JP2006020069A - Method and apparatus of radio communicating and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of radio communication which can autonomously realize the insurance of a radio resource necessary for each traffic in each radio communication apparatus. <P>SOLUTION: The radio communication apparatus includes (a) an arrival rate operation which obtains a data arrival rate according to a traffic, and (b) a radio resource control unit which controls autonomously the assigning amount of the radio resource corresponding to each traffic based on the obtained data arrival rate. Consequently, the radio resource necessary for each traffic can be autonomously assured. As a result, the influence of a delay and a jitter can be eliminated as much as possible. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

One aspect of the present invention relates to a wireless communication device constituting a wireless network. The wireless communication device is realized as a semiconductor chip, a processing board, a communication card, an adapter, and other circuit devices, and also as an electronic device on which the circuit device can be mounted or mounted.
The present invention is also realized as a wireless communication method. The present invention is also realized as a program for realizing a wireless communication method.

Today, wireless networks are rapidly spreading. Along with this, opportunities for transmission of moving image data and audio data are increasing. For this reason, establishment of QoS (Quality of Service) control technology according to the nature of these data is required. Note that most of the existing QoS control technologies employ a method of securing radio resources in advance along the communication path.
However, in the case of a wireless network, the communication state constantly changes due to the influence of fading and hidden terminals. For this reason, the amount of data that can be transmitted per unit time and the communication path frequently change. In such an unstable wireless environment, the resource reservation type QoS control technique used in the wired network does not function well.
Currently, there is IEEE 802.11e as one of QoS control technologies suitable for such wireless networks. IEEE 802.11e employs two QoS control technologies. One is EDCA (Enhanced
Distributed Channel Access), and one is HCCA (HCF Controlled Channel Access).
This IEEE802.11e prescribes a method of “how to request radio resources” but does not prescribe a method of “how much radio resources are allocated”.
For this reason, various proposals have been made for specific QoS control techniques. For example, there is Patent Document 1. Patent Document 1 discloses a technique for determining the communication line assignment order for each terminal and determining communication line (wireless resource) assignment according to the order.
JP 2002-112321 A

  However, Patent Document 1 employs a technique in which an access point and other specific devices centrally manage “communication lines”. For this reason, it cannot be applied to a wireless network that does not employ a centralized management method. Further, in the case of the centralized management method, the “communication line” that can be managed is limited to the range that can communicate with the access point.

The inventor proposes the following communication method based on the above fact recognition.
That is, a radio communication method is proposed that autonomously controls the allocated amount of radio resources corresponding to each traffic based on the data arrival rate for each traffic. Thereby, the radio | wireless resource required for each traffic can be ensured autonomously. As a result, the influence of delay and jitter can be eliminated as much as possible.
The “traffic” here can include not only the case where the final destination is the same, but also the case where traffic having the same relay destination is handled as a unit.
The data arrival rate for each traffic is preferably obtained for each service class. By controlling the amount of radio resources allocated for each service class, access control for each class can be realized.
In the control, it is desirable to determine a reference value for the allocation amount based on the data arrival rate for each traffic and to correct the reference value using the remaining packet length and the remaining radio resource amount for each traffic.
Here, it is desirable that the remaining packet length for each traffic is used for increasing the allocated amount, and the remaining radio resource amount is used for reducing the allocated amount.

  By adopting the communication method according to the invention, it is possible to autonomously secure radio resources necessary for each traffic in each radio communication device.

Hereinafter, exemplary embodiments of a communication apparatus that employs the communication method according to the invention will be described.
In addition, the well-known or well-known technique of the said technical field is applied to the part which is not illustrated or described in particular in this specification.
The embodiment described below is one embodiment of the present invention and is not limited thereto.

