JP5761802B2 - Wireless communication system - Google Patents

Description

  The present invention relates to a channel switching technique in wireless communication.

  In recent years, as a high-speed wireless access system using the 2.4 GHz band or the 5 GHz band, base station apparatuses (AP: Access point) based on the IEEE802.11g standard, the IEEE802.11a standard, and the like are widely spread. In systems based on these standards, an orthogonal frequency division multiplexing (OFDM) modulation method, which is a technology for stabilizing the characteristics in a multipath fading environment, is used, and a transmission rate of 54 Mbps is realized at the maximum. doing.

  However, the transmission rate described above is a transmission rate on the physical layer, and is not a data throughput effective for the user. Actually, since the transmission efficiency in the MAC (Medium Access Control) layer is about 50 to 70%, the throughput has an upper limit of about 30 Mbps.

  On the other hand, the communication speed of the wired LAN has been increasing due to the spread of FTTH (Fiber to the home). For this reason, it is assumed that further increase in transmission speed will be required in the wireless LAN in the future. Various spatial signal processing techniques such as MIMO and multi-user MIMO have been studied to increase the throughput of the radio section, but communication frequency bands have been expanded as other methods. In IEEE802.11a, a frequency band of 20 MHz is used for each channel. In IEEE802.11n, a frequency band of 40 MHz is used. Further, IEEE 802.11ac is considered up to 160 MHz when options are included. In this way, channel band expansion is progressing.

  Thus, the frequency band of the channel extends from IEEE802.11a to 11ac and is expanded eight times. However, no major expansion has been recognized for the entire frequency band that can be used for a wireless LAN. Therefore, with the spread of wireless terminals, frequency resources are becoming insufficient. For example, an environment in which a plurality of base station apparatuses use the same frequency band is increasing. For this reason, depending on the channel selected by the base station apparatus, the throughput may decrease due to the influence of packet signals from other base station apparatuses with which communication cells overlap each other, or the throughput efficiency of the entire system may decrease. There was a problem. Furthermore, channel bandwidths that can be selected by each base station apparatus are diversified. Therefore, when managing the frequency channel of the base station apparatus using the centralized control station (AC), it is necessary for the centralized control station to determine which frequency band to use as well as which frequency to use.

  A method of connecting a plurality of base station apparatuses to a central control station and assigning a channel to each base station apparatus is described in Patent Document 1. In Patent Document 1, a communication state such as throughput in a base station apparatus can be used as input information, and a channel can be selected so as to maximize system throughput.

JP 2007-74097 A

  However, the method of Patent Literature 1 should select a channel when an uncontrollable base station apparatus existing in the vicinity of a controllable base station apparatus is communicating using channels of various frequency bands. I have not taken into account. In an example of a wireless LAN, there are base station apparatuses using channels of 20 to 160 MHz in the vicinity, and what kind of channel the central control station uses and how to allocate them to the connected base station apparatuses It is not clear.

  In view of the above circumstances, an object of the present invention is to provide a technique for improving the efficiency of wireless communication in a wireless communication system in which other wireless communication devices using overlapping frequencies exist in the vicinity and an interference signal is generated. And

  One aspect of the present invention is a wireless communication system that includes a centralized control station and a plurality of base station apparatuses connected to the centralized control station, and performs wireless communication in an environment in which an interference signal is received in the base station apparatus. The base station apparatus includes interference estimation means for estimating interference relation information that is information on the interference signal, interference relation information notification means for notifying the central control station of the estimated interference relation information, and a terminal apparatus. Channel setting means for changing a channel used for radio communication with the channel notified from the central control station, and the central control station selects a plurality of channels having different frequency bands for each base station apparatus. A channel index for calculating a throughput index of two or more channel candidates from among the channel candidates including the channel candidates and determining a channel of the base station apparatus based on the calculated throughput index And Le determining means, the determined channel, a channel notifying means for notifying the base station apparatus, a wireless communication system comprising a.

  One aspect of the present invention is the above wireless communication system, wherein the channel determination means is another device that communicates within a frequency band range of a channel candidate of a connected base station device. The throughput index is calculated based on the number or communication frequency of devices other than the terminal device wirelessly communicating with a base station device and the frequency bandwidth or the number of data subcarriers of the channel candidate.

  One aspect of the present invention is the above-described wireless communication system, in which the base station device includes the number of the terminal devices communicating with the own device and a frequency bandwidth used for communication with the terminal device or Terminal information notifying means for notifying the central control station of terminal information including at least one of the number of data subcarriers and communication frequency information of the terminal device, and the channel determining means of the central control station And calculating the throughput index further based on the terminal information.

  One aspect of the present invention is the above-described wireless communication system, wherein the interference estimation unit of the base station device is a source address of information or a communication packet notified by a beacon received when communication is stopped in the own device. At least one of the transmission destination address of the communication packet, the address of the base station apparatus to which the communication packet belongs, and the time occupancy rate of the communication packet is acquired as interference relation information.

  One aspect of the present invention is the above-described wireless communication system, wherein the central control station further includes a communication stop unit that notifies the base station apparatus to collect interference-related information without performing communication for a certain period of time. When the stop of communication is designated by the central control station, the interference estimation unit of the base station apparatus stops data communication and collects interference relation information.

  One aspect of the present invention is the above wireless communication system, wherein the central control station further includes interference measurement designating means for designating a common channel to the base station apparatus, and the interference estimation means of the base station apparatus Determines whether the signal transmitted from each base station device can be detected using the channel specified by the central control station, and acquires it as interference relation information.

  One aspect of the present invention is the above-described wireless communication system, wherein the central control station independently uses a channel in a frequency band of a specific channel candidate based on information notified from the base station apparatus. And a channel determination means for determining a throughput index of the channel candidate using the number of base station apparatuses operating independently in the channel candidate. calculate.

  One aspect of the present invention is the above-described wireless communication system, wherein the centralized control station uses a frequency band of a channel used by an unconnected base station based on information notified from the base station apparatus. Among them, channels of the plurality of base stations that are in an interference relationship with the base station that is not connected and that are not in a relationship in which signals can be detected are allocated.

  One aspect of the present invention is the above-described wireless communication system, wherein the channel determination unit determines an order of evaluating the channels used by the base station apparatus, and sets a throughput index for each base station apparatus in the determined order. Determine the high channel.

  One aspect of the present invention is the above wireless communication system, wherein the channel determination unit determines a plurality of patterns in the order, determines the channel for each base station apparatus in an order determined for each pattern, Based on the combination of channels determined for each pattern and the throughput index, a combination of channels in any one pattern is selected.

  According to the present invention, it is possible to improve the efficiency of radio communication in a radio communication system in which other radio communication apparatuses using overlapping frequencies exist in the vicinity and an interference signal is generated.

It is a figure showing the outline of the radio | wireless system 1 as one Embodiment of this invention. 2 is a diagram illustrating channels used in the wireless system 1. FIG. It is a figure which shows the example of the detection result of interference AP and system AP. It is a table representing the throughput indicator T _j expected for each channel to be selected for each system AP. It is a figure which shows the specific example (channel determination flow 1-1) of the process which determines the channel which each base station apparatus uses. It is a figure which shows the specific example (channel determination flow 1-2) of the process which determines the channel which each base station apparatus uses. It is a figure which shows the example of the detection result of interference AP and system AP. It is a table representing the throughput indicator T _j expected for each channel to be selected for each system AP. It is a figure which shows the specific example (channel determination flow 2-1) of the process which determines the channel which each base station apparatus uses. It is a figure which shows the specific example (channel determination flow 2-1) of the process which determines the channel which each base station apparatus uses. It is a figure which shows the specific example (channel determination flow 2-2) of the process which determines the channel which each base station apparatus uses. It is a figure which shows the specific example (channel determination flow 2-2) of the process which determines the channel which each base station apparatus uses. It is a figure showing the channel detection result of interference AP. A communication frequency detection table is shown when channel E (X) and channel F (ε) are selected for AP4 and AP1, respectively. 7 shows a communication frequency detection table when channel A (T) is selected for AP2. It is a figure which shows the relationship when the centralized control station AC selects F ((epsilon)) as a channel of AP4. It is a figure which shows the specific example of a table | surface when the channel of AP0 is channel C (Q), the channel of AP7 is channel A (T), the communication frequencies are 90 and 100, respectively, and the signal of AP7 reaches AP3. . It is a table | surface showing the function of each terminal. It is a figure showing the detection table of the communication frequency when channel F is selected for AP1 and channel E is selected for AP4. It is a flowchart which shows the flow of a process of the channel determination method in this invention. It is a flowchart which shows the flow of a process of the modification of the channel determination method in this invention. It is a figure which shows the table | surface which added the star to the channel which a characteristic degradation by the influence of interchannel interference adds to the table of the detection result of the communication channel of FIG.

  Hereinafter, an embodiment of the present invention will be described. In the following description, the base station apparatus and the central control station may be described as “AP” and “AC”, respectively.

[Outline]
FIG. 1 is a diagram schematically illustrating a wireless system 1 as an embodiment of the present invention. The wireless system 1 includes a central control station AC and a plurality of base station devices. Each base station apparatus communicates with terminals connected to the base station apparatus by wireless communication. A plurality of base station devices (AP1, AP2, AP3, AP4, AP6) are connected to the central control station AC. On the other hand, the base station devices AP0, AP5, AP7, AP8, AP9, and AP10 are not connected to the central control station AC. In the following description, the base station apparatus connected to the central control station AC is referred to as “system AP”. A base station apparatus that is not connected to the central control station AC is referred to as an “interference AP”.

A certain base station device has another base station device that is located next to each other in the up / down / left / right direction, another base station device that is located obliquely next to the other, and another base station that is located next to the other in the up / down / left / right direction. It is assumed that the station apparatus and the station apparatus can detect signals with each other. For example, the base station device AP1 can detect signals with each other with the other base station devices AP0, AP2, AP3, AP4, AP5, and AP7. In other words, the base station device AP1 can detect a signal emitted from any of the other base station devices AP0, AP2, AP3, AP4, AP5, and AP7. On the other hand, the signal transmitted by the base station device AP1 is detected by all of the other base station devices AP0, AP2, AP3, AP4, AP5, and AP7.
In the wireless system 1, the throughput of the system AP is improved.

  FIG. 2 is a diagram illustrating channels used in the wireless system 1. As the channel arrangement in the wireless system 1, for example, as shown in FIG. 2, a band that can be used in the IEEE 802.11 5 GHz band can be applied. The channel described above is an example, and the channel used in the wireless system 1 may be another frequency.

  The centralized control station AC designates a channel to be used for communication from the channels A to η in the bands of 20 MHz, 40 MHz, 80 MHz, and 160 MHz for each system AP. The central control station AC selects a primary channel from the 20 MHz channels P to η. For example, when communicating using channel A, the centralized control station 20 sets one 20 MHz channel among the channels P to W as a primary channel and the other channels as secondary channels.

[First Embodiment]
A first embodiment of the present invention will be described. The centralized control station AC in the first embodiment calculates the throughput index using the number of system APs that perform communication in the selected channel or the number of system APs and terminals. Then, the central control station AC determines a channel based on the calculated index. Hereinafter, the first embodiment will be described in detail.

