JP2016015570A - Communication device, computer program, and channel evaluation method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To estimate the interference level of a channel.SOLUTION: A communication device 1 comprises a receiver 10 for receiving a signal and a processing unit 30. The processing unit 30 is configured to perform: monitoring processing 32 for obtaining an observation value of an index of interference per interference source group on the basis of a signal received by the receiver 10; and estimation processing 34 of the interference level on an estimation object channel. The estimation processing 34 includes: processing for selecting one or multiple interference source groups; and processing for calculating an estimation value of the interference level on the estimation object channel from the observation value for the selected one or multiple interference source groups.

Description

  The present invention relates to a communication device, a computer program, and a channel evaluation method.

As the number of wireless LAN communication devices increases, there is a problem in that communication quality is likely to deteriorate due to interference from other communication devices. As a countermeasure against interference, it is conceivable to select a channel with less interference.
Non-Patent Document 1 discloses a technique for allocating channels so that the ratio of the transmittable / receiveable data amount to the accommodated data amount of each access point is increased so that the throughput of the access point whose throughput is greatly deteriorated due to inter-cell interference is improved. It is disclosed.

B. A. Hirantha Sithira Abeysekera, 4 others, “Network-Linked Channel Allocation Method in IEEE 802.11 Wireless LAN”, IEICE Technical Report, IEICE, 2013-04

  In the technique disclosed in Non-Patent Document 1, it is necessary to periodically measure the throughput at other access points. In order to measure the throughput at other access points, the access points need to communicate for throughput measurement.

  An object of the present invention is to provide a new technique for easily estimating the interference level experienced by a channel.

A communication apparatus from a certain point of view includes a receiver that receives a signal and a processing unit, and the processing unit transmits an observation value of an index related to interference based on the signal received by the receiver to an interference source group. A monitoring process to be obtained every time, an estimation process of the interference level in the estimation target channel,
The estimation process includes: selecting one or a plurality of the interference source groups; and the interference level in the estimation target channel from the observation value for each of the selected one or a plurality of the interference source groups. And calculating an estimated value.

  A computer program from a certain point of view causes a computer in a communication apparatus including a receiver that receives a signal and a computer to execute processing. The processing includes a monitoring process in which the computer obtains an observation value of an index related to interference for each interference source group based on the signal received by the receiver, and the computer estimates an interference level in an estimation target channel. An estimation process, wherein the computer selects one or more of the interference source groups, and the computer uses the observation value for each of the selected one or more of the interference source groups, Calculating an estimated value of the interference level in the estimation target channel.

  According to a channel evaluation method from a certain viewpoint, an observation value of an index related to interference is obtained for each interference source group based on a received signal, one or more interference source groups are selected, and one or more selected interference source groups are selected. Calculating an estimated value of the interference level in the estimation target channel from the observed value for each of the interference source groups.

  According to the present invention, it is possible to easily estimate the interference level received by a channel.

It is a figure which shows a communication apparatus and an interference source group. It is a block diagram of a communication apparatus. It is a flowchart which shows a frame monitoring process. It is a figure which shows an observation value database. It is a figure which shows the channel of IEEE802.11b / g. It is a figure which shows the simulation conditions in the case of one interference source group. (A) shows the relationship between the frame arrival rate and the score, and (b) is a diagram showing the relationship between the transmission delay and the score. It is a figure which shows the simulation conditions in the case of two interference source groups. (A) shows the relationship between the frame arrival rate and the score, and (b) is a diagram showing the relationship between the transmission delay and the score. It is a flowchart which shows an evaluation process. It is explanatory drawing of a calculation object channel. It is a list of prediction functions. It is a figure which shows a conversion table. It is explanatory drawing of the communication quality evaluation by a score. It is explanatory drawing of the conversion table which has several conversion characteristics. It is a flowchart which shows an interference countermeasure process. It is explanatory drawing of the time series data regarding a channel occupation rate.

[1. Outline of Embodiment]
(1) A communication apparatus according to an embodiment includes a receiver that receives a signal and a processing unit, and the processing unit obtains an observation value of an index related to interference based on the signal received by the receiver. It is configured to perform a monitoring process obtained for each interference source group and an interference level estimation process in the estimation target channel, and the estimation process includes a process for selecting one or a plurality of the interference source groups and a selected one. Or processing for calculating an estimated value of the interference level in the estimation target channel from the observed values for each of the plurality of interference source groups.
In the estimation process, an estimated value of the interference level in the estimation target channel can be calculated from the observed values for one or a plurality of the interference source groups. If obtained, the estimated value of the interference level in the estimation target channel can be easily calculated.

(2) Each of the one or more interference source groups selected by the process of selecting one or more interference source groups is an interference source group that uses one of a plurality of calculation target channels, The calculation target channel preferably includes one or a plurality of channels in the vicinity of the estimation target channel.
Since interference may also be received from a channel near the estimation target channel, interference from a channel near the estimation target channel can be considered by selecting an interference source group that uses the channel near the estimation channel. This is particularly preferable when the frequency regions of adjacent channels overlap.