(1) Overview of Communication Method Hereinafter, an ad hoc network will be described. An ad hoc network refers to a wireless network composed only of wireless communication devices (hereinafter referred to as “nodes”) that do not require an access point.
Two types of beacons are defined in the MAC (Media Access Control) layer of the ad hoc network. There are two types: a normal beacon (hereinafter referred to as “regular beacon”) and an additional beacon (hereinafter referred to as “auxiliary beacon”).
A beacon is a packet that notifies the availability and presence of a wireless device.
A regular beacon is a beacon transmitted by each node constituting an ad hoc network.
On the other hand, an auxiliary beacon is a beacon transmitted separately from a regular beacon by a node that has acquired the right to transmit data with priority over surrounding nodes.
FIG. 1A shows an arrangement example of regular beacons 1 corresponding to nodes A, B, and C. FIG. 1B shows an arrangement example when the node A additionally transmits one auxiliary beacon 3.
Each node autonomously mediates the transmission timing of the auxiliary beacon so as not to collide with the transmission period of other beacons. The data is transmitted at an arbitrary timing within a certain period from the transmission of each beacon.
The node that has acquired the auxiliary beacon 3 can transmit data with priority over other nodes for a certain period of time after transmission of the auxiliary beacon. This period 3A is called a TGP (Transmission Guarantee Period) period. This TGP section is also called a TPP (Transmission Prioritized Period) section.
On the other hand, the period 3B other than the TGP section is FAP (Fairness Access).
Period) called a section. In the FAP section, each node competes for transmission opportunities fairly using CSMA (more precisely, PSMA).

In this embodiment, data with high priority is transmitted using a TGP section obtained by transmitting an auxiliary beacon.
In order to provide QoS control to the user or application, a mechanism for determining how many auxiliary beacons are allocated, a mechanism for determining the order in which packets are transmitted in the TGP period, and the like are required.
Hereinafter, examples of these mechanisms will be described. However, the embodiment does not provide narrowly defined QoS (absolute priority control with bandwidth reservation or the like). That is, transmission bandwidth, delay, jitter (delay fluctuation), etc. are not guaranteed.
Instead, the following embodiment provides a relative priority control called CoS (Class of Service). This provides a relatively higher quality service for traffic classified into the priority class than in the non-priority class.
In the following embodiment, a case where two types of classes are prepared will be described. The two types of classes are a priority class (TGP class) and a non-priority class (FAP class).
The priority class is a class that is preferentially transmitted using the TGP section.
On the other hand, the non-priority class refers to a class in which transmission opportunities are given fairly using the FGP section. The non-priority class is used when priority is not specified.
We propose a method that uses these two types of classes and does not guarantee service quality, but ensures a transmission band that can transmit data with as low delay and low jitter as possible. We also propose a method to realize CoS control in an autonomous and distributed manner.

(2) Node Configuration FIG. 2 shows a configuration example of each node configuring the ad hoc network. The node 11 includes a routing unit 13, a classification unit 15, an output waiting queue 17, a packet scheduler 19, a QoS management unit 21, a token generation unit 23, and an auxiliary beacon management unit 25 as main modules.
The routing unit 13 is a device that performs packet routing. There are three routing paths.
One is a path for outputting a packet input from the upper layer (application) 11A to an adjacent node. One is a path for outputting a packet received from an adjacent node to the upper layer (application) 11A. One is a path for transferring a packet received from an adjacent node to another adjacent node. An adjacent node refers to a node in a range where a beacon reaches.
In any case, the class (FAP class or TGP class) to which the packet belongs is specified before reaching the routing unit 13.
The routing unit 13 outputs the packet to the classification unit 15 when an adjacent node (hop destination) serving as a transfer destination is determined.
The classifying unit 15 is a device that distributes packets to the corresponding output waiting queue 17. In addition, the classification unit 15 is also used for measuring the data arrival rate and the like for each transfer destination (for each hop destination) of traffic of the TGP class. The various statistical values measured are to be referred to by the QoS management unit 21.