  FIG. 3 is a diagram illustrating an example of detection results of the interference AP and the system AP. FIG. 3A shows an example of the detection result of the primary channel of the interfering AP. FIG. 3B shows an example of the detection result of interference with the system AP. In FIG. 3B, a circle indicates that interference occurs when signals in the same frequency band are used. In the first embodiment, the system AP stops communication with the terminal based on an instruction from the central control station AC, and detects the primary channel used by the interfering AP.

  The detection result of the interfering AP shown in FIG. 3A is a table of primary channels used by the interfering AP detected by the system AP. There are two marks (double circles) in the same frame (AP2 ε and AP6 ε) because it is detected that another base station apparatus uses the same channel. According to FIG. 3A, AP0 uses channel T, AP5 uses channel X, AP7 uses channel Q, AP8 uses channel ε, AP9 uses channel β, and AP10 uses channel ε as the primary channel.

  The detection result of the system AP shown in FIG. 3B is a result of investigating whether or not the system APs interfere with each other. This can be measured by designating the same channel from the centralized control station AC to each system AP and observing each other's beacon or communication packet.

(Channel evaluation index 1)
Next, a throughput index will be described as one specific example of the channel evaluation index. The throughput index is an index used when the system AP determines which channel to select from a plurality of channels including different frequency bandwidths. The throughput index T _j of the j-th system AP is expressed by Equation 1 using the utilization frequency band index B _j and the throughput degradation rate ρ _j affected by other interference APs or system APs existing in the band. Can be expressed as

It can be assigned a number of sub-carriers are used for bandwidth B _c and data channels j-th system AP has selected use frequency band index B _j.
ρ _j is expressed by a function of m (j) and l (j) as in the following Expression 2.

m (j) is the number of base station apparatuses other than the j-th system AP existing in the same channel, and l (j) is the number of the j-th system AP (l is a small letter of L). m (j) may be defined as the number of base station apparatuses other than the j-th system AP and terminals communicating with base station apparatuses other than the j-th system AP existing in the same channel. In this case, the number of terminals communicating with the jth system AP and the number of the jth system AP may be defined as l (j). ρ _j may be expressed as Equation 3 below.

ρ_ｊは、以下の式４のように表されても良い。

ρ _j may be expressed as in Equation 4 below.

式５の場合、ｍ（ｊ）の関数で得られる値をｌ（ｊ）／（ｍ（ｊ）＋ｌ（ｊ））に足すことで、スループットの期待値を上げたり、下げたりすることもできる。 ρ _j may be expressed as in Equation 5 below.

In the case of Equation 5, the expected value of throughput can be increased or decreased by adding the value obtained by the function of m (j) to l (j) / (m (j) + l (j)). .

  The reason why f (m (j) + l (j)) is made positive and the expected value of throughput is increased is as follows. Actually, communication does not always occur at the same timing. For this reason, communication is performed at timings independent of each other, and if they happen to be out of time, they are not subject to communication degradation.

The reason why f (m (j) + l (j)) is negative and the throughput is lowered is as follows. This is because the execution throughput is reduced rather than simply dividing by time due to packet collision or the like.
When m (j) + l (j) is large, f (m (j) + l (j)) is made small or negative, and when m (j) + l (j) is small, f It is also possible to define f (m (j) + l (j)) so that (m (j) + l (j)) is increased or set to a positive value.

The value of f (m (j) +1 (j)) may be expressed as Equation 6. G, D and Z are given as positive numbers. As in Equation 6, when m (j) + l (j) increases, it can take a negative value.

ρ_ｊは、以下の式８のように表されても良い。 ρ _j may be expressed as the following Expression 7.

ρ _j may be expressed as in Equation 8 below.

この場合、ｍ（ｊ）＋ｌ（ｊ）の関数gまたは関数hによる補正を行うこともできる。ｇ（ｍ（ｊ）＋ｌ（ｊ））の関数で得られる値をｌ（ｊ）／（ｍ（ｊ）＋ｌ（ｊ））に乗算する、またはｈ（ｍ（ｊ）＋ｌ（ｊ））の階乗を行うことで、スループットの期待値を上げたり、下げたりすることもできる。ｍ（ｊ）＋ｌが小さい値をとるときは、ｇ（ｍ（ｊ）＋ｌ）を１より大きくする、またはｈ（ｍ（ｊ）＋ｌ（ｊ））を１より小さくしてスループットの期待値を上げることができる。これは、通信が実際には同じタイミングに発生するとは限らず、互いに独立のタイミングに通信し、たまたまそれらが時間的にずれていれば、お互いに通信の劣化を受けないためである。また、ｍ（ｊ）＋ｌ（ｊ）が大きい値をとる場合には、ｇ（ｍ（ｊ）＋ｌ（ｊ））を１より小さくしたり、ｈ（ｍ（ｊ）＋ｌ（ｊ））を１より大きくしたりしてスループット指標を下げることができる。これは、パケット衝突などにより、単に時間で分けるより、実行スループットが低下することを考慮するためである。また、ｆ（ｍ（ｊ）＋ｌ（ｊ））と同様にｇ（ｍ（ｊ）＋ｌ（ｊ））とｈ（ｍ（ｊ）＋ｌ（ｊ））も当該チャネルのデータ量などによる通信状況により変化させることができる。

In this case, the correction by the function g or the function h of m (j) + l (j) can also be performed. Multiply l (j) / (m (j) + l (j)) by a value obtained by a function of g (m (j) + l (j)) or h (m (j) + l (j)) By using factorials, you can raise or lower the expected throughput. When m (j) + l takes a small value, g (m (j) + l) is made larger than 1 or h (m (j) + l (j)) is made smaller than 1 to set the expected throughput. Can be raised. This is because communication does not actually occur at the same timing, but communicates at timings independent of each other, and if they happen to be out of time, they are not subject to communication degradation. When m (j) + l (j) takes a large value, g (m (j) + l (j)) is made smaller than 1 or h (m (j) + l (j)) is set to 1. The throughput index can be lowered by making it larger. This is because the execution throughput is reduced rather than simply dividing by time due to packet collision or the like. Similarly to f (m (j) + l (j)), g (m (j) + l (j)) and h (m (j) + l (j)) also depend on the communication status depending on the data amount of the channel. Can be changed.

In Equations 4 to 8, when m (j) is the number of base station devices other than the jth base station device, l (j) is the number of jth base station devices, that is, l ( j) = 1.
The throughput index T _j may be expressed as in Equation 9 below. In this case, the centralized control station AC investigates the number of groups of base station apparatuses (hereinafter referred to as “AP groups”) that operate independently in the channel selected by the j-th base station apparatus. When there are n AP groups that operate independently, the independent AP penalty Γ _{n, j} for the j-th base station corresponding to these numbers can be reflected in the throughput index.

  n is the number of AP groups having independent primary channels. n may also include the number of AP groups that have the same primary channel but are in a positional relationship where communication packets and beacons cannot be detected. Whether such a positional relationship (relationship having the same primary channel and operating independently) is based on whether one base station apparatus has acquired the access right by RTS / CTS or the like. Nevertheless, the determination is made based on whether or not the other base station apparatus is communicating. Whether or not the positional relationship is satisfied may be determined based on whether or not the other base station apparatus is transmitting a communication packet or a beacon while one base station apparatus is transmitting. The above determination can be made based on the observation result by the system AP. For example, when the system AP reports the interference relationship to the centralized control station AC, the interference AP observed in the same primary cell is operating independently, or the positional relationship where the communication packets can be received from each other. Thus, it is possible to determine whether the communication is performed in accordance with the rules of RTS / CTS and CSMA / CA, and to notify the centralized control station AC.

The different channel penalty Γ _{n, j} represents the different channel penalty amount for the j-th base station when n independently operating AP groups are on the same channel. For example, Γ _{1, j} = 1, Γ _{2, j} = 0.5, Γ _{3, j} = 0.1, Γ _{4, j} = 0.02 can be set.

A specific example of calculating the throughput index T _j will be described. In this description, the throughput index T _j is calculated using Equation 2 and Equation 9. Here, using a frequency bandwidth as an index _{B j,} widest in the case of 160MHz is the time of _B j = 160,80MHz the value of the bandwidth of the selected channel _B j = 80, the unit and so as MHz May be used. Here, B _j = 160, Ω = 0.6, and m (j) are the number of base station devices other than the j-th system AP. When one interfering AP primary channel is included, T _j = 160 × 1 × 0.6 ¹ = 96. When two primary channels of the interfering AP are included and these are independent primary channels, T _j = 160 × 0.5 × 0.6 ² = 14.4. When there are four base station apparatuses other than the base station apparatus and three independent channels are included, T _j = 160 × 0.1 × 0.6 ⁴ = 2.08.

FIG. 4 is a table showing the throughput index T _j expected for each channel selected for each system AP. In the table of FIG. 4, about 8 to 10 throughput indices T _j are listed in order from the largest value. The throughput index T _j is a value calculated using Equation 2 and Equation 9. Ω = 0.6, Γ ₁ = 1, Γ ₂ = 0.5, Γ ₃ = 0.1, Γ ₄ = 0.02, and m is the number of other interfering APs and system APs.

  The centralized control station AC may consider brute force as to which channel is actually allocated to each AP, or may determine one by one in order. When determining one by one, the order may be determined based on a predetermined priority of the base station apparatus, or the order may be determined based on the size of the throughput candidates obtained in FIG. The order may be determined (for example, in order from the largest, or from the smallest), or the order may be determined at random each time. The predetermined priority of the base station device may be determined by the usage fee of the user who owns each base station device, may be determined based on the function or performance of each base station device, It may be determined according to the information of the interference AP notified from the base station apparatus, may be determined by the amount of data waiting for transmission stored in each base station as a buffer, and each base station apparatus exchanges It may be determined by the amount of data to be performed.

(Channel determination flow 1-1)
FIG. 5 is a diagram illustrating a specific example (channel determination flow 1-1) of processing for determining a channel used by each base station apparatus. The specific example shown in FIG. 5 shows the flow of processing for determining one by one in descending order of throughput candidates when the throughput index T _j is obtained as shown in FIG. In FIG. 4, the use of channels between system APs is not considered. Therefore, as the channel of the system AP is determined, the candidate channel and its throughput index T _j need to be updated. First, the detection result of the system AP shown in FIG. 3B and the detection result of the interference AP shown in FIG. 3A are merged and represented on the vertical axis. Further, the used channels of the detected system AP and interference AP are shown on the horizontal axis. FIG. 5A shows the state of the table in this state. Since it is undecided which channel each system AP uses, the system AP relationship is circled.

  First, the centralized control station AC performs processing for AP3 having a candidate with the highest throughput. In AP3, communication of the throughput index 160 can be expected by using channel A as shown in FIG. Therefore, the central control station AC selects channel A as the channel used by AP3. FIG. 5B shows the state of the table in this state. Since central control station AC has selected channel A as the channel of AP3, the interference state of AP3 has been changed from A to A.

Next, the centralized control station AC performs processing for AP6 having the next highest throughput after AP3. According to FIG. 5B, since AP6 receives interference from channels A, X , β, and ε, it is necessary to determine a channel based on these. Also, the throughput index needs to be updated from the values shown in FIG. Since it is determined that channel A is used by AP3, channels A, C, and D are shared with AP3. Therefore, in Equation 1, ρ = 0.6. That is, new throughput candidates are A: 96, C: 48, D: 48, E: 48, G: 40, H: 40, I: 40, J: 40, L: 40, O: 40,. As obtained. However, if channels A, C, D, G, H, I, and J are selected, the throughput index of AP3 decreases from 160 to 66 and becomes 96. In this way, when reducing the throughput index of the system AP that has already been selected, a penalty may be given to the selected channel.
For example, the attenuation coefficient γ can be multiplied by the number of system APs that reduce the throughput. Here, when γ is 0.8, the throughput candidates are A: 76.8, E: 48, L: 40, O: 40, C: 38.4, D: 38.4, G: 32, H : 32, I: 32, J: 32,. Even in this case, since the channel A takes the largest value, the central control station AC selects the channel A as the channel of the AP 6. FIG. 5C shows a relationship table indicating this state.