(3) It is preferable that the index includes an inter-channel distance between the estimation target channel and a channel used by each of the selected one or a plurality of the interference source groups. In this case, for each interference source group, how far away the channel used by the interference source group is from the estimation target channel can be reflected in the estimated value of the interference level.

(4) It is preferable that the index includes an intensity of an interference signal from the interference source group. In this case, the intensity of the interference signal for each interference source group can be reflected in the estimated value of the interference level.

(5) It is preferable that the index includes a channel occupation ratio of the interference source group. In this case, the channel occupation rate for each interference source group can be reflected in the estimated value of the interference level.

(6) The processing unit further performs interference countermeasure processing based on a result of the estimation processing, and the estimated value of the interference level in the estimation target channel is calculated for each of the plurality of estimation target channels, and the interference The countermeasure process preferably includes a process of selecting a communication channel of the communication device from the plurality of estimation target channels based on the estimated value of the interference level calculated for each of the plurality of estimation target channels. .
In this case, it is possible to select an appropriate communication channel based on the estimated value of the interference level of each of the plurality of estimation target channels.

(7) A storage unit is further provided, and the storage unit stores a conversion table for converting the estimated value of the interference level into a communication quality value, and the interference level calculated for each of the plurality of estimation target channels. The estimated value is converted into a communication quality value using the conversion table, and the interference countermeasure process determines a communication channel of the communication apparatus from a plurality of the estimation target channels based on the communication quality value. It is preferable that the process to select is included.
The communication quality value of the estimation target channel can be obtained from the conversion table. Furthermore, a communication channel can be appropriately selected according to the obtained communication quality value.

(8) The storage unit further stores correction information, and the conversion table is configured to convert the estimated value of the interference level into a communication quality value when the number of the interference source groups is a predetermined number. The correction information is for correcting the communication quality value converted by the conversion table to a value corresponding to the number of the interference source groups, and is selected by the process of selecting the interference source groups. In addition, when the number of the interference source groups is other than the predetermined number, the communication quality value converted by the conversion table is preferably corrected based on the correction information. In this case, the size of the conversion table can be suppressed.

(9) The interference countermeasure process is a process of selecting an interference countermeasure to be performed based on the estimated value of the interference level in the estimation target channel and a temporal change of the estimated value calculated in the past. Is preferably included.
In this case, an appropriate interference countermeasure can be selected in consideration of the temporal change at the time of estimation calculated in the past.

(10) The calculation of the estimated value of the interference level in the estimation target channel is preferably performed by further considering the type of the interference source group. In this case, a more appropriate estimated value can be calculated in consideration of the type of the interference source group.

(11) It is preferable that the type of the interference source group includes a type of an access point included in the interference source group. In this case, a more appropriate estimated value can be calculated in consideration of the type of access point included in the interference source group.

(12) It is preferable that the type of the interference source group includes a type according to whether or not the interference source group is a preset interference source group. In this case, a more appropriate estimated value can be calculated in consideration of whether or not the interference source group is a preset interference source group.

(13) It is preferable that the number of the interference source groups selected in the process of selecting one or a plurality of the interference source groups is equal to or less than a preset upper limit value. In this case, it is possible to suppress the number of selected interference source groups to be equal to or lower than the upper limit value of the number of interference source groups for which an estimated value can be calculated in the estimation process.

(14) It is preferable that the selection of the interference source group in the process of selecting one or a plurality of the interference source groups is performed based on an intensity of an interference signal from the interference source group. In this case, the interference source group can be appropriately selected according to the intensity of the interference signal.

(15) The selection of the interference source group in the process of selecting one or a plurality of the interference source groups is preferably performed based on a channel occupancy rate of the interference source group. In this case, the interference source group can be appropriately selected according to the channel occupation ratio.

(16) A computer program according to an embodiment causes a computer in a communication apparatus including a receiver that receives a signal and a computer to execute processing. The processing includes a monitoring process in which the computer obtains an observation value of an index related to interference for each interference source group based on the signal received by the receiver, and the computer estimates an interference level in an estimation target channel. An estimation process, wherein the computer selects one or more of the interference source groups, and the computer uses the observation value for each of the selected one or more of the interference source groups, Calculating an estimated value of the interference level in the estimation target channel. The computer program may be stored on a recording medium.

(17) The method according to the embodiment obtains an observation value of an index related to interference for each interference source group based on the received signal, selects one or a plurality of the interference source groups, Calculating an estimated value of the interference level in the estimation target channel from the observed value for each of the plurality of interference source groups.

[2. Details of Embodiment]
Hereinafter, details of the embodiment will be described with reference to the drawings.