The output waiting queue 17 is a set of queues prepared for each transfer destination (for each hop destination) and for each class. In the case of FIG. 2, N queues are prepared for each transfer destination for each of the TGP class and the FAP class.
When the number of adjacent nodes is small with respect to the prepared N queues, some queues are not used.
The packet scheduler 19 is a device that extracts and transmits packets from each queue. Specifically, it is output to the physical layer.
The packet scheduler 19 employs a token bucket method in order to smooth the transmission rate. In other words, the packet scheduler 19 employs a technique in which tokens generated at a constant speed are stored in a token bucket, and packets are transmitted by the amount of the stored tokens.
The packet scheduler 19 reduces the number of tokens in the token bucket by the amount transmitted after the packet is transmitted.
The QoS management unit 21 is a device that determines the number of auxiliary beacons to be acquired using the traffic arrival rate and other values measured by the classification unit 15 and notifies the auxiliary beacon management unit 25 of the number.
Further, the QoS management unit 21 determines a token rate corresponding to each queue of the TGP class, and notifies the token generation unit 23 of the token rate. Here, the token rate refers to the generation speed of tokens accumulated in the token bucket.
The token generation unit 23 is a device that generates a token at a token rate specified for each queue. The generated tokens are stored in a token bucket prepared for each queue.
The auxiliary beacon management unit 25 is a device that secures the number of auxiliary beacons determined by the QoS management unit 21. When the auxiliary beacons are insufficient, the auxiliary beacon management unit 25 tries to acquire as many auxiliary beacons as possible. However, the auxiliary beacon management unit 25 does not particularly perform exception processing even if a shortage occurs.
As described above, the QoS management unit 21, the token generation unit 23, and the auxiliary beacon management unit 25 provide a function as a radio resource control unit. That is, it provides a function of autonomously acquiring auxiliary beacons necessary for transmission of each traffic based on the data arrival rate.

(3) Processing contents of each unit Next, specific processing contents executed in each module will be described in detail. However, since the processing content of the routing unit 13 is an existing technology, a description thereof will be omitted.
(A) Processing of Classification Unit The packet for which routing processing has been completed is passed to the classification unit 15. Based on the priority (class) included in the header information and the address of the adjacent node that is the transfer destination, the classifying unit 15 distributes the packet to the queue prepared by the number of adjacent nodes × 2. In the case of FIG. 2, the number of adjacent nodes is N (N is a natural number).
The classification unit 15 is provided with a mechanism for discarding packets that exceed the maximum number of packets that can be held in the MAC layer. This method is called a tail drop method.
Note that if most of the buffer reserved by the MAC layer is occupied by one class of traffic, the other class of traffic hardly flows.
The classifying unit 15 is also provided with a function for avoiding this problem. That is, the classifying unit 15 has a function of checking whether or not the maximum number of packets for each class is exceeded when the packets of both classes are distributed (enqueued).
When exceeding the maximum number of packets for each class, the classifying unit 15 discards the packet. The maximum number of packets for each class is given from the QoS management unit 21.

FIG. 3 shows an example of a processing procedure for realizing this processing function.
First, the classification unit 15 determines whether the input packet is a priority packet or a non-priority packet (process P1). That is, the classification unit 15 determines whether the input packet is a TGP class packet (TGP packet) or an FAP class packet (FAP packet).
First, a case where a negative result is obtained in the process P1 will be described.
If it is an FAP packet, the classification unit 15 compares the total number of packets in the output waiting queue 17 with the threshold value LimBUF (process P2). Here, the threshold value LimBUF is a limit value (constant) of the number of packet buffers that each node can hold.
When the total number of packets is equal to or greater than the threshold value LimBUF, the classification unit 15 discards the input packet (process P3).
After discarding, the classification unit 15 returns to the process P1 and starts the above-described determination for the next input packet.
On the other hand, when the total number of packets is less than the threshold value LimBUF, the classifying unit 15 compares the total number of FAP packets with the threshold value LimFAP (process P4). Here, the threshold LimFAP is the upper limit number (variable) of FAP packets. The threshold LimFAP is given from the QoS management unit 21.
When the total number of FAP packets is equal to or greater than the threshold value LimFAP, the classification unit 15 discards the input packet (process P3).
On the other hand, when the total number of FAP packets is less than the threshold LimFAP, the classification unit 15 adds the input packet to the end of the corresponding queue (process P5).