  In the above example, there is no primary channel of the interfering AP in channel A used by AP3 and AP6. Therefore, the centralized control station AC needs to newly determine a primary channel. The centralized control station AC may determine a new primary channel from P to W at this time, or may determine a new primary channel after all channels are determined. By deciding last, it is possible to adjust so that a plurality of primary channels do not occur in the channel selected by the system AP.

  Next, the centralized control station AC performs processing for AP1, which has the next highest throughput after AP6. According to FIG. 5C, since AP1 receives interference from channels A, Q, T, and X, it is necessary to determine a channel based on these. Also, the throughput index needs to be updated from the values shown in FIG. New throughput candidates are B: 96, F: 80, E: 48, L: 40, N: 40, M: 40, O: 40, H: 40 × 0.6 = 24, J: 40 × 0.6 = 24,... Here, the reason why C and D disappear from the choices is that AP3 has selected channel A, and since channels Q and T of the interfering AP are not detected from AP3, APs that operate independently in the same channel This is because there are two. The centralized control station AC selects channel B with high throughput as the channel of AP1. Here, since the primary channel X is observed, the primary channel of channel B is set to X. FIG. 5D shows the state of the table in this state. The description “B (X)” in FIG. 5D represents channel B with X as the primary channel.

  Next, the centralized control station AC performs processing for AP2, which has the next highest throughput after AP1. According to FIG. 5D, since AP2 receives interference from channels A, B, Q, T, and X, it is necessary to determine a channel based on these. Also, the throughput index needs to be updated from the values shown in FIG. New throughput candidates are C: 48 (primary channel is P to S), O: 40, E: 28.8, G: 24 (primary channel is P or Q), H: 24 (primary channel is R or S) It is obtained as. However, when channel C is selected, the primary channels of AP3 and AP6 must be selected from P to S. Furthermore, the same primary channel as AP3 and AP6 is shared. Therefore, the throughputs of AP3 and AP6 are reduced. Multiplying γ (= 0.8) by the number of system APs that will reduce the throughput, the throughput candidates are O: 40, C: 48 × 0.8 × 0.8 = 30.7, E: 28 .8, G: 24 × 0.8 × 0.8 = 15.4, H: 24 × 0.8 × 0.8 = 15.4,. Therefore, in order to increase the system throughput, the central control station AC selects channel O as the channel of AP2.

  Next, the centralized control station AC performs processing for AP4 having the next highest throughput after AP2. AP4 interferes with AP6 when signals in the same frequency band are used, but does not interfere with AP3. However, AP6 interferes with both AP3 and AP4 when signals in the same frequency band are used. For this reason, when AP 4 selects channel A and lower channels (C, D, G, H, I, J, P to W) related to channel A, the throughput is significantly reduced. In addition, since AP4 observes channel B of AP1 (primary channel is X), it is not possible to select a lower channel related to channel B whose primary channel is not X. Therefore, the channels that can be selected by AP4 are only E (X): 80 × 0.6 × 0.6 = 28.8, and O: 40 × 0.6 = 24.

When channel E (primary channel X) is selected as the channel of AP4, it is shared with channel B used by AP1. In this case, for AP1, the primary channel (channel X) of another AP exists on the same channel. Therefore, the throughput index of AP1 decreases from 160 × 0.6 = 96 to 160 × 0.6 ² = 57.6.
On the other hand, when channel O is selected as the channel of AP4, the throughput index of AP2 decreases to 40 × 0.6 = 24.

In any case, in order to reduce the throughput of one system AP, the AP 4 can select the channel E (X) having a high throughput index.
The central control station AC selects a channel according to a preset policy. When a policy that maximizes the system throughput is set, when the throughput index of another system AP is reduced, the reduced throughput index can be subtracted from the throughput index candidates as a penalty γ ′. That is, if E (X) is selected in the channel selection of the last AP4, the throughput index of AP1 decreases by 38.4, and if O is selected, the throughput index of AP4 decreases by 16. When this value is subtracted from the throughput index as γ ′, the selectable channels of AP4 are E (X): 80 × 0.6 × 0.6−36.6 = −7.6, O: 40 × 0.6 −16 = 8, and channel O is selected.

  The throughput indices of APs 1, 2, 3, 4, and 6 finally obtained in FIG. 5E are AP1: 57.6 (the primary channel X is used with one interference AP and one system), AP2: 40 (40 MHz channel) ), AP3: 96 (channel A is shared with AP6), AP4: 28.8 (primary channel X is shared with one interfering AP and AP1), and AP6: 96. Further, the throughput index of the entire system is 318.4.

(Channel decision flow 1-2)
FIG. 6 is a diagram illustrating a specific example (channel determination flow 1-2) of processing for determining a channel used by each base station apparatus. The specific example shown in FIG. 6 shows the flow of processing for determining one by one in order from the lowest throughput candidate when the throughput index _Ti is obtained as shown in FIG. By performing processing in this order, the effect of obtaining fairness and the effect of maximizing the minimum throughput of the system AP can be obtained.

  First, the centralized control station AC performs processing for the AP 4 having the smallest throughput candidate. As shown in FIG. 4, by using any one of channels C, D, E, and F, the throughput of AP4 is maximized. Here, the centralized control station AC may randomly select a channel from these channel candidates. The centralized control station AC may select a channel with a small number of detections in the entire primary channel of the interference AP detected by the system AP. Further, the centralized control station AC may select a channel having a large number of detections in the entire primary channel of the interference AP detected by the system AP.

  In this description, the central control station AC selects a channel with a large number of detections in the system AP. By selecting a channel with a large number of detections in the system AP, it is possible to select a channel that is difficult for the system AP other than the system AP (AP4) to use. Therefore, it is possible to expand channel selection options for other system APs.

  From the detection result of the interference AP in FIG. 3, the number of primary channel detections is Q: 2, T: 3, X: 5, β: 2, and ε: 7. Further, the total number of primary channels of interfering APs belonging to channels C, D, E, and F is as follows: C (P, Q, R, S): 2, D (T, U, V, W): 3, E (X, Y, Z, α): 5 pieces, F (β, χ, δ, φ): 9 pieces. In this case, there are many primary channels corresponding to channel F. Therefore, the central control station AC selects F (primary channel: ε) as the channel of AP4. FIG. 6A shows the state of the table in this state. Since central control station AC has selected channel F (primary channel: ε) as the channel of AP4, the interference state of AP4 has been changed from a circle to F (ε).

  Next, the centralized control station AC performs processing for AP2, which has the smallest throughput candidate next to AP4. According to FIG. 6A, the channel candidates that can be selected for AP2 are not affected by the increase in F (ε) of AP4. Therefore, the central control station AC selects channel A (primary channel: T) as the channel of AP2. FIG. 6B shows the state of the table in this state. Since central control station AC has selected channel A (primary channel: T) as the channel of AP2, the interference state of AP2 has been changed from a circle to A (T).

  Next, the centralized control station AC performs processing for AP1, which has the smallest throughput index candidate next to AP2. According to FIG. 6B, since the primary channel T of AP2 and the primary channel ε of AP4 are added, the channel selection candidates are F: 48, C: 48, E: 48, H: 40, L: 40, N : 40, O: 40,... Channels C, E, and F are expected values for the same throughput. However, channels C and F affect other system APs (AP2 and AP4). Therefore, the central control station AC selects the channel E (primary channel: X) as a channel having little influence on other system APs. FIG. 6C shows the state of the table in this state. Since central control station AC has selected channel E (primary channel: X) as the channel of AP1, the interference state of AP1 has been changed from a circle to E (X).

  Next, the centralized control station AC performs processing for AP6, which has the next smallest throughput candidate after AP1. Note that the AP3 candidate and the AP6 candidate have the same maximum throughput. However, the AP 6 that has the lowest expected throughput value (candidate) and that is associated with AP 4 (communication cells overlap each other) is AP 6. Therefore, the centralized control station AC can perform processing for AP6 before AP3.

According to FIG. 6C, since the primary channel T of AP2 and the primary channel ε of AP4 are added, the channel selection candidates are A: 96, C: 80, D: 48, E: 48, L: 40, O : 40,... However, when channel A is selected, two independent primary cells are created in channel A of AP2, and the throughput index of AP2 is significantly reduced from 96. Specifically, in ρ _j of Equation 1, when Γ _{n, j} of Equation 9 is considered, Γ _{n, j} × Ω = 0.2 × (0.6) is used as an index of two independently operating APs. Is considered, the throughput index of AP2 is 160 × 0.2 × 0.6 = 19.2. Thus, when the different channel penalty Γ _{n, j} occurs, the central control station AC may not select the channel (here, channel A). Next, when channel C is selected, AP2 has a plurality of primary channels in the channel. Therefore, the channel selected by AP2 is changed from A (T) to D (T), and the throughput index is set to 96. By reducing the value from 48 to 48, a different channel penalty can be avoided. When D (T) is reselected, the throughput index is larger than A (T) and multiplied by the different channel penalty. Similarly to channel A, channel D (primary channel: T) also has two independent primary channels in the channel of AP2. Channel X, which is the primary channel of channel E, has already been selected by AP1. Therefore, when channel E is selected as the channel of AP6, different channel penalties are generated for APs 2, 3, and 4 that overlap with both AP1 and AP6, and the throughput index is significantly reduced. Selecting channel L will cause AP1 to generate a plurality of primary channels in the channel and incur a different channel penalty. For this reason, the channel of AP1 can be changed to channel K (X) to avoid a different channel penalty. In this case, the throughput index decreases from 48 to 40 × 0.6 × 0.6 = 14.4. Therefore, here, the central control station AC shows a case where the channel O that does not affect other system APs is selected as the channel of the AP 6. FIG. 6D shows a correspondence table representing this state. Since central control station AC has selected channel O as the channel of AP6, the interference state of AP6 has been changed from a circle to O (O).

  Next, the centralized control station AC performs processing for the remaining AP3. According to FIG. 6D, channel candidates and throughput indices are: A: 96, C: 80, D: 48, G: 40, J: 40, H: 40, L: 40, E: 28.8,.・ It becomes. However, when channel A or channel D is selected, two primary channels that operate independently exist in channel A of AP2, and the throughput index is greatly reduced due to the different channel penalty. This is because the primary channel T of the interfering AP observed at AP2 is not observed from AP3. When channels C, G, and H are selected, it is necessary to change channel A (T) of AP2 to channel D (T) in order to avoid a different channel penalty. At this time, the throughput index is changed from 96 to 80. × 0.6 = 48 When channel J is selected, it is necessary to change channel A (T) to channel I (T) to avoid a different channel penalty for AP2, where the throughput index is from 96 to 40 × 0.6 = 24. To drop. When channel L is selected, channel E (X) needs to be changed to channel K (X) to avoid a different channel penalty for AP1, and the throughput index drops from 48 to 40 × 0.6 = 24. . Thus, even if any channel is selected, the throughput of other system APs is affected. In order to select a channel, the central control station AC may multiply the coefficient γ by the number of APs in the system that reduce the throughput. In addition, in order to select a channel, the central control station AC may determine a channel that maximizes the sum or product of the throughput indices of AP3 and the selected AP. Here, since the channel cannot be determined only by the number of APs that reduce the throughput, a case where the latter processing is performed will be described. When channel C is selected as the channel of AP3 and the channel of AP2 is D (T), the throughput indices of AP2 and AP3 are 48 and 80, respectively, and the sum and product of the throughput indices are maximized. Therefore, the centralized control station AC changes the channel of AP2 to channel D (T) and selects channel C as the channel of AP3.