[2.1 Overview of communication devices]
The communication apparatus 1 shown in FIG. 1 is for ITS (Intelligent Transport Systems). The communication device (base station) 1 is installed near an intersection or the like, and is used for driving support such as traffic signal information, vehicle / pedestrian approach information, etc. with respect to a mobile station (on-vehicle device mounted on a vehicle) 100. Provides information (ITS service).
The communication device 1 can also communicate with other ITS communication devices (base stations). For example, an ITS communication device (base station) 1 connected to a camera installed on the shoulder of the road transmits vehicle sensing information acquired from the camera to another IT communication device (base station), thereby enabling the communication device Information can be shared between (base stations).

The communication apparatus 1 according to the embodiment uses a 2.4 GHz band, which is an ISM (Industry Science Medical) band, for communication, and performs communication based on the IEEE 802.11 standard, which is a wireless LAN standard. The cost can be reduced by performing the ITS service by the widely used wireless LAN technology.
However, when a wireless LAN is used for ITS communication, communication interference from other wireless LAN base stations 101a, 102a unrelated to ITS and terminals 101b, 102b, 102c connected to these base stations 101a, 102a is a problem. Become. In particular, in urban areas, there are a large number of wireless LAN base stations and public wireless LAN base stations installed in offices and homes, so communication quality is likely to deteriorate due to interference.
As the wireless LAN base stations 101a and 102a, there are mobile base stations (such as mobile routers) in addition to fixed base stations installed in offices. The influence on interference differs depending on whether the base station is a fixed station or a mobile station.

  Moreover, since the ISM band is also used for a device (non-wireless LAN device) 103a other than the wireless LAN, interference from the non-wireless LAN device 103a also becomes a problem.

  Further, in IEEE802.11b / g, the 2.4 GHz band channels are a total of 13 channels. However, since different channels are overlapped at physical frequencies, even different channels may interfere with each other.

As described above, in the ISM band, there are various interference sources and various ways of affecting the interference.
Therefore, the communication device 1 according to the embodiment has a function for switching to a frequency channel with a low interference level when there is radio wave interference. Note that the communication device 1 of the present embodiment is an ITS communication device using the IMS band, but is not limited to such an application.

  FIG. 2 shows the configuration of the communication apparatus (base station) 1. The communication device 1 includes a communication device (receiver / transmitter) 10 and a computer 20. The communication device 10 performs wireless communication with another communication device (base station) or a mobile station (on-vehicle device). The computer 20 includes a processing unit 30 and a storage unit 40. The processing unit 30 executes a computer program (not shown) stored in the storage unit 40. Necessary information processing is performed by the computer program being executed by the processing unit 30. This information processing includes a frame monitoring process 32, an evaluation process (estimation process) 34, and an interference countermeasure process (channel selection process) 36.

When the communication device 10 receives signals (for example, communication frames) from the other ISM band use devices 101a, 101b, 102a, 102b, 102c, and 103a, the processing unit 30 performs frame monitoring based on the received signals. Processing (monitoring processing) 32 is performed to obtain an observed value of an index related to interference. The monitoring process 32 is executed periodically or as necessary.
The processing unit 30 performs an interference level evaluation process (estimation process) 34 based on the observed value, and performs an interference countermeasure process 36 such as a channel selection process for switching to a channel with less interference based on the result of the evaluation process.
Details of the processes 32, 34, and 36 will be described later.

The storage unit 40 stores an observation value database 42, a conversion table 44, correction information 46, time series data 48, and the like. The observation value database 42 is a database that records observation values obtained by the monitoring process 32. The conversion table 44 is for converting the interference level estimation value (score) obtained by the evaluation process (estimation process) 34 into a communication quality value (packet arrival rate, transmission delay, etc.). The correction information 46 is for correcting the score obtained by the conversion table 44. The time-series data 48 is data indicating changes over time such as past interference level estimation values (scores). The time series data 48 may include data indicating time series changes such as observed values and communication quality.
Details of the stored information 42, 44, 46, and 48 will be described later.

[2.2 Frame monitoring processing]
Using CSMA (Carrier Sense Multiple Access / Collision Detection) adopted in IEEE 802.11, it is necessary to suppress the frame delay and loss to a certain level and maintain an appropriate throughput in order to provide the ITS service described above. There is. Here, in order to suppress the frame delay, it is required to reduce as much as possible the effects of frame error and accompanying frame retransmission, and communication opportunity loss due to spatial contention.

The communication device 1 according to the embodiment evaluates how much the interference source affects the device 1 and its client terminal using a plurality of indices. In the frame monitoring process 32, observed values of a plurality of indices p ₁ , p ₂ , and p ₃ used for interference evaluation are obtained. Frame monitoring for obtaining observation values is performed for all channels (channel 1 to channel 13) that can be used by the communication device 1.

For example, a) interference signal strength p ₁ , b) channel occupancy p ₂ , and c) interchannel distance p ₃ are adopted as the plurality of indices.
a) The interference signal strength p ₁ is the interference signal strength from the interference source groups 101, 102, and 103.
b) The channel occupancy p ₂ is the channel occupancy of the interference source groups 101, 102, and 103.
c) The inter-channel distance p ₃ is a distance (absolute value of the distance) between the used channel of the interference source groups 101, 102, and 103 and the evaluation (estimation) target channel.

