i in any access category AC _i is varied, the control unit 4 sets α ₀ = α ₁ = α ₂ = α ₃ = 1 (step S32). .

On the other hand, when it is determined in step S31 that the number of terminals N _i in any access category AC _i does not vary, the control means 4 further determines whether or not the channel occupancy ATR is smaller than the reference value ATRmin. (Step S33).

When it is determined in step S33 that the channel occupancy ATR is smaller than the reference value ATRmin, the control means 4 determines α _{0 to} α _{3 according} to α _i = MIN (UB _i / OB _i , 1) (step S33). S34).

  On the other hand, when it is determined in step S33 that the channel occupancy ATR is not smaller than the reference value ATRmin, the control means 4 further determines whether or not the channel occupancy ATR is greater than or equal to the reference value ATRmax (step S35). ).

When it is determined in step S35 that the channel occupancy ATR is equal to or greater than the reference value ATRmax, the control unit 4 determines that α _{i to} MIN (C × UB _i / OB _i , 1), C> 1, α _{0 to} α. ₃ is determined (step S36).

On the other hand, when it is determined in step S35 that the channel occupancy ATR is not greater than or equal to the reference value ATRmax, the control means 4 maintains the previously determined value without changing the values of α _{0 to} α ₃ (step S35). S37).

  And after any of step S32, S34, S36, S37, a series of operation | movement transfers to step S4 shown in FIG.

In step S31, when the wireless device 10 is activated, it is determined that there is a change in the number of terminals N _i of any access category AC _i . Therefore, when the flowchart shown in FIG. 4 is executed first, α _i is usually set to “1”. In addition, as long as it is determined in step S31 that the number of terminals N _i in any access category AC _i is varied, α _{0 to} α ₃ are set to “1”.

This is because, as described above, it is considered that the channel state by variations in the configuration of a network including a wireless device 10 (the number of terminals) is changed, the contention window CW ₀ ~CW ₃ in order to determine by only the number of terminals.

In step S31, when it is determined that there is no change in the number of terminals N _i of any access category AC _i , α _{0 to} α ₃ are determined based on the channel occupancy ATR. In this case, when the channel occupation ratio ATR is smaller than the reference value ATRmin, α _{0 to} α ₃ are determined by α _i = MIN (UB _i / OB _i , 1) (see step S34).

Since the reference value ATRmin indicates that the band UB _i actually used in one access category is smaller than the band OB _i allocated to one access category, as described above, ATR <ATRmin is established. In this case, UB _i <OB _i . Therefore, UB _i / OB _i is smaller than “1”, and α _i (i = 0 to 3) is set to a value smaller than “1”. As a result, the relative channel access frequency ρ _i (i = 0 to 3) calculated by the equation (2) becomes smaller than the previous value, and the contention window CW _i (i = 0 to 3) Larger than the value. Therefore, the rate of accessing the channel decreases when packets belonging to the access category AC _i (i = 0 to 3) are transmitted. Of the bands OB _i assigned to the access category AC _i , bands that are not actually used are assigned to other access categories. As a result, throughput can be improved.

If the contention window _CW 0 _{~CW 3} is calculated using the alpha ₀ to? ₃ determined in step S34, the contention window _CW 0 _{~CW 3,} the terminal number _{n k} of the access categories _AC 0 to Ac ₃ There increased if increased, the number of terminals _{n k} access categories _AC 0 to Ac ₃ is calculated to be smaller if reduction (see equation (6)). Then, the collision of packets is suppressed if large contention window _CW i (i = _0~3), contention window _CW i (i = _0~3) is increased the smaller. Therefore, the calculated contention window _CW 0 _{~CW 3} with alpha ₀ to? ₃ determined in step S34, to transmit and receive packets using a contention window _CW 0 _{~CW 3} that its operation is wireless This corresponds to suppressing packet collision due to an increase in the number of devices (number of terminals) and improving throughput.

Further, as described above, ATRmax is a reference value indicating that more packets are generated than the number of packets that can be transmitted using the band UB _i actually used. Therefore, when it is determined in step S35 that ATR ≧ ATRmax, α _i = MIN (C × UB _i / OB _i , 1), and C> 1, the values of α _{0 to} α ₃ are changed from the previous values. Increase (see step S36). As a result, the contention windows CW _{0 to} CW ₃ become smaller than the previous value, and more generated packets can be transmitted. That is, correct priority control corresponding to the fluctuation (generally, increase) of traffic in the access categories AC _{0 to} AC ₃ that has been in a non-saturated state by the previous setting of α _{0 to} α ₃ is performed. .