  AP1, 2, 3, 4 and 6 throughput indices finally obtained are AP1: 48 (primary channel X is used as one interference AP), AP2: 48 (primary channel T is used as one interference AP), AP3: 80 (no channel C sharing), AP 4:48 (the primary channel ε is shared with one interfering AP), AP 6:40 (no channel O sharing), and the total system throughput index is 264. Compared to the channel determination flow 1-1, the total throughput index was reduced, but the minimum throughput index could be improved from 28.8 to 40.

  The wireless system 1 is a wireless communication system in which other base station devices using overlapping frequencies are present in the vicinity. In such a situation, the efficiency of wireless communication can be improved. Specifically, in the wireless system 1, the channel used by each system AP can be determined so as to improve the throughput of the system AP according to the channel used by the interference AP.

[Second Embodiment]
A second embodiment of the present invention will be described. The centralized control station AC in the second embodiment determines the channel of each system AP based not only on the primary channel but also on all the secondary channels used for communication. FIG. 7 is a diagram illustrating an example of detection results of the interference AP and the system AP. FIG. 7A is a table of primary channels and secondary channels used by interfering APs detected by the system AP connected to the centralized control station AC. A circle represents a secondary channel that exists in the same 40 MHz channel as the basic 20 MHz channel. The triangles represent the other 40 MHz secondary channel that is present in the same 80 MHz channel. A square mark represents the secondary channel corresponding to the other 80 MHz existing in the same 160 MHz channel. In FIG. 7B, a circle indicates that interference occurs when signals in the same frequency band are used. Here, AP0 is channel A (primary channel is channel T), AP5 is channel E (X), AP7 is channel A (Q), AP8 is channel N (ε), AP9 is channel M (β), and AP10 is channel F (ε) is used as the channel.

FIG. 8 is a table showing a throughput index T _j expected for each channel selected for each system AP. In the table of FIG. 8, about 8 to 10 throughput indices T _j are listed in order from the largest value. The throughput index T _j is a value calculated using Equation 5, Equation 6, and Equation 9. In Equation 6, each constant was calculated as G = 0.5, D = 1, and Z = 0. Considering the number n of AP groups operating independently in the channel, Γ _{1, j} = 1, Γ _{2, j} = 0.5, Γ _{3, j} = 0.1, Γ _{3, j} = 0.02 It was. In the second embodiment, unlike the first embodiment, the value of n is calculated including the secondary channel. In Equation 4, m (j) is the number of other base station apparatuses in the selected channel, and is calculated as l (j) = 1.

In FIG. 8, the channel and the throughput index T _j shown in parentheses correspond to the selection of the channel corresponding to the secondary channel of the interfering AP. When such a channel is selected as a communication channel, there is a possibility that the interference AP does not understand the MAC protocol such as RTS / CTS, and an interference packet arrives to reduce the throughput. Therefore, normally, such a channel is not selected. However, there are cases where such a channel is selected as an exception. Details will be described later.

(Channel determination flow 2-1)
9 and 10 are diagrams illustrating a specific example (channel determination flow 2-1) of processing for determining a channel used by each base station apparatus. The specific examples shown in FIGS. 9 and 10 show the flow of processing for determining one by one in descending order of throughput candidates when the throughput index _Tj is obtained as shown in FIG. In FIG. 8, the use of channels between system APs is not considered. Therefore, as the channel of the system AP is determined, the candidate channel and its throughput index T _j need to be updated. First, the detection result of the system AP shown in FIG. 7B and the detection result of the interference AP shown in FIG. 7A are merged and represented on the vertical axis. Further, the used channels of the detected system AP and interference AP are shown on the horizontal axis. FIG. 9A shows the state of the table in this state. Since which channel each system AP uses is undecided, the relationship between each system AP is circled.

First, the centralized control station AC performs processing for AP3 having a candidate with the highest throughput. In AP3, communication of the throughput index 160 can be expected by using channel A as shown in FIG. Therefore, the central control station AC selects channel A as the channel used by AP3. FIG. 9B shows the state of the table in this state. Since central control station AC has selected channel A as the channel of AP3, the interference state of AP3 has been changed from A to A.

Next, the centralized control station AC performs processing for AP6 having the next highest throughput after AP3. According to FIG. 9B, for AP6, it is necessary to determine a channel based on interference from channels A, X, Y, Z, α, β, χ, δ, and ε. Also, the throughput index needs to be updated from the values shown in FIG. Since it is determined that channel A is used by AP3, channels A, C, and D are shared with AP3. Therefore, according to Equation 5 and Equation 6 (G = 0.5, D = 1, Z = 0), ρ ₃ of the _third AP, AP3 is ρ ₃ = 1 / (m (3) +1 (3)) + 0 .5 / (m (3) +1 (3) +1) = 1/2 + 1/6 = 0.667. That is, new throughput candidates are A: 180 × 1 × 0.667 = 106.6, C: 80 × 1 × 0.667 = 53.3 (when sharing a primary channel with AP3) or 80 (AP3 channel) Is changed to channel D), D: 53.3 (when the primary channel is shared with AP3) or 80 (when the channel of AP3 is changed to channel C), E: 53.3, G: 40 (AP3 H: 40 (when the AP3 channel is changed to channel D), I: 40 (when the AP3 channel is changed to channel C), J: 40 (AP3 channel) ) Is obtained as O: 40,. Since channel A takes the largest value, centralized control station AC selects channel A as the channel of AP6. FIG. 9C shows the state of the table in this state.

Next, the centralized control station AC performs processing for AP2, which has the next highest throughput after AP6. The centralized control station AC determines a channel based on interference signals in channels A, T, U, V, W, X, Y, Z, α, β, χ, δ, and ε. Channel candidates and their throughput indices are obtained as follows. A: 160 × 0.5 × (1/3 + 1/8) = 36.7 (the penalty is 0.5 because AP of channel D of AP2 is not visible to AP3 and AP6 C: 80 × 1 × (1/3 + 1/8) = 36.7 (AP6 and AP3 channels are channel A, and the primary channel is channel C), because there are two independent primary channels (n = 2)) P to S) or 80 (when the channels of AP6 and AP3 are changed to channel D), D: 53.3 (when the channels of AP3 and AP6 are changed to channel C), E: 53. 3, G: 40 (when the channel of AP3 and AP6 is changed to channel D), H: 40 (when the channel of AP3 and AP6 is changed to channel D), O: 40. A high throughput index without decreasing the throughput index of another system AP is when channel E is selected as the channel of AP2. Therefore, the central control station AC selects channel E (primary channel: X) as the channel of AP2. FIG. 9C shows the state of the table in this state.
Here, the channel E is selected, but it can be determined based on various policies in the same manner as in the first embodiment in consideration of reducing the throughput of the system AP. For example, γ may be multiplied by the number of system APs that reduce the throughput, and the throughput index may be evaluated low. It is also possible to select a channel having the maximum sum or product with the throughput index of the system AP that affects the throughput.

  Next, the centralized control station AC performs processing for AP1, which has the next highest throughput candidate after AP2. The central control station AC is based on the interference of channels A, E (X), P, Q, R, S, T, U, V, W, X, Y, Z, and α from FIG. To determine the channel. Channel candidates and their throughput indices are obtained as follows. B: 160 × 1 × (1/3 + 1/8) = 73.3, F: 80, E: 80 × 1 × (1/3 + 1/8) = 36.7, N: 40, M: 40, O: 40. When channel F is selected, it is a high throughput index and does not decrease the throughput of the system AP. Therefore, the central control station AC selects channel F as the channel of AP2. FIG. 10A shows the state of the table in this state.

  The centralized control station AC finally performs processing for AP4. From FIG. 10A, AP4 is based on the interference in channels A, F, E (X), P, Q, R, S, T, U, V, W, X, Y, Z, α, δ, ε. To decide. Channel candidates and their throughput indices are obtained as follows. E: 80 × 1 × (1/3 + 1/8) = 36.7, F: 80 × 1 × (1/3 + 1/8) × 0.5 = 18.4, M: 40 × 1 × (1/2 + 1) /6)=26.7 (when AP1 sets β to χ as the primary channel), O: 40. Since the high throughput index is when channel O is selected, centralized control station AC selects channel O as the channel of AP4. FIG. 10B shows the state of the table in this state.

  In this way, the centralized control station AC of the second embodiment determines the channel of the system AP based on the primary channel and the secondary channel. Therefore, it is possible to assign a channel that increases the total throughput index of the wireless system 1 to each system AP. In this specific example, the throughput index of each system AP is obtained as 80, 53.3, 106.6, 40, 106.6, and the total throughput index is obtained as 386.7.

(Channel determination flow 2-2)
11 and 12 are diagrams illustrating a specific example (channel determination flow 2-2) of processing for determining a channel used by each base station apparatus. The specific examples shown in FIGS. 11 and 12 show a flow of processing for determining channels one by one in order from the system AP having a large number of detected interference AP channels. In FIG. 8, the use of channels between system APs is not considered. Therefore, as the channel of the system AP is determined, it is necessary to update the candidate channel and its throughput index. The detection result of the system AP of FIG. 7B and the detection result of the channel of the interfering AP of FIG. 7A are merged and represented on the vertical axis. The used channels of the detected system AP and interference AP are shown on the horizontal axis. The contents of the table at this point are the same as those shown in FIG. 10A.

  First, the centralized control station AC performs processing for the AP 4 having the largest number of detected channels in 20 MHz units. In AP4, communication of throughput index 53.3 is obtained by using channel E or F as shown in FIG. Here, the central control station AC may select either one. Further, the processing of the system AP having the next largest number of channel detections may be performed with two candidates remaining. Here, the centralized control station AC performs the processing of the next system AP as either E (X) or F (ε). FIG. 11A shows the state of the table in this state.

  The central control station AC performs processing for the AP 1 with the next largest number of detected channels (number of interference channels). Assuming that AP4 has selected E (X), the channel candidate and throughput index of AP1 are as follows: B: 160 × 1 × (1/3 + 1/8) = 73.3, F: 80, E: 80 × 1 × (1/3 + 1/8) = 36.7, N: 40, M: 40, O: 40. Assuming that AP4 has selected F (ε), the channel candidate and throughput index of AP1 are B: 160 × 0.5 × (1/3 + 1/8) = 36.7, F: 80 × 1 × (1/2 + 1/6) = 53.3, E: 80 × 1 × (1/2 + 1/6) = 53.3, N: 40 × 1 × (1/2 + 1/6) = 26.7, O: 40. If E (X) is selected as the channel of AP4 and F is selected as the channel of AP1, the throughput index of AP1 becomes the highest without decreasing the throughput of AP4. Therefore, the central control station AC sets the AP4 channel to E (X) and the AP1 channel to F. FIG. 11B shows the state of the table in this state.