Here, each of the interference source groups 101, 102, and 103 includes at least one signal generation device (communication device and other devices that generate radio waves) 101a, 102a, and 103a, as shown in FIG. When there are other apparatuses 101b, 102b, and 102c that are wirelessly connected to the generation apparatuses 101a, 102a, and 103a, the other apparatuses 101b, 102b, and 102c are also included.
Typically, one interference source group 101 includes a wireless LAN access point 101a and one or more client terminals 101b connected to the access point. In addition, a single device (such as a medical device that emits radio waves) 103a may constitute one interference source group.

In the monitoring process 32, a plurality of indices p ₁ , p ₂ , and p ₃ are observed for each of the interference source groups 101, 102, and 103 and recorded in the observation value database 42 for each of the interference source groups 101, 102, and 103 ( (See FIG. 4).

FIG. 3 shows the procedure of the frame monitoring process 32. First, the interference source groups 101, 102, and 103 (interfering access points 101a and 102a) are recognized (step S11).
Interference source group recognition is performed using the SSID or MAC address included in the Beacon frame transmitted from the wireless LAN base stations 101a and 102a of the interference source groups 101 and 102. The Beacon frame is periodically transmitted from the wireless LAN base stations (access points) 101a and 102a. The SSID or MAC address is used as an identifier of the interference source groups 101 and 102.

The Beacon frame is a management frame used for notifying various client terminals of various information necessary for communication, and the source MAC address and SSID are stored in the address field of the Beacon frame.
In step S11, the SSID or MAC address included in the received Beacon frame is acquired and recorded in the observation value database 42 (see FIG. 4).

  The wireless LAN base station (access point) may include a Passive access point that does not transmit a Beacon frame. For such Passive access points, the SSID or MAC address included in the Probe Request transmitted by the client terminal or the Probe Response transmitted by the access point is acquired as the identifier of the interference source group.

  For the interference source group 103 including the devices 103a that are not performing wireless communication but are emitting radio waves, the processing unit 30 may assign an appropriate identifier (for example, “xxx”, see FIG. 4).

Subsequently, when receiving a frame not addressed to the own apparatus 1 (step S12), the processing unit 30 confirms the SSID, the source MAC address, or the destination MAC address in the received frame, and classifies the received frame for each interference source group. (Step S13).
The processing unit 30 obtains the average interference signal strength p ₁ , the channel occupation ratio p ₂ , and the observation value of the used channel for each interference source group (step S14). Incidentally, the channel is used for calculating the distance between channels p _3.
The used channel is indicated by a channel number. As shown in FIG. 5, the 2.4 GHz band in IEEE802.11b / g is from channel 1 to channel 13, and the channel is specified by numbers 1 to 13.
Method of calculating the distance between channels p ₃ will be described later.

The average interference signal strength p ₁ for each interference source group is obtained by calculating the average of the received signal strength (interference signal strength) RSSI of each of the plurality of received frames classified for each interference source group.
The channel occupancy p ₂ for each interference source group measures the number of a plurality of received frames classified for each interference source group existing within a predetermined observation time, or the time length that existed within a predetermined observation time. Can be obtained.
The channel used for each interference source group can be obtained from the channel number (Channel Number) included in the received frame or can be obtained by measuring the frequency of the received frame (signal).

The obtained observation values (average interference signal intensity p ₁ , channel occupation rate p ₂ , used channel) are recorded in the observation value database 42 for each interference source group (step S15).

FIG. 4 shows an observation value database 42 in which observation values are recorded. In the observation value database 42, interference source group identifiers (such as SSID) are recorded in the interference source identifier 42a column. In the interference source identifier 42a column of FIG. 4, “AP-a” that is the SSID of the wireless LAN access point 101a, “AP-b” that is the SSID of the wireless LAN access point 102a, and the non-wireless LAN device 103a. The identifier “xxx” given by the processing unit 30 is recorded as the identifier of each of the interference source groups 101, 102, and 103.
In the average interference signal strength p ₁ column of FIG. 4, the average RSSI of each of the plurality of interference source groups 101, 102, and 103 is recorded as the average signal strength of the interference source group.
The channel occupancy _{p 2} column of FIG. 4, a plurality of interferers groups 101, 102, 103 each channel occupancy is recorded.
In the use channel of FIG. 4, the channels used by the plurality of interference source groups 101, 102, and 103 for communication are recorded.

Incidentally, when the device constituting the communication apparatus 1 and the interference source groups 101 and 102 is in compliance with the IEEE 802.11k, the average interference signal strength _{p 1,} transmitted from a base station of the interference source group (interfering access point) An average received channel power indicator (average RCPI) included in the generated Frame Report may be used. When the interference source group 103 is configured by the non-wireless LAN device 103a, an IEEE 802.11k Noise Histogram Report ANPI can be used.
The channel occupancy p ₂ can use the Frame Count included in the IEEE 802.11k Frame Report, and the channel used can use the Channel Number included in the IEEE 802.11k Beacon Report.