Next, a case where a positive result is obtained in the process P1 will be described.
Even in the case of a TGP packet, the classification unit 15 compares the number of all packets in the output waiting queue 17 with the threshold value LimBUF (process P6). When the total number of packets is equal to or greater than the threshold value LimBUF, the classification unit 15 discards the input packet (process P7). After discarding, the above-described determination is started for the next input packet.
On the other hand, when the total number of packets is less than the threshold value LimBUF, the classifying unit 15 compares the total number of TGP packets with the threshold value LimTGP (process P8). Here, the threshold LimTGP is the upper limit number (constant) of TGP packets. The threshold LimTGP is given from the QoS management unit 21.
When the total number of TGP packets is equal to or greater than the threshold value LimTGP, the classifying unit 15 discards the input packet (process P7).
On the other hand, when the total number of TGP packets is less than the threshold LimTGP, the classification unit 15 adds the input packet to the end of the corresponding queue (process P9). Thereafter, the classification unit 15 updates the maximum accumulated packet number N_PRI (process P10) and updates the data arrival rate (process P11).
Here, the maximum accumulated packet count N_PRI refers to the maximum value of TGP packets in the queue within a certain period. In this example, a certain period is one superframe.
The data arrival rate refers to the total number of bytes of TGP packets that arrive within a certain period. In this example, a certain period is one superframe.
Note that the maximum accumulated packet number N_PRI and the data arrival rate are obtained for the TGP packets written in the TGP class queue.

(B) Processing of QoS Management Unit The QoS management unit 21 calculates the auxiliary beacon number BN, the token generation rate TR, and the maximum packet numbers LimFAP and LimTGP for each class. Information necessary for the calculation includes three items: a data arrival rate, a maximum number of stored packets N_PRI, and an effective transfer rate at that time.
These pieces of information are periodically fetched from the classification unit 15. In this embodiment, these pieces of information are acquired every superframe.
The generated auxiliary beacon number BN is used by the auxiliary beacon management unit 25. The generated token generation rate TR is used by the token generation unit 23. The maximum packet numbers LimFAP and LimTGP are used in the classifying unit 15.
Next, calculation examples of these values will be shown.
First, a calculation example of the maximum number of packets LimFAP and LimTGP will be shown.
LimFAP = LimBUF-N_PRI-N_SURPLUS
LimTGP = LimBUF-N_MIN
Here, LimFAP is the maximum number of FAP packets that can be captured.
LimBUF is the maximum number of packets that each node can hold.
N_PRI is the maximum accumulated number of TGP packets. N_PRI is given as a constant at the time of initialization (at the time of activation), and its value is held.
N_SURPLUS is the number of packet buffers for absorbing the increase in TGP packets.
LimTGP is the maximum number of TGP packets that can be captured.
N_MIN is the minimum number of packet buffers to be reserved for FAP packets.
In this example, the maximum accumulated packet number N_PRI after the smoothing process is used for the calculation.

Next, a calculation example of the auxiliary beacon number BN and the token generation rate TR will be shown.
Basically, these values are calculated so as to be proportional to the traffic data arrival rate AR and inversely proportional to the effective transfer rate.
However, in this embodiment, the number of bytes remaining in the output queue 17 (remaining packet length) and the amount of tokens remaining in the token bucket are also used in the calculation so that packets and tokens do not stay. The remaining amount of the token is used as a value that gives the remaining radio resource.
That is, a modification is made to increase the number of auxiliary beacons and the token rate according to the remaining packet length. Also, a modification is made to reduce the token rate according to the remaining token amount (remaining token amount).
If the remaining packet length and token amount are used as they are, the correction amount may become too large. For this reason, in this embodiment, the remaining packet length and the token amount calculated by a quarter are used for correction.
Therefore, the calculation formula of the token rate TR in the embodiment is as follows.
TR = AR + (remaining packet length−remaining token amount) / 4
Moreover, the calculation formula of the number of auxiliary beacons BN in the embodiment is as follows. In this embodiment, when the number of auxiliary beacons BN is calculated, the number of remaining packets is corrected to one half.
BN = (AR + remaining packet length / 2) / (number of transferable bytes per auxiliary beacon)