Therefore, the calculated contention window _CW 0 _{~CW 3} with alpha ₀ to? ₃ determined in step S36, to transmit and receive packets using a contention window _CW 0 _{~CW 3} that its operation is preferentially This is equivalent to improving the throughput while accurately performing the degree control.

Furthermore, when it is determined in step S35 that ATR ≧ ATRmax is not satisfied, that is, when ATRmin ≦ ATR <ATRmax, α _{0 to} α ₃ are not changed (see step S37). In this case, it is determined that the channel is sufficiently effectively used and priority control is also performed.

  FIG. 5 is a flowchart for explaining another detailed operation of step S3 shown in FIG.

  The flowchart shown in FIG. 5 is the same as the flowchart shown in FIG. 4 except that steps S31 and S32 in the flowchart shown in FIG. 4 are replaced with step S30.

Referring to FIG. 5, when it is determined in step S2 shown in FIG. 3 that RSVB _i > P / I, control means 4 sets α ₀ = α ₁ = α ₂ = α ₃ = 1. (Step S30). Thereafter, Steps S33 to S37 described above are executed.

As described above, step S32 shown in FIG. 4, in step S31, it is performed when it is determined that there is a variation in the number of terminals _{N i} of any access category _{AC i, α 0 = α 1} = α 2 = α ₃ = 1 is set. Then, when the wireless device 10 is activated, it is determined that there is a change in the number N _i of terminals in any access category AC _i .

Therefore, in the flowchart shown in FIG. 5, in step S30, α ₀ = α ₁ = α ₂ = α ₃ = 1 is set, and α _i is controlled based on the channel occupation ratio ATR.

Even if α _i is determined according to the flowchart shown in FIG. 5, when ATR <ATRmin, step S34 described above is executed, and when ATR ≧ ATRmax, step S36 described above is executed. Therefore, when ATR <ATRmin, the throughput is improved while suppressing packet collision due to an increase in the number of wireless devices (number of terminals), and when ATR ≧ ATRmax, the throughput is improved while accurately performing priority control. To do.

Since the flowchart shown in FIG. 4 or FIG. 5 is executed for all of the access categories AC _{0 to} AC ₃ , even when the number of terminals N _i is biased among the access categories AC _i , the priority control is performed accurately. Throughput can be improved.

  FIG. 6 is a schematic diagram of a wireless network according to the embodiment of the present invention. With reference to FIG. 6A, the wireless network 100 includes an access point 110 and terminal devices 121 to 128.

Each of the access point 110 and the terminal devices 121 to 128 includes the wireless device 10 shown in FIG. Then, the access point 110 transmits and receives packets belonging to the access category AC _i between the terminal device 121 to 128 by the communication method in a radio apparatus 10 described above.

  Thus, the wireless network 100 is a centralized wireless network.

With reference to (b) of FIG. 6, the wireless network 200 includes terminal devices 131 to 138. Each of the terminal devices 131 to 138 includes the wireless device 10 shown in FIG. Then, the terminal device 131 to 138 transmits and receives packets belonging to the access category AC _i by a communication method in a radio apparatus 10 described above to each other.

  Thus, the wireless network 200 is a distributed wireless network.

Therefore, the wireless device 10 according to the embodiment of the present invention is applied to both the centralized wireless network 100 and the distributed wireless network 200. As a result, in the wireless networks 100 and 200, it is possible to improve the throughput while accurately performing the priority control of the access category AC _i .

When the transmission rate is 6 Mbps and the priorities of the access categories AC ₁ , AC ₂ , and AC ₃ are 1/7, 2/7, and 4/7 (fixed values), respectively, each system parameter and channel capacity (formula ( Table 1 shows the theoretical calculation results of the band (OB _i ) assigned to each access category AC _i in parentheses 7).

The purpose of this embodiment of the invention is to set the total throughput of the network to 3.8 Mbps (= channel capacity) by maximizing the effective efficiency of the channel, and each access category AC _i is saturated. , The throughputs of the access categories AC ₁ , AC ₂ , AC ₃ are set to 0.544 Mbps, 1.09 Mbps and 2.18 Mbps (= allocated bandwidth), respectively, without depending on the number of terminals in each access category AC _i It is.

Next, simulation results will be described. In the simulation, an access network composed of one access point and 30 terminal devices is assumed, and each terminal device transmits a frame in _{one of} the access categories AC ₁ , AC ₂ , and AC ₃ . Parameters such as payload length and transmission rate are as shown in Table 1.

  FIG. 7 is a diagram illustrating the relationship between the total throughput when each access category is saturated and the number of terminals in each access category.