[2.3 Evaluation process (estimation process)]
[2.3.1 Interference level prediction function (interference level prediction model)]
In the interference level evaluation process 34, a score, which is an estimated value of the interference level, is obtained from the observed values p ₁ , p ₂ , and p _{3 for} each interference source group using an interference level prediction function.
Here, first, an interference level prediction function (interference level prediction model) will be described.

The interference level prediction function (interference level prediction model) is generated in advance by simulation.
As shown in FIG. 6, the simulation shows a set of an access point (interfered AP; target AP) that receives interference and its client terminal (target client; camera), and an interfering access point (interfering fixed AP) that gives interference. And a set of the client terminals (interfering fixed clients) are assumed to be installed.
For the interference source group composed of the fixed interference AP and the fixed interference client, the simulation was performed by comprehensively changing the combinations of the values of the indices p ₁ , p ₂ , and p ₃ .
A function for indicating the interference level received by the interfered AP as a relative value by a single interference source group (interfering fixed AP and interfering fixed client) by performing multiple regression analysis on the result obtained by the simulation. Can be obtained.

In FIG. 6, the target client is a client terminal connected to a vehicle monitoring camera that periodically transmits vehicle sensing information detected from an image to a target AP that is an intersection roadside machine. The distance between the interfered AP (target AP) and the target client was fixed at 100 m. The interfering fixed AP and the interfering fixed client are assumed to be a general office or hotspot access point and the client, and the distance between them is 5 m.
The client (camera) connected to the interfered AP transmits vehicle sensing information (1 kByte packet every 1.75 ms) to the interfered AP (target AP) at regular time intervals using UDP.

In the simulation, the interference signal strength p ₁ was changed by changing the distance between the interfered AP (target AP) and the interfering fixed AP from 10 m to 300 m at 10 m intervals.
The interfering fixed client shall transfer a plurality of files to the interfering fixed AP using TCP, and the file reading interval when the plural files are transferred (from the end of the file transmission to the start of the next file transmission). by varying the time), it was changed channel occupancy p _2. Note that the transfer file size was 2 Mbytes on average.

The communication channel (used channel) used by the interfered AP (target AP) was fixed to “6”, and the interfering fixed AP was allowed to use a total of 4 channels from 6 to 9. Thus, it is possible to change the channel distance p ₃ between the interference fixed AP uses channel given to the use channel of the interfered AP (target AP) to 0-3.
If the interchannel distance to be changed is 4 or more, as is apparent from FIG. 5, the overlap between the channels is reduced or completely eliminated. Therefore, the case where the interchannel distance is 4 or more is not considered. However, the simulation may be performed including the case where the inter-channel distance is 4 or more.

As an index of communication quality (communication performance), the arrival rate and frame transmission delay time of a data frame transmitted by UDP from the camera client of the interfered AP (target AP) are used.
120 seconds of simulation was performed for all 3600 combinations of p ₁ of 30, p ₂ of 30, and p ₃ of 4, and the arrival of the frame as a communication quality index under the conditions of each combination The rate and transmission delay time were obtained.

By performing multiple regression analysis using p ₁ , p ₂ , and p ₃ used in the simulation as explanatory variables and the frame arrival rate or transmission delay time obtained by the simulation as an objective variable, an interference level prediction function (interference level prediction) Model). Note that the explanatory variable is a normalized value because the units are different from each other. The explanatory variables, interactions _{p 3} from _{p 1} was also added.

In the present embodiment, the interference level prediction function includes a first interference level prediction function f ₁ ¹ for obtaining a score (interference level estimation value) S ₁ related to a frame arrival rate, and a score (interference level estimation value) related to transmission delay time. ) A second interference level prediction function f ₂ ¹ for obtaining S ₂ was obtained. In any of the functions f ₁ ¹ and f ₂ ¹ , the scores S ₁ and S ₂ can be obtained from p ₁ , p ₂ and p ₃ (see FIG. 12).
Therefore, if there is one interference source group, when a set of observation values p ₁ , p ₂ , and p ₃ for the interference source group is obtained, the score S ₁ regarding the frame arrival rate and the score regarding the transmission delay time A plurality of interference level estimates S ₂ can be obtained.

FIG. 7A shows a case where the number of interference source groups is one, using the first interference level prediction function f ₁ ^{1 based} on the frame arrival rate obtained by the simulation and the p ₁ , p ₂ , and p ₃ used in the simulation. the resulting score (score for frame arrival rate; interference level estimate) shows the S _1.
7 (b) is the transmission delay was determined by simulation and (Delay), _p 1 used in the _simulation, p 2, _{p 3} from the second interference level prediction function _f ^{2 1} Score obtained using (transmission delay The time score; interference level estimate) S ₂ is shown.
7A and 7B, the horizontal axis indicates the number of combinations of p ₁ , p ₂ , and p ₃ . 7A and 7B, the frame arrival rate and transmission delay (Delay) obtained by simulation are indicated by ◯, and the score obtained by the interference level prediction function is indicated by ×. In FIG. 7, the circles are densely arranged, so that they appear as one thick black line as a whole.