FIG. 4 shows a processing procedure example for realizing this processing function.
First, the QoS management unit 21 acquires the traffic data arrival rate AR from the classification unit 15 (process P21).
In this embodiment, the QoS management unit 21 smoothes the acquired data arrival rate AR. The QoS management unit 21 takes in the smoothed data arrival rate AR and sets it to a temporary value of the token rate TR (process P22).
Next, the QoS management unit 21 determines whether or not a packet remains in the output waiting queue for priority traffic (process P23).
If the packet remains, the QoS management unit 21 adds a quarter of the remaining packet length to the token rate TR (process P24). By this addition process, the value of the token rate TR becomes larger than before the correction. By increasing the token rate TR, transmission of the TGP packet is promoted.
When the addition process is completed or there is no packet, the QoS management unit 21 determines whether or not a token remains in the token bucket (process P25).
If no token remains, the QoS management unit 21 subtracts a quarter of the remaining token amount from the token rate TR (process P26). By this subtraction process, the value of the token rate TR becomes smaller than before the correction. By reducing the token rate TR, excessive token accumulation is suppressed.
When this subtraction process ends or when there is no token, the QoS management unit 21 notifies the token generation unit 23 of the determined token rate TR (process P27).

Subsequently, the QoS management unit 21 acquires the current communication speed (effective transfer rate), and calculates the transferable byte number TB per auxiliary beacon (process P28).
Next, the QoS management unit 21 divides the data arrival rate AR by the token rate TR and obtains the necessary number of auxiliary beacons BN (process P29).
Thereafter, the QoS management unit 21 determines whether or not a packet remains in the priority traffic output waiting queue 17 (process P30).
When the packet remains, the QoS management unit 21 divides the half of the remaining packet length by the transferable byte number TB per auxiliary beacon, and obtains the number of auxiliary beacons necessary for the transmission. The QoS management unit 21 adds the calculated number of auxiliary beacons to the number of auxiliary beacons BM calculated in process P29 (process P31).
When this addition process ends or when there is no packet, the QoS management unit 21 notifies the beacon management unit 25 of the auxiliary beacon number BM (process P27).

(C) Processing of Token Generation Unit and Packet Scheduler The token generation unit 23 generates a token according to the token rate TR determined by the QoS management unit 21 and adds it to the corresponding token bucket. In the case of the embodiment, the token generation unit 23 adds the generated token for each head of the super frame.
The packet scheduler 19 extracts and transmits packets from each of the TGP traffic queue and the FAP traffic queue.
The packet scheduler 19 selects a TGP traffic queue so that output opportunities for all transfer destinations are equal during the TGP interval. That is, the packet scheduler 19 cyclically selects N output waiting queues for TGP traffic according to the round robin method.
At the same time, the packet scheduler 19 uses a token bucket to limit the amount of TGP packets sent and smooth the traffic. When transmitting the TGP packet, the packet scheduler 19 reduces the token length from the token bucket to the packet length for transmission.
On the other hand, the packet scheduler 19 selects a queue for FAP traffic so that output opportunities for all transfer destinations are equal during the FAP period according to the round robin method. That is, the packet scheduler 19 cyclically selects N output queues for FAP traffic according to the round robin method. However, the packet scheduler 19 does not limit the transmission rate using a token bucket or the like.