In FIG. 7, the vertical axis represents the total throughput, and the horizontal axis represents the number of terminals in each access category AC ₁ , AC ₂ , AC ₃ . Curve k1 shows the relationship between the total throughput when the method according to the embodiment of the present invention is used and the number of terminals in each access category, and curve k2 shows the total throughput when the conventional EDCA method is used. The relationship with the number of terminals in each access category is shown.

  Referring to FIG. 7, it can be seen that when the method according to the embodiment of the present invention is used, the total throughput is about 4 Mbps, and the channel is effectively used (see curve k1).

  On the other hand, the total throughput when using the conventional EDCA method is significantly lower than the channel capacity (= 3.81 Mbps).

  FIG. 8 is a diagram showing the throughput for each access category when the conventional EDCA method is used. FIG. 9 is a diagram showing the throughput for each access category when the method according to the embodiment of the present invention is used.

8 and 9, the vertical axis represents the throughput, and the horizontal axis represents the number of terminals in each access category AC ₁ , AC ₂ , AC ₃ .

Curve k3 shows the throughput of access category AC ₁ when using the conventional EDCA method, curve k4 shows the throughput of access category AC ₂ when using the conventional EDCA method, and curve k5 shows The throughput of access category AC ₃ when using the conventional EDCA method is shown.

Further, curve k6 shows the throughput of access category AC ₁ when using the scheme according to the embodiment of the present invention, and curve k7 shows the access category AC ₂ when using the scheme according to the embodiment of the present invention. It shows the throughput curve k8 shows the throughput of the access categories AC ₃ when using scheme according to the present invention.

Referring to FIG. 8, when the conventional EDCA scheme is used, the throughput obtained by any access category AC _i greatly depends on the number of terminals in that access category AC _i and other access categories AC _j (i ≠ j). However, an access category having a large number of terminal devices tends to obtain a high throughput regardless of its priority.

Referring to FIG. 9, when the system according to the embodiment of the present invention is used, the bandwidth according to the priority setting is set independently of the number of terminals in each access category AC ₁ , AC ₂ , AC ₃ . It can be seen that access categories AC ₁ , AC ₂ , and AC ₃ are assigned. The throughputs of the access categories AC ₁ , AC ₂ , and AC ₃ are about 0.5 Mbps, 1.1 Mbps, and 2.2 Mbps, respectively (see curves k3 to k5), which are as calculated theoretically.

Next, a case where any access category is non-saturated will be described. Specifically, the number of terminals in the access categories AC ₁ , AC ₂ , AC ₃ is fixed to ₃ , ₃ and 24, respectively, and the data generation rate of a single terminal in the access category AC ₁ and the access category AC ₂ is 1 Mbps, It was set to 216 kbps.

Since there are three terminals in each of the access categories AC ₁ and AC ₂ , the total data generation rate of the access category AC ₁ is 3 Mbps, which greatly exceeds the bandwidth (= 0.5 Mbps) allocated by the MAC. .

On the other hand, the total data generation rate of the access category AC ₂ is 648 kbps, which is lower than the bandwidth (= 1.1 Mbps) allocated by the MAC.

  FIG. 10 is a diagram illustrating the total throughput when an arbitrary access category is in a non-saturated state.

10, the vertical axis represents the total throughput, the horizontal axis represents the traffic generation rate in the access category AC _3. A curve k9 indicates the total throughput when the method according to the embodiment of the present invention is used, and a curve k10 indicates the total throughput when the conventional EDCA method is used.

Referring to FIG. 10, when the conventional EDCA method is used, the total throughput increases with a decrease in the traffic generation rate in the access category AC _3, but overall, the method according to the embodiment of the present invention is used. Lower than when used (see curve k10).

On the other hand, when the method according to the embodiment of the present invention is used, the total throughput does not depend on the traffic generation rate in the access category AC ₃ and maximizes the throughput of the entire network (that is, the channel capacity is 3.81 Mbps). Match) (see curve k9).

FIG. 11 is a diagram showing the throughput, data generation rate, and bandwidth allocation amount for each of the access categories AC ₁ , AC ₂ , AC ₃ when the system according to the embodiment of the present invention is applied.

11, the vertical axis represents the throughput, the horizontal axis represents the traffic generation rate in the access category AC _3.

Referring to FIG. 11, it can be divided into two areas with different access category saturation conditions. On the left side is an area where the access category AC ₂ is in a non-saturated state RB ₂ <OB ₂ (AC ₁ and AC ₃ are in a saturated state RB> OB), and on the right side, the access category AC ₂ and the access category AC ₃ are non-saturated This is the case.

In the left area of FIG. 11, the unused portion of the band (OB ₂ ) allocated to the access category AC ₂ is used in the access category AC ₁ and the access category AC ₃ . In the right area, the unused band (OB ₂ , OB ₃ ) allocated to the access category AC ₂ and the access category AC ₃ is used in the access category AC ₁ . As a result, it has been found that the total throughput is always high as shown in FIG.