The determination coefficient (correlation) between the frame arrival rate and the score S ₁ shown in FIG. 7A is 0.79, and it is recognized that the frame arrival rate and the score S ₁ have a strong positive correlation. The coefficient of determination of the transmission delay (Delay) and the score _{S 2} shown in FIG. 7 (b) is 0.77, is recognized that there is a strong positive correlation to transmission delay and scores _{S 2.}
Therefore, the scores S ₁ and S ₂ obtained by using the interference level prediction functions f ₁ ¹ and f ₂ ¹ from the actually observed p ₁ , p ₂ , and p ₃ are appropriate as relative estimates of the interference level. It can be seen that it is. The interference level prediction functions f ₁ ¹ and f ₂ ¹ are obtained when the communication channel (used channel) used by the interfered AP (target AP) is a specific channel (CH6). It can also be used to predict the interference level in the channel. This is because any of the channels (CH1 to CH13) is a close channel within a common band of 2.4 GHz band, and thus has similar interference characteristics.

In the simulation condition shown in FIG. 6, there is one interference source group (interfering AP and interfering client). However, even when there are a plurality of interference source groups, a similar simulation is performed to predict the interference level. You can get a function.
FIG. 8 shows simulation conditions when there are two interference source groups. For each interference source group, p ₁ , p ₂ , and p ₃ are changed, and the frame arrival rate and transmission delay time in the interfered AP (target AP) are obtained by simulation.

Here, a set of indices (p ₁ , p ₂ , p ₃ ) related to one interference source group (first interference source group) in FIG. 6 is represented as (p ₁ ¹ , p ₂ ¹ , p ₃ ¹ ), A set of indices (p ₁ , p ₂ , p ₃ ) relating to the other interference group (second interference source group) is represented as (p ₁ ² , p ₂ ² , p ₃ ² ).
P ₁ ¹ , p ₂ ¹ , p ₃ ¹ , p ₁ ² , p ₂ ² , and p ₃ ^{2 used} for the simulation are explanatory variables, and the frame arrival rate or transmission delay time obtained by the simulation is the target variable. By performing the regression analysis, an interference level prediction function (interference level prediction model) is obtained as in the case of a single interference source group.

Even when there are two interference sources, the interference level prediction function includes a first interference level prediction function f ₁ ² for obtaining a score (interference level estimation value) S ₁ regarding the frame arrival rate, and a score regarding the transmission delay time. (Interference level estimation value) A second interference level prediction function f ₂ ² for obtaining S ₂ was obtained. In any of the functions f ₁ ² and f ₂ ² , scores S ₁ and S ₂ can be obtained from p ₁ ¹ , p ₂ ¹ , p ₃ ¹ , p ₁ ² , p ₂ ² , and p ₃ ² (FIG. 12). reference).

Therefore, if there are two interference source groups, the observed values (p ₁ ¹ , p ₂ ¹ , p ₃ ¹ ) and (p ₁ ² , p ₂ ² , p ₃ ² ) are obtained for each of the interference source groups. Then, a plurality of interference level estimation values such as a score S ₁ related to the frame arrival rate and a score S ₂ related to the transmission delay time can be obtained.

FIG. 9A shows the frame arrival rate obtained by the simulation and the p ₁ ¹ , p ₂ ¹ , p ₃ ¹ , p ₁ ² , p ₂ ² , p used in the simulation when there are two interference source groups. ₃ scores obtained by using the ² first interference level prediction function f ₁ ² (score for frame arrival rate; interference level estimate) shows the S _1.
FIG. 9B shows the transmission delay (Delay) obtained by the simulation and the second interference level from the p ₁ ¹ , p ₂ ¹ , p ₃ ¹ , p ₁ ² , p ₂ ² , and p ₃ ² used in the simulation. It shows the S _2; (interference level estimate score for transmission delay time) prediction function f ₂ ² scores obtained using.
9A and 9B, the horizontal axis indicates the number of combinations of (p ₁ ¹ , p ₂ ¹ , p ₃ ¹ , p ₁ ² , p ₂ ² , p ₃ ² ). 9A and 9B, the frame arrival rate and transmission delay (Delay) obtained by simulation are indicated by ◯, and the score obtained by the interference level prediction function is indicated by ×. In FIG. 9 as well, the circles are densely arranged, so that they appear as one thick black line as a whole.

The determination coefficient (correlation) between the frame arrival rate and the score S ₁ shown in FIG. 9A is 0.80, and it is recognized that the frame arrival rate and the score S ₁ have a strong positive correlation. The coefficient of determination of the transmission delay (Delay) and the score _{S 2} shown in FIG. 9 (b) is 0.86, is recognized that there is a strong positive correlation to transmission delay and scores _{S 2.}
Therefore, the score S ₁ obtained from the actually observed p ₁ ¹ , p ₂ ¹ , p ₃ ¹ , p ₁ ² , p ₂ ² , and p ₃ ^{2 using} the interference level prediction functions f ₁ ² and f ₂ ^2. , S ₂ is also found to be suitable as a relative estimate of the interference level.