5 and 6 show an example of a processing procedure for realizing this processing function.
First, the packet scheduler 19 determines whether or not the current time belongs to the TGP section (process P41).
Here, a case where a negative result is obtained will be described.
If it is the FAP section, the packet scheduler 19 refers to the queue next to the last-referenced FAP traffic queue (process P42). Then, the packet scheduler 19 determines whether or not there is a packet in the referred queue (process P43).
If there is no packet, the packet scheduler 19 returns to the process P41. On the other hand, if there is a packet, the packet scheduler 19 takes out the packet located at the head of the queue (process P44).
Subsequently, the packet scheduler 19 determines whether or not further packet transmission is possible in the current FAP section (process P45).
If transmission is possible, the packet scheduler 19 determines whether there is another packet in the referenced queue (process P46). Here, when a positive result is obtained, the packet scheduler 19 returns to the process P44 and reads the next head packet.
On the other hand, when it is determined that the process P45 cannot transmit or when the process P46 determines that there is no packet, the packet scheduler 19 sets the packet group extracted in the process so far to the current FAP section. Send all at once. Thereafter, the packet scheduler 19 returns to the process P41 and repeats the above-described determination.

Next, a case where a positive result is obtained in process P41 will be described.
If it is the TGP section, the packet scheduler 19 refers to the queue next to the last-referenced TGP traffic queue (process P48). Then, the packet scheduler 19 determines whether or not there is a packet in the referenced queue (process P49).
If there is no packet, the packet scheduler 19 returns to the process P41. On the other hand, when there is a packet, the packet scheduler 19 determines whether or not the length of the packet located at the head of the queue is larger than the token amount (process P50).
When the packet length is larger than the token amount, the packet scheduler 19 does not extract the first packet, and transmits the previously extracted packet group collectively (process P54). If no packet is extracted, nothing is transmitted.
On the other hand, when the packet length and the token amount are the same or small, the packet scheduler 19 takes out the first packet and subtracts the amount from the token amount (process P51).
Subsequently, the packet scheduler 19 determines whether another packet can be transmitted in the current TGP section (process P52).
Here, when a positive result is obtained, the packet scheduler 19 determines whether there is another packet in the referred queue (process P53). If a positive result is also obtained here, the packet scheduler 19 returns to the process P50 and compares the length of the next top packet with the token amount.
On the other hand, if it is determined that the process P52 cannot transmit or if the process P53 determines that no packet exists, the packet scheduler 19 sets the packet group taken out from the referenced queue to the current TGP section. Send all at once. Thereafter, the packet scheduler 19 returns to the process P41 and repeats the above-described determination.
The packet scheduler 19 repeatedly executes such processing.

(D) Experimental Results The present inventor conducted a transmission experiment on the nodes described above. Hereinafter, experimental results will be described for two experimental examples.
(D-1) Experimental example 1
FIG. 7 shows an example of the network topology. In this network, each node can receive the beacons of all nodes other than itself.
In the figure, circles indicate each node. The numbers in the circles are numbers for identifying each node. Here, the distance between adjacent nodes is 5 [m].
The experiment was conducted by flowing the following two traffics with different priorities on this network.
(I) FAP traffic 20 [Mbps] CBR (Constant Bit Rate) traffic on the route given by the nodes “1” → “2” → “3” → “5” from the 0.5 second to 9. Run for 9 seconds to 5 seconds.
(Ii) TGP traffic 20 [Mbps] CBR (Constant Bit Rate) traffic on the route given by the nodes “0” → “2” → “3” → “4” from the 9th to the 9th. Run for 5 seconds until the 5th second.