  Next, the centralized control station AC performs processing for AP2, which has the largest number of detected channels after AP1. The channel candidates for AP2 and the throughput index are shown in FIG. 11B. A: 160 × 1 × (1/2 + 1/6) = 106.7, C: 80, E: 80 × 1 × (1/3 + 1/8) = 36.7, G: 40, H: 40, O: 40. Therefore, the centralized control station AC selects channel A (primary channel: T) as the channel of AP2.

  Next, the centralized control station AC determines the base station apparatus having the next largest number of detected channels after AP2. Since AP3 and AP6 have the same number of detected channels, the central control station AC selects one of the base station apparatuses to be processed. At this time, the centralized control station AC may randomly select a base station device to be processed. In addition, the centralized control station AC may execute the process with the system AP (AP6) having the largest number of observation channels and the system AP (AP6) in which communication cells overlap each other being processed first. The centralized control station AC may first execute the process on the system AP (AP3) in which the primary channel of the system AP (AP1) in which communication cells overlap with each other is not determined.

  Here, an example in which AP6 is selected will be described. In this case, the channel candidates and the throughput index are A: 160 × 1 × (1/2 + 1/6) = 106.7, C: 80 × 1 × (1/2 + 1/6) = 53.3, D: 80 × 1 * (1/2 + 1/6) = 53.3, E: 80 * 1 * (1/3 + 1/8) = 36.7, O: 40. However, when channels A, D, and C are selected as AP6 channels, a plurality of independent primary channels are generated in AP2 channel A (T), and the throughput index of AP2 becomes large due to the different channel penalty. descend. This is because the AP 6 does not receive the communication of the interfering AP whose channel T is the primary. Therefore, the channel candidates are C: 80 (AP2 is changed to channel D (T) and the throughput index of AP2 is decreased from 106.7 to 53.3), E: 36.6, and O: 40. The central control station AC selects a channel (channel O) from which the high throughput index can be obtained without affecting the throughput index of the system AP. Again, a channel can be selected according to a predetermined policy. Considering the product with the throughput index of the influential system AP, when channel C is selected, the throughput index of AP6 × the throughput index of AP2 is 80 × 53.3 = 4264, and 40 × 106.7 for channel O. = 4268, and also in this case, the channel O is selected.

  Finally, the centralized control station AC performs processing for the remaining AP3. Channel candidate and throughput index for AP3 are: A: 160 × 1 × (1/2 + 1/6) = 106.7, C: 80 (change AP2 channel to channel D (corrected from 106.7 to 53.3) )), D: 80 × 1 × (1/2 + 1/6) = 106.7, E: 80 × 1 × (1/2 + 1/6) = 53.3, O: 40 × 1 × (1/2 + 1 / 6) = 26.7. However, as in the case of AP6, since the throughput index of channel A of AP2 is greatly reduced, the centralized control station AC does not select channel A and channel D in the definition of the throughput index that is greatly affected by the different channel penalty. When channel E (X) is selected, channel E (X) from AP3 to AP4 is not observed, so AP2 and AP4 can be observed simultaneously, and channels (B, E, Check whether there is a system AP using K, L, X, Y, Z, α). Here, other system APs do not use these channels. Therefore, the centralized control station AC can select E (X) as the channel of AP3. Therefore, the centralized control station AC selects the channel E (X) that does not affect other system APs and provides high system throughput.

In this way, the channel of the system AP can be determined in consideration of the secondary channel. Therefore, for five system APs, the throughput index is obtained as 80, 106.7, 53.3, 53.3, 40, and the total throughput index is obtained as 333.3.
Further, the throughput of each system AP changes as described above depending on the order of the system APs to be processed. From this, the central control station AC may determine the channel in the selection order of the plurality of system APs and select the optimum channel according to the communication state. For example, when the channel of the system AP is determined as shown in FIGS. 9 and 10, the AP4 throughput is the lowest. On the other hand, when determined as shown in FIGS. 11 and 12, the throughput of AP5 is the lowest. For example, when the system AP having a high data communication frequency, data communication bit amount, priority, and the like is AP4, the central control station AC obtains the result of the processing shown in FIGS. The obtained channel arrangement is selected, and each system AP is notified to use the channel.

[Third Embodiment]
A third embodiment of the present invention will be described. The system AP in the third embodiment estimates information (communication frequency information) regarding the communication frequency of the interference AP in addition to the interference AP, the system AP, and the address information of the terminals that communicate with them. FIG. 13 is a diagram illustrating a channel detection result of an interference AP.

(Channel evaluation index 2)
When the central control station AC is instructed to estimate interference-related information, each system AP obtains information (communication frequency information) on the frequency of receiving communication packets of the interference AP, the time rate at which the access right is acquired, and the like. Estimate and get. That is, the communication frequency information is a specific example of interference relation information. The centralized control station AC acquires the frequency of communication of the interference APk as the interference frequency index Ψ _k . The interference frequency index Ψ _k may be a coefficient representing the communication frequency of the interference APk, or a coefficient representing the communication time rate of the interference APk and the communication frequency of the entire number of terminals belonging to the interference APk. The number of terminals belonging to the interference APk is U _k , the communication frequency of a terminal communicating with the i-th interference APk is Λ _{k, i,} and the communication frequency of the interference APk is Λ _{k, 0} . In this case, the interference frequency index Ψ _k is expressed as in Expression 10 below.

In Equation 10, if the communication frequency coefficient is a common constant, such as Λ ₀ = Λ ₁ =..., Λ, the sum of the AP and the terminal can be simply considered. If Λ ₀ = 0, only the number of terminals can be considered.

Further, a communication frequency index Φ _j for the j-th system AP is determined. The communication frequency index Φ _j may be common to all system APs, with Φ _j = Φ. Further, from the number of terminals belonging to each AP and the communication frequency, the communication frequency index Φ _j may be expressed as in Expression 11 below.

Equations 10 and 11 can also be evaluated for each channel at 20 MHz. In this case, U _k or U _j is the number of terminals communicating with the interfering APk operating on the channel or the number of terminals communicating with the system APj.
As an example, when Λ _{j, 0} = 0, Λ _{j, i} = 10 and the number of terminals with which APs 1, 2, 3, 4, 6 communicate are 10, 2, 5, 7, 3, The channel determination method is shown. The communication frequency index for each system AP (AP1, AP2, AP3, AP4, AP6) is 100, 20, 50, 70, 30 respectively.

When a certain system APj tries to select a channel, the throughput of APj is calculated using the communication frequency index, the interference frequency index, and the different channel penalty in the 20 MHz channel having the highest communication frequency among the base station devices that affect the channel. The index T _j is expressed as in Expression 12.

θ is a set of system APs belonging to the channel selected by APj. ψ is a set of interfering APs belonging to the channel selected by APj. Furthermore, based on the sum of the communication frequency and the interference frequency obtained as Equation 13 below, the correction by a function f of the [pi _j as in Equation 14 may be performed.

By adding the value obtained by the function of f (Π _j ) to Φ _j / Π _j , the expected value of throughput can be increased or decreased. Increasing the expected value of throughput by making f (Π _j ) positive does not necessarily mean that communication actually occurs at the same timing, but if communication occurs at mutually independent timings and they happen to be shifted in time This is because the communication is not deteriorated. This is because, when f (Π _j ) is negative and the throughput is lowered, the execution throughput is lowered rather than simply dividing by time due to packet collision or the like. For this reason, the centralized control station AC changes f ( _{ｊ j} ) according to the communication status, increases the amount of data to be communicated, takes a negative value when the line gets tight, and is positive when the communication frequency is low. It can also be adaptively controlled to take the value of. Similarly, central control station AC, like the equation 15 and equation 16, may be corrected by the function g or function h [pi _j.

The centralized control station AC increases the expected value of throughput by multiplying Φ _j / Π _j by a value obtained by a function of g (Π _j ) or by factoring h (Π _j ). It may be lowered or lowered. Increasing the expected value of throughput by making g (Π _j ) greater than 1 or making h (Π _j ) less than 1 does not necessarily mean that communication actually occurs at the same timing, but timings independent of each other This is because if they communicate with each other, and they happen to be out of time, they will not suffer communication degradation. When g (Π _j ) is made smaller than 1 or h (Π _j ) is made larger than 1 to reduce the throughput, it is considered that the execution throughput is lowered rather than simply dividing by time due to packet collision or the like. It is to do. The central control station AC may change g (Π _j ) and h (Π _j ) depending on the communication status, as in f (Π _j ).

  In the third embodiment, as described above, channels including candidates with different frequency bands can be determined based on the communication frequency of the interference AP. For channel selection, the channel index that can be taken by the system AP can calculate a throughput index for all combinations, and can determine an appropriate channel combination, or the system AP can determine channels in order. You can also.

  Based on the example of FIG. 13, a channel determination method of the system AP will be described. In FIG. 13, the numerical value shown below the symbol indicating the primary channel or the secondary channel is the interference communication frequency coefficient. Although the interference communication frequency coefficient for the same interference AP is the same, each system AP may take different values depending on the system AP in order to evaluate the reception frequency of the communication packet of the interference AP. In addition, there is an example in which the interference communication frequency coefficient is different between the primary channel and the secondary channel because the terminal with which the interference AP is communicating does not necessarily use all the channels acquired by the system AP or has no function to use. Because there is.

(Channel decision flow)
In the third embodiment, a specific example of processing in which the central control station AC determines a channel in order for each system AP will be described. The order of the system APs for determining the channel is the same as in the first embodiment and the second embodiment described above. In this description, processing is performed in descending order of the sum of communication frequencies of interference APs.
When the sum of interference frequencies is taken for all channels, 90 × 4 + 20 × 8 + 30 × 8 = 760, 20 × 8 + 30 × 6 + 20 × 2 + 10 × 2 = 400, 30 × 4 + 70 × 2 + 30 × 2 + 10 × 2 = for AP1 to AP6, respectively. 340, 90 × 4 + 20 × 10 + 30 × 8 = 800, 30 × 6 + 70 × 2 + 20 × 2 + 10 × 2 = 380.

In this description, processing is performed in order from the worst interference condition (that is, in order of increasing sum of communication frequencies of interference APs). Therefore, processing for determining channels in the order of AP4, AP1, AP2, AP6, and AP3 is performed.
First, the centralized control station AC calculates the channel candidate and throughput index of AP4 based on Equation 12. The channel candidate and throughput index of AP4 are: A: 160 × 0.5 × (70 / (70 + 90 + 20)) = 31.1, B: 160 × 0.5 × (70 / (70 + 30 + 20)) = 46.7, C : 80 × 0.5 × (70 / (70 + 90 + 20)) = 15.6, D: 80 × 0.5 × (70 / (70 + 30 + 20)) = 23.3, E: 80 × 1 × 70 / (70 + 30) = 56, F: 80 × 1 × 70 / (70 + 20) = 62.2, G: 7.8, (H: 7.8), I: 11.7, (J: 11.7), K: 40 × 1 × 70 / (70 + 30) = 28, (L: 28), M: 40, N: 31.1, O: 40. Since the high throughput index is obtained from the channel E or F, the central control station AC selects one of these as the channel of the AP 4. The centralized control station AC may select a channel F having a vacant channel M, may select one at random, or may proceed to processing of the next system AP while holding a plurality of candidates. good. In this description, the centralized control station AC will explain the flow in the case where both E (X) and F (ε) are held as candidates and the process proceeds to the next system AP process. Here, an example is shown in which the centralized control station AC stores a plurality of candidates for the same throughput index, but even if the same throughput index is not obtained, a plurality of candidates having a high throughput index are stored, and other candidates are stored for each. It may be evaluated what value the throughput index of the system AP takes.