Similarly, when the number of interference source groups is 3, simulations can be performed to obtain the interference level prediction functions f ₁ ³ and f ₂ ³ (see FIG. 12). In this way, the interference level prediction functions f ₁ ⁿ and f ₂ ⁿ can be obtained for each number n of interference groups (n is 1 to N is a positive integer).
In the communication device 1, interference level prediction functions f ₁ ⁿ and f ₂ ⁿ are set for each n of the interference source groups (see FIG. 12), and the processing unit 30 performs the evaluation of the interference source groups in the evaluation process 34. Depending on the number n, the interference level prediction functions f ₁ ⁿ and f ₂ ⁿ are selectively used.
Since only a finite number of interference level prediction functions f ₁ ⁿ and f ₂ ⁿ can be set, the value of n has a predetermined upper limit value N. That is, the estimated interference level can be obtained by the interference level prediction function when the number of interference source groups is up to N.

[2.3.2 Interference level evaluation using interference level prediction function]
FIG. 10 shows the procedure of the interference level evaluation process 34 performed by the processing unit 30 using the interference level prediction function.
The evaluation process 34 shown in FIG. 10 is performed for each of a plurality of evaluation target channels, and an interference level estimation value is obtained for each of the plurality of evaluation channels.
The evaluation target channels are, for example, all channels (channel 1 to channel 13) that can be used by the communication device 1. However, only a part of the channels (channel 1 to channel 13) may be the evaluation target channel. For example, only a channel sufficiently separated from the communication channel currently used by the communication device 1 can be set as the evaluation target channel.
Note that only one specific channel may be evaluated by the procedure of FIG.

In the evaluation process 34 shown in FIG. 10, first, a calculation target channel is determined, and an interference source group that uses the calculation target channel is selected (step S21).
The calculation target channel is a channel that may interfere with the evaluation target channel when communication is performed by the interference source group. The calculation target channel is one or a plurality of channels in the vicinity of the evaluation target channel (for example, the distance between the channels from the evaluation target channel is within a range of ± 3 with reference to the evaluation target channel (the channel for which an estimated interference level is to be obtained). Included channels). As shown in FIG. 11, when the evaluation target channel is channel 6 (CH6), seven channels from channel 3 (CH3) to channel 9 (CH9) are selected as calculation target channels.

When determining the calculation target channel based on the evaluation target channel, the processing unit 30 selects an interference source group that uses the calculation target channel with reference to the observation value database 42.
For example, when the calculation target channel is from channel 3 (CH3) to channel 9 (CH9) as shown in FIG. 11, the processing unit 30 refers to the use channel 42d of the observation value database 42 shown in FIG. Since at least three interference source groups having the interference source identifier (SSID) 42a of “AP-a”, “AP-b”, and “xxx” use the calculation target channel, at least these three An interference source group is selected.

  When selecting an interference source group in step S21, not all of the interference source groups using the calculation target channel are selected, but a predetermined condition among the interference source groups using the calculation target channel is selected. It is possible to select only the interference source group that matches the above or to preferentially select the interference source group that matches a predetermined condition.

For example, since the interference is weak interference source group and outside consideration, it is possible to average interference signal strength p ₁ in FIG. 4 is a selection target interference source group above the first threshold. Further, since the low interference source groups of channel occupancy and outside consideration, it is possible to channel occupancy p ₂ in FIG. 4 as a selection target interference source group of more than a second threshold value.
In addition, an interference source group in which the average interference signal strength p ₁ in FIG. 4 is equal to or higher than the first threshold and the channel occupation ratio p _{2 in} FIG. 4 is equal to or higher than the second threshold can be selected.

  The number of interference source groups selected in step S21 is limited to N or less, which is the upper limit value of n described above. Even if more than N interference source groups are selected, the interference level cannot be estimated because there are no interference level prediction functions corresponding to N or more interference source groups. By selecting the number of interference source groups equal to or less than the preset upper limit value N, it is possible to reliably calculate the estimated value of the interference source level.

When the number of interference source groups exceeds N, for example, the influence on interference is less based on at least one of the average interference signal strength p _{1 in} FIG. 4 and the channel occupation ratio p _{2 in} FIG. By removing the interference source groups (interference source groups having smaller p ₁ and p ₂ ) from the selection target, the number of selected interference source groups can be limited to N or less.
The conditions for limiting the number of interference source groups may be other appropriate conditions.

When one or a plurality of interference source groups are selected from the observation value database 42 in step S21, the processing unit 30 determines the observation value p of each selected interference source group based on the observation value database 42. ₁ , p ₂ and p ₃ are obtained. As the observed values p ₁ and p ₂ , the values of the average interference signal intensity 42b and the channel occupation rate 42c of the observed value database 42 can be used as they are. Observations p ₃ is the distance between the channels is evaluated channel (e.g., CH6) and the communication channel interferers group indicated by the used channel 42 (e.g., CH3) and the absolute value of the inter-channel distance (e.g., channel It is calculated | required as distance between = 6-3 = 3).