FIG. 8 shows the throughput at this time, FIG. 9 shows the delay amount, and FIG. 10 shows the number of auxiliary beacons.
As shown in FIG. 8, it was confirmed that the throughput of the TGP traffic almost coincided with the expected value. That is, the throughput is almost 20 [Mbps], and it has been confirmed that transmission is preferentially performed over FAP traffic.
On the other hand, it has been confirmed that the throughput of FAP traffic decreases with the start of transmission of TGP traffic and decreases to approximately 10 [Mbps].
Moreover, as shown in FIG. 9, the delay amount of the TGP traffic decreased with the passage of time and stabilized at about 3 to 40 [msec]. On the other hand, the delay amount of FAP traffic increased with the start of transmission of TGP traffic, and stabilized at about 8 to 90 [msec].
Also, as shown in FIG. 10, the numbers of auxiliary beacons at nodes “0”, “2”, and “3” through which TGP traffic flows are all stable. Moreover, it was confirmed that the total number of auxiliary beacons at each node also changed stably.

(D-2) Experimental example 2
FIG. 11 shows another example of the network topology. However, in this network, it is assumed that each node can receive only beacons from adjacent nodes and cannot receive beacons from non-adjacent nodes.
FIG. 12 shows this relationship. Note that the distance between adjacent nodes is at least 8 [m].
The experiment was conducted by flowing the following two traffics with different priorities on this network.
(I) FAP traffic 15 [Mbps] CBR (Constant Bit Rate) traffic on the route given by the nodes “1” → “2” → “3” → “5” from the 0.5 second to 9. Run for 9 seconds to 5 seconds.
(Ii) TGP traffic 15 [Mbps] CBR (Constant Bit Rate) traffic on the route given by the nodes “0” → “2” → “3” → “4” Run for 5 seconds until the 5th second.

The throughput at this time is shown in FIG. 13, the delay amount is shown in FIG. 14, and the number of auxiliary beacons is shown in FIG.
As shown in FIG. 13, it was confirmed that the throughput of the TGP traffic almost coincided with the expected value. That is, it was confirmed that the throughput was approximately 15 [Mbps] and was transmitted with priority over the FAP traffic.
On the other hand, it has been confirmed that the throughput of FAP traffic decreases with the start of transmission of TGP traffic and decreases to approximately 6 [Mbps].
Moreover, as shown in FIG. 14, the delay amount of the TGP traffic decreased with the passage of time. On the other hand, the delay amount of FAP traffic increased with the start of transmission of TGP traffic.
Also, as shown in FIG. 15, the numbers of auxiliary beacons at nodes “0”, “2”, and “3” through which TGP traffic flows are all stable. In addition, although the total number of auxiliary beacons at each node also slightly increased, it was confirmed that it was relatively stable.

(4) Effect of Embodiment By adopting the control function described above, a transmission band necessary for each traffic can be secured in an autonomous and distributed manner at each node. As a result, data transmission with a priority according to the application can be realized with practical quality.
(5) Other Embodiments (a) In the above-described embodiment, the maximum number of packets LimFAP that can be assigned to a non-priority class is defined as LimBUF-N_PRI-N_SURPLUS.
In this case, the value of LimFAP decreases as the maximum number of TGP packets N_PRI increases.
On the other hand, if the value of LimFAP becomes too small, the non-priority class can hardly be transmitted. Therefore, when LimFAP falls below a preset reference value, it is desirable to raise at least the value of LimFAP to the reference value.
For example, the number of packet buffers N_SURPLUS for absorbing the increase in TGP packets may be forcibly reduced. Further, the maximum number of TGP packets N_PRI may be forcibly reduced.
(B) In the above-described embodiment, a method is adopted in which securing of the auxiliary beacon is continued for a while after the traffic is interrupted.
However, if the duration time is too long, the adverse effect of consuming radio resources unnecessarily increases. Therefore, it is desirable to install a function for forcibly setting the data arrival rate to zero when there is no traffic for a certain period. For example, it is desirable to employ a method for forcibly setting the data arrival rate to zero when the no-input period continues for 10 superframes.
Thus, by changing the data arrival rate to zero, radio resources can be released early.