As described above, when the channel occupation ratio ATR in each access category AC _i is smaller than the reference value ATRmin, the number of terminals is set by setting α _i to a value smaller than “1” reflecting the number of terminals N _i. When throughput is improved while suppressing packet collision due to an increase in N _i and the channel occupancy ATR is equal to or greater than the reference value ATRmax, by increasing α _i reflecting the number of terminals N _i , The throughput is improved while correct priority control is performed in response to the increase in the access category AC _i .

Therefore, even when there is a deviation in the number of terminals between access categories AC _i, it is possible to accurately control priority while suppressing a decrease in throughput.

  In the above description, ATRmin is 80%. In the embodiment of the present invention, the ATRmin is not limited to this, and ATRmin may be a value other than 80%.

  In the above description, ATRmax is 88%. In the embodiment of the present invention, the ATRmax is not limited to this, and ATRmax may be a value other than 88%.

  Furthermore, in the above description, background, best effort, video, and audio have been described as access categories having different priorities. However, the present invention is not limited to this, and any access category having a different priority may be used. Any access category may be used.

  Further, Steps S33 to S37 shown in FIG. 4 or FIG. 5 and Step S4 shown in FIG. 3 constitute “first arithmetic processing”. Steps S31 and S32 shown in FIG. 4 and Step S4 shown in FIG. The “second arithmetic processing” is configured.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is shown not by the above description of the embodiments but by the scope of claims for patent, and is intended to include meanings equivalent to the scope of claims for patent and all modifications within the scope.

  The present invention is applicable to a wireless device.

  DESCRIPTION OF SYMBOLS 1 Antenna 2 Communication means 3 Detection means 4 Control means 5-8 Buffer 9 Classification means 10 Radio apparatus 11 Application 12 Admission control means 100, 200 Wireless network 110 Access point 121-128 131-138 Terminal equipment.

Claims (5)

Detecting means for detecting the number of terminals and channel occupancy in each of a plurality of access categories having different priorities;
The detected channel occupancy is less than a first reference value indicating that a first band actually used in one access category is smaller than a second band assigned to the one access category. When the number of detected terminals is small, the first contention window that increases with an increase in the number of detected terminals and has a value larger than the value at the previous packet transmission is calculated. When the number of detected terminals is equal to or larger than a second reference value indicating that more packets than the number of packets that can be transmitted using the first band are generated, the number of detected terminals increases. In addition, the first calculation processing for calculating the second contention window having a value smaller than the value at the previous packet transmission is applied to all of the plurality of access categories. Operating means for executing Te,
A wireless device comprising: a communication unit that transmits and receives a packet using each of the first contention window and the second contention window in each of the plurality of access categories.   The computing means increases with an increase in the number of detected terminals when the number of detected terminals varies from the number of terminals at the previous packet transmission in any of the plurality of access categories. And, a second calculation process for calculating a contention window so as to decrease as the number of detected terminals decreases is performed for all of the plurality of access categories, and in all of the plurality of access categories, When the detected number of terminals does not vary from the number of terminals at the time of previous packet transmission, the first arithmetic processing is executed for all of the plurality of access categories based on the detected channel occupancy rate. The wireless device according to claim 1.   The communication means uses the first contention window or the second contention window when an allocatable bandwidth obtained by subtracting the first bandwidth from the second bandwidth is larger than a traffic request bandwidth. The wireless device according to claim 1, wherein communication processing for transmitting and receiving packets is executed for all of the plurality of access categories. An application for generating and outputting the requested bandwidth in each of the plurality of access categories;
When it is arranged in the MAC layer and receives the requested bandwidth from the application, it is determined whether the allocatable bandwidth is larger than the requested bandwidth, and when the allocatable bandwidth is larger than the requested bandwidth, permission is granted. Admission control means for executing control processing for outputting a signal to the application for all of the plurality of access categories;
The wireless device according to claim 3, wherein the application generates the packet and outputs the packet to the communication unit when receiving the permission signal from the admission control unit in each of the plurality of access categories. In each of the plurality of access categories, an application that generates and outputs an inquiry signal that inquires about the allocatable bandwidth, and
An admission control unit that is arranged in the MAC layer and that executes a process of outputting the allocatable bandwidth to the application when the inquiry signal is received from the application, for all of the plurality of access categories;
In each of the plurality of access categories, the application receives the allocatable bandwidth from the admission control means, and generates the packet when the received allocatable bandwidth is larger than the requested bandwidth, and the communication means The wireless device according to claim 3, wherein the wireless device outputs to.

Images