  Next, the centralized control station AC evaluates the throughput index of AP1, which has the second most interference condition after AP4. When AP4 selects channel E (X), the throughput indices are channel A: 160 × 0.5 × (100 / (100 + 90 + 20)) = 38.1, B: 160 × 1 × (100 / (100 + 70 + 30)) = 80, C: 80 × 0.5 × (100 / (100 + 90 + 20)) = 19, D: 80 × 0.5 × (100 / (100 + 30 + 20)) = 53.3, E: 80 × 1 × (100 / (100 + 30 + 70)) = 40, F: 80, K: 40 × 1 × (100 / (100 + 30 + 70)) = 20, (L: 20), M: 40, N: 40, O: 40. On the other hand, when the AP 4 selects the channel F (ε), the throughput index is A: 160 × 0.5 × (100 / (100 + 90 + 20)) = 38.1, B: 160 × 0.5 × (100 / ( 100 + 70 + 30)) = 40, C: 19, D: 80 × 0.5 × (100 / (100 + 30 + 20)) = 53.3, E: 80 × 1 × (100 / (100 + 30)) = 61.5, F: 80 × 1 × (100 / (100 + 70)) = 47.1, K: 30.8, (L: 30.8), N: 40 × 1 × (100 / (100 + 70)) = 23.5, O: 40 is obtained.

  As described above, the throughput index 80 is obtained by selecting the channel E (X) as the channel of the AP4 and selecting the channel B (X) or F (ε) as the channel of the AP1. As to which channel B (X) or F (ε) should be selected as the channel of AP1, the central control station AC may select a channel (channel F) with a smaller frequency bandwidth. By selecting to obtain a high throughput with a small bandwidth, it is possible to prevent unnecessary interference from being given to other systems.

  FIG. 14 shows a communication frequency detection table when channel E (X) and channel F (ε) are selected for AP4 and AP1, respectively. When the channel of the system AP is determined in this way, the interference condition increases accordingly. Therefore, each time the channel of the system AP is determined, the centralized control station AC updates the number of channels to be measured and the sum of interference frequencies to determine which system AP channel of the remaining system APs. You can also change the order. At the time shown in FIG. 14, the system APs whose channels are not determined are AP2, AP3, AP6, and the sum of interference frequencies is 20 × 10 + 100 × 4 + 30 × 6 + 70 × 4 + 10 × 2 = 1080, 30 × 4 + 100. * 4 + 70 * 2 + 20 * 2 = 700, 30 * 6 + 70 * 6 + 20 * 2 + 10 * 2 = 660.

  Next, the central control station AC performs processing for AP2. Channel candidates and throughput indices are: A: 160 × 1 × (20 / (20 + 20)) = 80, (C: 80 × 1 × (20 / (20 + 20)) = 40), D: 80 × 1 × (20 / (20 + 20)) = 40, E: 80 × 1 × (20 / (20 + 30 + 100)) = 10.7, F: 80 × 0.5 × (20 / (20 + 20 + 30 + 70)) = 11.4, O: 40 It is done. Since the highest throughput is obtained by selecting channel A (T), the central control station AC selects channel A (T) as the channel of AP2. FIG. 15 shows a communication frequency detection table when channel A (T) is selected for AP2.

  Next, the central control station AC performs processing for AP6. The channel candidates and the throughput index are derived as follows from the relationship of FIG. A: 160 × 1 × (30 / (30 + 20)) = 96, B: 160 × 0.1 × (30 / (30 + 30 + 70 + 70 + 30)) = 2.1, (C: 80 × 1 × (30 / (30 + 20)) = 48), D: 80 × 1 × (30 / (30 + 20)) = 48, E: 80 × 1 × (30 / (30 + 30)) = 40, F: 80 × 0.5 × (30 / (30 + 30 + 70 + 70) ) = 6, O: 40. Here, although channel A is a high throughput index, when channels A, C, and D are selected as channels of AP6, AP2 that is a system AP has two independent APs in the same channel. When the different channel penalty is set to a large value, the throughput index of the system AP is greatly reduced, so that the channels A, C, and D are not selected as the channels of the AP 6. By selecting channel E (X) or selecting channel O, the throughput index 40 can be obtained. Here, the centralized control station AC may select the channel O as a channel having a smaller channel bandwidth. The central control station AC may construct two independently operating APs on the channel E used by the interfering AP by using the channel E. That is, the interference AP (AP5) of channel E detected by AP4 and AP1 is the same AP, and AP4 and AP1 are in a positional relationship where they cannot be observed. Thus, access rights can be acquired independently of each other. Therefore, when AP4 selects channel E, it becomes difficult for the interfering AP (AP5) to acquire an access right due to a different channel penalty. Therefore, the communication frequency of AP5 is effectively reduced from 30, and AP1 and AP4 Can be expected to increase the throughput.

FIG. 16 shows the relationship when the central control station AC selects F (ε) as the channel of AP4. Next, the centralized control station AC performs processing for AP3. The channel candidate and throughput index of AP3 are A: 160 × 1 × (50 / (50 + 20)) = 114.3, (C: 80 × 1 × 50 / (50 + 20) = 57.1), D: 80 × 1 × 50 / (50 + 20) = 57.1, F: 80 × 0.5 × 50 / (50 + 70 + 20) = 14.3, O: 40 However, when channels A, C, and D are selected as the channels of AP3, two primary channels that operate independently are generated in channel A (T) of AP2 that is the system AP. Therefore, the central control station AC does not select the channel as the channel of AP3. Therefore, the central control station AC selects the channel O that is the remaining channel.
Alternatively, the channel can be determined according to another policy. Even if the central control station AC changes the channel of AP2 from A (T) to D (T) and decreases the throughput index of AP2 from 80 to 40, channel C: 80 is selected as the channel of AP3. good. Considering the product of throughput indices, the throughput index of AP2 × the throughput index of AP3 is equal to 40 × 80 = 3200 when channel C is selected, and 80 × 40 = 3200 when channel O is selected. Either may be determined at random, or may be determined so that the throughput index 80 is obtained for the system AP having a high priority.
Further, the central control station AC may switch the channel combinations of the channels AP2 and AP3 at regular intervals.

  In this way, the channel of each system AP is selected by the centralized control station AC as AP1: 61.5, AP2: 80, AP3: 40, AP4: 56, AP6: 40. Since AP1 and AP6 tend to lower the throughput of interference AP on channel E, the throughput of AP1 and AP6 is expected to be higher when the communication frequency index of AP7 is lower than 30.

  Further, with the above method, it is not possible to use channel A that can be used without the influence of the interfering AP in AP3 and AP6 that have been observed in FIG. This is because AP2 uses channel A and AP2 observes the interfering AP, so that AP3 and AP6 can no longer select a channel related to channel A. For this reason, it is important for maximizing system throughput to determine a free channel in a large frequency band with the highest priority, and the maximum throughput that each system AP can take is calculated first and determined from the one with the highest throughput. Thus, system throughput can be increased.

  In addition, an example in which the secondary channel indicated in parentheses in the second embodiment and the third embodiment is selected as a channel by the above method will be described. When the secondary channel is selected, there is a risk that the other base station apparatus does not recognize a control signal such as RTS / CTS and the throughput is greatly reduced. Therefore, normally, the secondary channel is not selected. However, by using two independent APs in which the base station apparatus and the communication cell overlap and operating on the primary channel and the secondary channel, an independent primary channel can be generated in the channel of the interfering AP. it can. Therefore, the throughput of the system AP can be increased by reducing the throughput of the interference AP.

  FIG. 17 changes from the relationship of FIG. 13 that the channel of AP0 is channel C (Q) and the channel of AP7 is channel A (T), and the communication frequencies are 90 and 100, respectively. It is a figure which shows the specific example of a table | surface at the time of assuming that a signal arrives. In the case of FIG. 17, AP7 is detected by the system AP (AP1, AP2, AP3, AP4), and the channel selection of the system AP is limited. When such a terminal exists, for example, by selecting channel D (T) as the channel of AP1 and selecting channel C as the channel of AP2, two independent base station apparatuses are generated in the channel of AP7. And make it difficult to obtain access rights. By doing so, the throughput of the main interference source can be reduced. If channel D (T) is selected as the channel of AP3 and AP4 and channel C is selected as the channel of AP2, AP3 and AP4 are out of the communication area with each other, so that the primary channel that operates independently within the channel of AP7 3 can be generated, and the communication of the interference source can be greatly restricted.

[Fourth Embodiment]
Next, a fourth embodiment of the present invention will be described. The centralized control station AC in the fourth embodiment further performs processing based on the performance with respect to the frequency band of the terminal with which the system AP is communicating.

  In the description of the first to third embodiments, the frequency band index B uses the frequency bandwidth of the channel. However, in the fourth embodiment, the utilization frequency band index B is calculated using the function of the terminal and the frequency bandwidth of the channel. The evaluation method in the case where the terminals with which the system AP communicates has various frequency bandwidth functions from 20 MHz to 160 MHz will be described. At this time, even if a 160 MHz channel is assigned to a system AP whose terminal frequency bandwidth function is 20 MHz, only up to 20 MHz is used.

(Channel evaluation index 3)
For the reasons described above, the value B _j obtained by the following Expression 17 can be applied as B _j in Expression 1.

Here, B _c is a number corresponding to the frequency bandwidth of the allocated channel or the number of subcarriers. Here, for simplicity, it is assumed that B _c = 160 for a 160 MHz channel. B _{j, i} is a function of the frequency bandwidth of the i-th terminal with which the j-th system AP is communicating. For example, when the terminal has a function that considers up to frequency bands of 20 MHz, 40 MHz, and 80 MHz, B _{j and i} are set to 20, 40, and 80, respectively. U _j is the number of terminals with which the j-th system AP is communicating.

第４実施形態におけるｊ番目のシステムＡＰは、通信を行う端末の周波数帯域情報Ｂ_ｊ，ｉや式１７や式１８から得られるＢ_ｊや端末の通信頻度σ_ｊ，ｉを集中制御局ＡＣに通知する。 As B _{j in} Expression 1, a value B _j obtained by Expression 18 below may be applied. _{Incidentally, σ j, 1 ~σ j,} Uj is the communication frequency of the terminal I through U _j.

J-th system AP in the fourth embodiment, the frequency band information B _j of the terminal that performs _{communication, i} and equation 17 and the communication frequency of B _j or terminal from Equation 18 sigma _{j, i} to the central control station AC Notice.

(Channel decision flow)
The example which selects a channel in 4th Embodiment is shown. In the examples so far, the channel bandwidth and the throughput are proportional. On the other hand, in the fourth embodiment, the throughput B _j is calculated using Equation 18 based on the function information of the terminals of each system AP. In Equation 18, the communication frequency is an equal value 1 / U _j . FIG. 18 is a table showing the function of each terminal. It is assumed that the number of terminals with which APs 1, 2, 3, 4, and 6 communicate is 10, 2, 5, 7, and 3, and the function of each terminal is as shown in FIG. Processing is performed in order from the system AP with the highest communication frequency. Therefore, processing is performed in the order of AP1, AP4, AP3, AP6, AP2. The interference condition from the interference AP is as shown in FIG.