(C) In the above-described embodiment, the case where the communication function is realized by hardware has been described. However, the same function can also be realized as firmware or software. These programs are stored in an optical storage medium (optical disk, follower medium, etc.), magnetic recording medium (hard disk, etc.), and semiconductor memory.
(D) The above-described wireless communication device includes various devices depending on the product form. For example, portable information terminals (portable computers, mobile phones, portable game machines, electronic books, etc.), wireless LAN devices (wireless routers, slave units, access points), watches, game machines, computers, video cameras, digital cameras , A scanner, a printing device, an image reproducing device (for example, an optical disc device, a home server), a monitor, a television receiver, and a processing board and a processing card equipped with the functions according to the invention.
In any case, the casing, the antenna, and the transmission / reception circuit have a common configuration, and are configured by combining peripheral devices according to the product form. For example, a portable information terminal has an operation input unit, a display unit, and other user interfaces in addition to the above-described configuration. For example, in the case of a video camera or a digital camera, in addition to the above-described configuration, a camera unit and a writing circuit for storing captured video data in a storage medium are provided.
(E) Various modifications can be considered in the above-described embodiments within the scope of the gist of the invention. Various modifications and application examples created based on the description of the present specification are also conceivable.

It is a figure which shows the example of arrangement | positioning of a regular beacon and an auxiliary beacon. It is a figure which shows the internal structural example of a radio | wireless communication apparatus. It is a figure which shows the example of a process sequence of a classification part. It is a figure which shows the process sequence example of a QoS management part. It is a figure which shows the process sequence example of a packet scheduler. It is a figure which shows the process sequence example of a packet scheduler. It is a figure which shows the topology example of an ad hoc network. It is a figure which shows the throughput measured about the network shown in FIG. It is a figure which shows the amount of delay measured about the network shown in FIG. It is a figure which shows the number of auxiliary beacons measured about the network shown in FIG. It is a figure which shows the topology example of an ad hoc network. It is a figure which shows the reception state of a beacon. It is a figure which shows the throughput measured about the network shown in FIG. It is a figure which shows the amount of delay measured about the network shown in FIG. It is a figure which shows the number of auxiliary beacons measured about the network shown in FIG.

Explanation of symbols

11 Node 13 Routing unit 15 Classification unit 17 Output queue 19 Packet scheduler 21 QoS management unit 23 Token generation unit 25 Auxiliary beacon management unit

Claims (9)

A radio communication method characterized by autonomously controlling the allocated amount of radio resources corresponding to each traffic based on a data arrival rate for each traffic. The wireless communication method according to claim 1,
A wireless communication method, wherein the data arrival rate of the traffic is obtained for each service class. The wireless communication method according to claim 1,
In the control, the reference value of the allocation amount is determined based on the data arrival rate for each traffic, and the reference value is corrected using the remaining packet length and the remaining radio resource amount for each traffic. The wireless communication method according to claim 3,
The radio communication method according to claim 1, wherein the correction is performed by increasing an allocation amount according to a remaining packet length for each traffic and reducing an allocation amount according to a remaining radio resource amount. An arrival rate calculator for determining the data arrival rate for each traffic;
A radio communication apparatus comprising: a radio resource control unit that autonomously controls an allocation amount of radio resources corresponding to each traffic based on a determined data arrival rate. The wireless communication device according to claim 5, wherein
The wireless communication apparatus characterized in that the data arrival rate of the traffic is obtained for each service class. The wireless communication device according to claim 5, wherein
In the control, the reference value of the allocation amount is determined based on the data arrival rate for each traffic, and the reference value is corrected using the remaining packet length and the remaining radio resource amount for each traffic. The wireless communication device according to claim 7, wherein
The radio communication apparatus according to claim 1, wherein the correction is performed by increasing an allocation amount according to a remaining packet length for each traffic and reducing an allocation amount according to a remaining radio resource amount. In a computer installed in a wireless communication device,
A program characterized by executing a process for autonomously controlling the allocated amount of radio resources corresponding to each traffic based on a data arrival rate for each traffic.

Images