First, the central control station AC evaluates the throughput index of AP1 having the largest communication frequency index as follows.
Channel A: (5 × 160 + 3 × Min (160, 80) + 2 × Min (160, 40)) / 10 × 0.5 × (100 / (100 + 90 + 20)) = 26.7, B: (5 × 160 + 3 × Min (160,80) + 2 × Min (160,40)) / 10 × 1 × (100 / (100 + 70)) = 86.2, C: (5 × Min (80,160) + 3 × Min (80,80) + 2 × Min (80,40)) / 10 × 0.5 × (100 / (100 + 90 + 20)) = 17.1, D: (5 × Min (80,160) + 3 × Min (80,80) + 2 × Min (80, 40)) / 10 × 0.5 × (100 / (100 + 30 + 20)) = 24, E: (5 × Min (80,160) + 3 × Min (80,80) + 2 × Min (80,40) ) / 10 × 1 × (100 / (100 + 30)) = 55. F: 80, K: (5 × Min (40,160) + 3 × Min (40,80) + 2 × Min (40,40)) / 10 × 1 × (100 / (100 + 30)) = 30.8 (L: 30.8), M: 40, N: 40, O: 40

  The centralized control station AC can select the channel B with the highest throughput index as the channel of AP1. However, the function of the AP1 terminal does not necessarily support 160 MHz. Therefore, the difference in throughput index between the selection of the channel B of 160 MHz and the channel F of 80 MHz is slight (86.2-80 = 6.2). The central control station AC may select a narrow-band channel when only a slight effect is obtained even when the double band is used. Specifically, a threshold value is set in advance, and the central control station AC changes the channel to be selected according to whether the degree of the effect (for example, the index difference value or the index ratio) is larger than this threshold value. You may do it. Here, if the threshold is set as 10%, selection of channel B cannot be said to have a sufficiently large effect on selection of channel F. Therefore, in this case, the central control station AC selects the channel F.

  Next, the central control station AC evaluates the throughput index of AP4 as follows. A: (5 × 160 + 1 × Min (160, 80) + 1 × Min (160, 40)) / 7 × 0.5 × (70 / (70 + 90 + 20)) = 25.6, B: (5 × 160 + 1 × Min ( 160,80) + 1 × Min (160,40)) / 7 × 0.5 × (70 / (70 + 100 + 30 + 30)) = 20, C: (5 × Min (80,160) + 1 × Min (80,80) +1 × Min (80, 40)) / 7 × 0.5 × (70 / (70 + 90 + 20)) = 12.8, D: (5 × Min (80,160) + 1 × Min (80,80) + 1 × Min ( 80, 40)) / 7 × 0.5 × (70 / (70 + 30 + 20)) = 21.7, E: (5 × Min (80,160) + 1 × Min (80,80) + 1 × Min (80,40) )) / 7 × (70 / (70 + 30)) = 52, F: (5 × Min (80, 60) + 1 * Min (80,80) + 1 * Min (80,40)) / 7 * 0.5 * (70 / (70 + 100 + 30)) = 13, K: (5 * Min (40,160) + 1 * Min (40,80) + 1 × Min (40,40)) / 7 × 1 × (70 / (70 + 30)) = 26, (L: 26), (M: 40 × 0.5 × 70 / (70 + 100) = 8.3), N: 40 × 0.5 × 70 / (70 + 30 + 100) = 7, O: 40.

  Channel E (X) provides the highest throughput. Therefore, the central control station AC selects channel E (X) as the channel of AP4.

FIG. 19 is a diagram illustrating a communication frequency detection table when channel F is selected for AP1 and channel E is selected for AP4.
Next, the central control station AC evaluates the throughput index of AP3 as follows. A: (3 × 160 + 2 × Min (160, 80)) / 5 = 128, C: (3 × Min (80, 160) + 2 × Min (80, 80)) / 5 = 80, D: 80, E: 80 × 1 × 50 / (50 + 30) = 50, O: 40.

The highest throughput is obtained by selecting channel A. Therefore, the central control station AC selects channel A as the channel of AP3. The throughput when channel A is selected is higher than the threshold (10%) than when channel C is selected.
Next, the central control station AC evaluates the throughput index of the AP 6 as follows. A: (2 × 160 + 1 × Min (160, 40)) / 3 × 1 × 30 / (30 + 50) = 45, C: (2 × Min (80, 160) + 1 × Min (80, 40)) / 3 × 30 / (30 + 50) = 25, D: 25, E: (2 × Min (80,160) + 1 × Min (80,40)) / 3 × 1 × 30 / (30 + 70 + 30) = 15.4, O: 40 .

The highest throughput is obtained by selecting channel A. Therefore, the centralized control station AC selects channel A as the channel of AP6.
Finally, the centralized control station AC performs processing for AP2. AP2 supports up to 40 MHz at the maximum terminal function. Therefore, it can be evaluated as follows based only on a channel of 40 MHz or less. I: (Min (40,40) + Min (40,20)) / 2 × 0.5 × 20 / (20 + 50 + 30) = 2.5, K: (Min (40,40) + Min (40,20)) / 2 × 1 × 20 / (20 + 70 + 30) = 5, N: (Min (40,40) + Min (40,20)) / 2 × 0.1 × 20 / (20 + 100 + 30 + 20) = 0.4, O: (Min ( 40,40) + Min (40,20)) / 2 = 30
The centralized control station AC selects the channel O that is the highest throughput index for the AP2.

In this way, the centralized control station AC has AP1: 80, AP2: 40, AP3: (3 × 160 + 2 × Min (160,80)) / 5 × 50 / (50 + 30) = 80, AP4: 52, AP6: A channel arrangement that obtains a throughput index of 45 can be determined.
Further, the information on the interference relationship between APs as shown in FIGS. 3, 7 and 13 is obtained by instructing from the central control station AC, stopping communication of the system AP, and using the received signal from the adjacent interference AP. Also good. Alternatively, the central control station AC may instruct the system AP to select the same frequency channel, and the mutual interference relationship may be measured and acquired. Further, it may be acquired from information on a channel selected by the system AP for communication or a channel that can be detected by setting a reception band wider than a transmission band. When collecting information on the channels used by each of the latter system APs for communication, information on interference APs and information on system APs cannot be obtained completely as shown in FIGS. However, even if it is incomplete, the method of the present embodiment determines the order of channel determination of the system AP, and each system AP has a higher throughput index from the information in the range acquired by the central control station AC. The channel can be determined.

(principle)
FIG. 20 is a flowchart showing the flow of processing of the channel determination method in the present invention. Hereinafter, the channel determination method in the present invention will be described with reference to FIG. The central control station AC collects channel information used by the interference AP and communication frequency information from each system AP. Then, the central control station AC stores interference relation information representing the influence of the interference AP on the system AP as shown in FIGS. 3, 7 and 13 (step S101).

  When the interference related information is obtained, the centralized control station AC determines the channels of the system AP in order. For example, the centralized control station AC calculates a throughput index using Equations 1 to 16, and determines the order of channel determination of the system AP based on the calculation result. Specifically, the order of increasing throughput index or the order of decreasing throughput index may be used. Further, the central control station AC does not necessarily determine the order of channel determination based on the throughput index, and may determine the order in the following order, for example. The order of the number of primary channels that can detect communication. The order in which the number of primary channels is small. The order of the sum of the indicators of the communication frequency of the observed interfering APs is large. The order in which the sum of the above indicators is small. The order in which the number of terminals performing communication is large. The order in which the number of terminals performing communication is small. Order of priority determined in advance for the system AP. The order of the most data bits flowing to the system AP. The order in which the functions of the terminals with which the system AP communicates is high. The above functions are in ascending order. The order in which the amount of bits stored in the buffer is large. The order with the most packet collisions.

The central control station AC initializes the value of the counter. Specifically, the central control station AC sets j = 1 (step S102).
The central control station AC calculates a throughput index for the selectable channel for the j-th system AP based on any one or a plurality of formulas 1 to 16 (step S103). The central control station AC determines the channel having the highest throughput index as the channel of the system AP (step S104). If the channels of all the system APs have not been determined, the centralized control station AC updates the interference relationship information based on the channel assignments determined so far, sets j = j + 1, and returns to the process of step S103 ( Step S106).

When the central control station AC determines the channels of all the system APs, the centralized control station AC notifies each system AP of the determined channels (step S107). Each system AP changes the channel used for wireless communication with the terminal to the channel notified from the central control station AC, and starts communication (step S108).
The timing at which the above-described processing is executed may be executed from step S101 at regular intervals. In addition, the process of step S101 may always be repeatedly executed, and the processes after step S102 may be executed at regular time intervals. Each system AP notifies the centralized control station AC of the changed information when the communication state such as the channel used by the interference AP and the communication frequency of the packet signal is changed (step S109). The station AC may use this notification as a trigger for starting step S101.

The change in communication state detected by each system AP in step S109 is, for example, the following state. When it is detected that a new interfering AP has started communication or beacon transmission on the observable channel in the system AP. When it is detected that communication or beacons from interfering APs that have been communicating for a certain time or longer cannot be received. When it is detected that the time occupancy rate of a communication packet of an interfering AP using the same channel has changed for a predetermined time. When the number of terminals belonging to the interference AP changes by a predetermined value or more. When it detects that the interfering AP is sending a notification to change the channel.
Further, instead of determining one by one as in S102 to S106, the combination of channels that can be taken by the system may be checked in a round-robin manner, and the best combination may be determined as channel assignment to each system AP. . In this case, channels for all system APs are determined in S102, and each system AP is notified in S107.

<Modification>
Even in the case of a device other than the interference AP described above, when there is a device (interference device) that outputs an interference signal such as a radar, the system AP can autonomously change the channel. In addition, after the system AP that has observed the signal from the interference device autonomously changes the channel, it notifies the central control station AC that the channel has been changed, and does not use the channel used by the interference device. The central control station AC may newly determine the channel of the system AP.

In addition, when a new system AP is connected to the AC, it is possible to notify that the new system AP is connected to the AC, and start from S101 to re-determine the channel of the system AP.
In the wireless system 1, the system AP and the interference AP may be detected at the same time. Specifically, it is performed as follows. The system AP observes the status of available channels at a timing specified in advance by the centralized control station AC. Based on the received beacon or communication packet, the system AP collects information such as the address of the transmission source terminal and the address of the base station apparatus to which the transmission source terminal is connected. By collecting these pieces of information, the interference AP is detected. At this time, if each system AP also transmits a beacon, a communication packet, and the like on the observed channel, it is possible to collect information on each system AP. That is, each system AP can simultaneously detect other system APs and interference APs. Each system AP notifies the collected information to the central control station AC. The central control station AC determines whether the address notified from each system AP is an address of a terminal or system AP belonging to the system AP. This determination makes it possible to determine whether each piece of information is information related to the interference AP or information related to the system AP.

In step S104, when the central control station AC determines the channel of the j-th system AP, the highest throughput among cases where the channel corresponding to the highest throughput index selects a channel with a lower frequency bandwidth. It may be determined whether or not the index is higher than a predetermined value. If the value is higher than a predetermined value, the central control station AC may select the channel corresponding to the highest throughput index as the channel of the j-th system AP. On the other hand, if it is not higher than the predetermined value, the centralized control station AC selects the channel that becomes the highest throughput index among the channels of the lower frequency bandwidth as the channel of the j-th system AP. Also good. By selecting the channel in this way, it is possible to avoid using a broadband channel with low efficiency.
In step S104, the central control station AC can consider a penalty when reducing the throughput index of another system AP when determining the channel of the j-th system AP. By multiplying the coefficient γ by the number of system APs that reduce the throughput, or by using a policy that selects a channel having a high system throughput sum or product value, the throughput index of other system APs is increased. Decreasing can be prevented.

  In step S104, when the central control station AC determines the channel of the j-th system AP, priority is given to a channel that is a condition for allowing a plurality of independently operating primary channels to exist in the channel of the interfering AP. You may do it.

  FIG. 21 is a flowchart showing the flow of processing of a modification of the channel determination method in the present invention. In the flowchart shown in FIG. 21, q types of channel determination order criteria for the system AP are set in advance in the centralized control station AC. The centralized control station AC determines a determination order according to each criterion, and determines a channel by the processing of S102-1 to S106-q according to each determination order. The centralized control station AC may select an optimal channel arrangement based on the obtained q channel assignments and the throughput index (step S200). The optimum channel arrangement may be the one that maximizes the sum of the throughput indices, or the minimum value of the throughput indices may be the maximum, or the throughput index of the priority system AP is maximized. It may be a thing.

  Further, when the interference relationship information between the system APs cannot be obtained, the centralized control station AC may perform the above processing on the assumption that the system APs detect signals with each other. In a similar case, the central control station AC may perform the above-described process by separately acquiring the position information of the system AP and creating an interference relationship.

Further, when the transmission power per band differs depending on the frequency bandwidth of the channel selected for communication, the interference relationship between the system APs shown in the lower diagram of FIG. 3 is prepared depending on the channel bandwidth condition. You can also. For example, in the case of a system in which transmission power per band decreases as the frequency bandwidth increases, the interference relationship decreases as the bandwidth increases, and the interference relationship increases as the bandwidth decreases. The central control station AC may store a table representing such an interference relationship in advance for each bandwidth.
Further, when the central control station AC selects a channel, the central control station AC may determine the channel on the condition that a channel that includes a plurality of different primary channels is not selected in the channel.

  It is also possible to select a channel in consideration of the influence of interchannel interference on adjacent channels of the channel being used. FIG. 22 is a diagram showing a table in which a star is added to a channel in which characteristic deterioration occurs due to the influence of inter-channel interference in the communication channel detection result table of FIG. When using a certain channel, it is designed so that the signal is attenuated by a band-pass filter, etc. so that a large interference signal does not appear in the adjacent channel. Depending on the relationship, it is conceivable that inter-channel interference occurring in adjacent channels is not sufficiently suppressed and the characteristics of the channel are deteriorated. In the channel indicated by the asterisk in FIG. 22, there is a possibility that the communication of the adjacent channel is leaked without being sufficiently attenuated. Therefore, a condition is added so that the channel indicated by the star is not selected, the attenuation coefficient ω is multiplied by the number of stars (ω is 0 <ω <1), or the throughput penalty ω ′ is subtracted. The throughput index when the channel is selected can be reduced or the like according to the communication frequency of the main signal causing the inter-channel interference (the number in parentheses in FIG. 22).

  Further, the channel can be determined without using the information on the interference relationship between the system APs. In this method, as information in FIG. 2B and FIG. 7B, all the system APs are in an interference relationship, or all the APs are not in an interference relationship, or system APs that are similar to the detected interference AP are mutually connected. It can be detected. As a determination of whether or not the detected interference AP is similar, it can be considered that the system APs whose detected AP types are the same by X percent or more are in the mutually detectable area. For example, in the example of FIG. 7, assuming that the system APs in which the detected interference AP is the same by more than half (50%) are in the signal detection area, the interference AP that can be detected by AP1 is Since they are AP0, AP5, and AP7, it is possible to know the interference relationship between AP2 and AP4, assuming that they are in the signal detectable area with each other, without actually receiving signals or receiving each other's signals.

  In the wireless system 1 configured as described above, in the centralized control station AC that controls the frequency channels of a plurality of system APs, when there is a device (interference AP) that uses the same frequency in the vicinity, each system is determined from the interference condition. The frequency channel to be used by the AP is determined, and the system throughput can be improved.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes designs and the like that do not depart from the gist of the present invention.

1 ... Wireless system, AC ... Central control station, AP ... Base station equipment

Claims (8)

A wireless communication system comprising a centralized control station and a plurality of base station devices connected to the centralized control station, and performing wireless communication in an environment where an interference signal is received in the base station device,
The base station device
Interference estimating means for estimating interference relation information which is information relating to the interference signal;
Interference relation information notifying means for notifying the central control station of the estimated interference relation information;
Channel setting means for changing a channel used for wireless communication with the terminal device to a channel notified from the central control station,
The central control station is
For each base station apparatus, a throughput index of two or more channel candidates is calculated from channel candidates including a plurality of channels having different frequency bands, and the channel of the base station apparatus is calculated based on the calculated throughput index. Channel determining means for determining;
Channel notification means for notifying the determined channel to the base station apparatus ,
The channel determination means is a device other than the terminal device that is communicating with the base station device and is communicating with the base station device in a frequency band range of a channel candidate of the connected base station device. A wireless communication system that calculates the throughput index based on the number or communication frequency and the frequency bandwidth or number of data subcarriers of the channel candidate . A wireless communication system comprising a centralized control station and a plurality of base station devices connected to the centralized control station, and performing wireless communication in an environment where an interference signal is received in the base station device,
The base station device
Interference estimating means for estimating interference relation information which is information relating to the interference signal;
Interference relation information notifying means for notifying the central control station of the estimated interference relation information;
Channel setting means for changing a channel used for wireless communication with the terminal device to a channel notified from the central control station;
A terminal including at least one of the number of the terminal devices communicating with the own device, the frequency bandwidth or the number of data subcarriers used for communication with the terminal device, and the communication frequency information of the terminal device information, and a terminal information notification means for notifying the central control station,
The central control station is
For each base station apparatus, a throughput index of two or more channel candidates is calculated from channel candidates including a plurality of channels having different frequency bands, and the channel of the base station apparatus is calculated based on the calculated throughput index. Channel determining means for determining;
Channel notification means for notifying the determined channel to the base station apparatus,
It said channel determination unit calculates the throughput indicators based on the terminal information, radio communications system. A wireless communication system comprising a centralized control station and a plurality of base station devices connected to the centralized control station, and performing wireless communication in an environment where an interference signal is received in the base station device,
The base station device
Interference estimating means for estimating interference relation information which is information relating to the interference signal;
Interference relation information notifying means for notifying the central control station of the estimated interference relation information;
Channel setting means for changing a channel used for wireless communication with the terminal device to a channel notified from the central control station,
The central control station is
For each base station apparatus, a throughput index of two or more channel candidates is calculated from channel candidates including a plurality of channels having different frequency bands, and the channel of the base station apparatus is calculated based on the calculated throughput index. Channel determining means for determining;
Channel notification means for notifying the determined channel to the base station apparatus;
Without fixed time communication to the base station apparatus, and a communication stopping means for notifying to collect interference related information,
Before Kihi Wataru estimating means, when the specified stop communication from the central control station, to stop the data communication, transmission source information or communication packet is notified in the beacon received during communication stop in its own device address, destination address of the communication packet, belongs address of the base station apparatus collects the time occupancy of the communication packet, at least one of the interference-related information, radio communications system. A wireless communication system comprising a centralized control station and a plurality of base station devices connected to the centralized control station, and performing wireless communication in an environment where an interference signal is received in the base station device,
The base station device
Interference estimating means for estimating interference relation information which is information relating to the interference signal;
Interference relation information notifying means for notifying the central control station of the estimated interference relation information;
Channel setting means for changing a channel used for wireless communication with the terminal device to a channel notified from the central control station,
The central control station is
For each base station apparatus, a throughput index of two or more channel candidates is calculated from channel candidates including a plurality of channels having different frequency bands, and the channel of the base station apparatus is calculated based on the calculated throughput index. Channel determining means for determining;
Channel notification means for notifying the determined channel to the base station apparatus;
And a interferometric measuring designating means for designating a common channel to said base station apparatus,
Before Symbol interference estimation means uses the channel specified from the central control station, it determines whether it is possible detect the signal transmitted from each base station apparatus, acquires the interference related information, radio communications system. A wireless communication system comprising a centralized control station and a plurality of base station devices connected to the centralized control station, and performing wireless communication in an environment where an interference signal is received in the base station device,
The base station device
Interference estimating means for estimating interference relation information which is information relating to the interference signal;
Interference relation information notifying means for notifying the central control station of the estimated interference relation information;
Channel setting means for changing a channel used for wireless communication with the terminal device to a channel notified from the central control station,
The central control station is
For each base station apparatus, a throughput index of two or more channel candidates is calculated from channel candidates including a plurality of channels having different frequency bands, and the channel of the base station apparatus is calculated based on the calculated throughput index. Channel determining means for determining;
Channel notification means for notifying the determined channel to the base station apparatus;
Wherein based on the information notified from the base station apparatus, using the channel within the frequency band of a particular channel candidates, provided with counting means for determining the number of groups of base station apparatus that operate independently of,
The channel determining means uses a number of base stations that operate independently in the counting means the inner channel candidates determined in calculates the throughput indication of the channel candidates, radio communications system. A wireless communication system comprising a centralized control station and a plurality of base station devices connected to the centralized control station, and performing wireless communication in an environment where an interference signal is received in the base station device,
The base station device
Interference estimating means for estimating interference relation information which is information relating to the interference signal;
Interference relation information notifying means for notifying the central control station of the estimated interference relation information;
Channel setting means for changing a channel used for wireless communication with the terminal device to a channel notified from the central control station,
The central control station is
For each base station apparatus, a throughput index of two or more channel candidates is calculated from channel candidates including a plurality of channels having different frequency bands, and the channel of the base station apparatus is calculated based on the calculated throughput index. Channel determining means for determining;
Channel notification means for notifying the determined channel to the base station apparatus,
The centralized control station, based on the information notified from the base station apparatus, in a frequency band of a channel connected to have no base station apparatus is used, the interference with a base station apparatus which is not the connection There, and allocates a channel of the plurality of the base station apparatus is not a detectable relationship signal to each other, radio communications system. A wireless communication system comprising a centralized control station and a plurality of base station devices connected to the centralized control station, and performing wireless communication in an environment where an interference signal is received in the base station device,
The base station device
Interference estimating means for estimating interference relation information which is information relating to the interference signal;
Interference relation information notifying means for notifying the central control station of the estimated interference relation information;
Channel setting means for changing a channel used for wireless communication with the terminal device to a channel notified from the central control station,
The central control station is
For each base station apparatus, a throughput index of two or more channel candidates is calculated from channel candidates including a plurality of channels having different frequency bands, and the channel of the base station apparatus is calculated based on the calculated throughput index. Channel determining means for determining;
Channel notification means for notifying the determined channel to the base station apparatus,
The channel determining means, said determining the order in which to evaluate the channels used by the base station apparatus determines the high the channel throughput metrics for each base station apparatus determined order, radio communications system. The channel determination means determines a plurality of patterns for the order, determines the channel for each base station device in the order determined for each pattern, and based on the combination of channels determined for each pattern and the throughput index , Select a combination of channels in any one pattern,
The wireless communication system according to claim 7 .

Images