Further, the processing unit 30 selects an interference level prediction function corresponding to the number n of interference source groups selected in step S21 (step S22). FIG. 12 shows a list of interference level prediction functions set in the storage unit 40 of the computer 20. As shown in FIG. 12, for each number n of interference source groups, a first interference level prediction function f ₁ ⁿ for determining a score S ₁ related to the frame arrival rate and a _second for determining a score S ₂ related to transmission delay. An interference level prediction function f ₂ ⁿ is set.
For example, when the number of interference source groups selected in step S21 is 3 (n = 3), the processing unit 30 selects the interference level prediction functions f ₁ ³ and f ₂ ³ corresponding to the number of groups = 3.

Subsequently, the processing unit 30 estimates the interference level using the selected interference level prediction functions f ₁ ³ and f ₂ ³ from the observed values p ₁ , p ₂ , and p _{3 for} each selected interference source group. Values (scores S ₁ and S ₂ ) are calculated (step S23).

More specifically, the observed values of the three selected interference source groups are (p ₁ ¹ , p ₂ ¹ , p ₃ ¹ ), (p ₁ ² , p ₂ ² , p ₃ ² ), (p ₁ ³ , p ₂ ³ , p ₃ ³ ), these observations are variables, and based on the selected interference level prediction function f ₁ ³ , f ₂ ³ , the score S ₁ regarding the frame arrival rate and the transmission score S ₂ is calculated on delay.
Note that different interference level prediction functions may be prepared depending on the evaluation target channel, and different interference level prediction functions may be selected depending on the evaluation target channel. However, this embodiment is simple because a common interference level prediction function is used regardless of the difference between the evaluation target channels (CH1 to CH13).

The processing unit 30 converts each of the calculated scores S ₁ and S ₂ into a communication quality value (frame arrival rate, transmission delay) using the conversion table 44 (step S24).
The storage unit 40 converts the conversion table 44, a first conversion table for converting the score _{S 1} about the frame arrival rate to the frame arrival rate (see FIG. 13 (a)), a score _{S 2} to the transmission delay (Delay) A second conversion table (see FIG. 13B) is stored.

In FIG. 13, a thick curve with the letters “conversion characteristic curve” indicates a conversion characteristic for converting the scores S ₁ and S ₂ into communication quality values. In FIG. 13, the relationship between the scores S ₁ and S ₂ and the communication quality values (frame arrival rate, transmission delay) shown in FIGS. The conversion characteristic curve in FIG. 13 is set in accordance with the characteristic indicated by ◯.

The communication quality value (frame arrival rate, transmission delay) obtained in step S24 becomes the evaluation value of the evaluation target channel.
For example, if the frame arrival rate is 0.8 (80%) or higher, the evaluation target channel can be evaluated as a channel that can ensure good communication quality. If the transmission delay is 4 ms or less, the evaluation target channel can be evaluated as a channel that can ensure good communication quality.

Processing unit 30, based on the interference level estimated values S ₁ is the result of the evaluation process _34, S 2 _{(communication} quality value estimated based on the interference level estimated values S _1, S2), among all the evaluated channel Therefore, the channel with the best communication quality is set as the switching destination channel candidate in the communication device 1.

As described above, the communication device 1 can estimate communication quality when an evaluation target channel is used for communication with an arbitrary channel that is not currently used for communication as an evaluation target.
Further, since a plurality of communication qualities (frame arrival rate, transmission delay) can be estimated, the communication quality can be grasped more accurately, and channel selection (channel switching) according to the content (QoS) of transmitted data becomes possible. .

When the desired communication quality value is determined in advance, the scores S ₁ and S ₂ corresponding to the desired communication quality value can also be determined in advance, so that the scores S ₁ and S ₂ can be directly used without using the conversion table. You may evaluate the communication quality of an evaluation object channel.
For example, in FIG. 13A, in order for the desired frame arrival rate to be 0.8 (80%) or more, the score S ₁ regarding the frame arrival rate may be 0.775 or more. Further, in FIG. 13 (b), the desired transmission delay is 4ms or less, the score _{S 2} regarding transmission delay, may be at 2.25 or less. Thus, for the score S _1, as a threshold of 0.775, communication quality if it exceeds the threshold can be evaluated as good. With respect to the score S _2, as a threshold value to 2.25, the communication quality if falls below the threshold value can be evaluated as good.

As shown in FIG. 14, beyond the 0.775 score S ₁ is a threshold, and, when in the range A below 2.25 score S ₂ is the threshold value, evaluates the communication quality to be good May be. In this case, since it is not necessary to convert the scores S ₁ and S ₂ into communication quality using the conversion table in step 24 of FIG. 10, it is not necessary to hold the conversion table.

The conversion characteristics of the conversion table shown in FIG. 13 may be common regardless of the number of interference source groups considered for calculating the scores S ₁ and S ₂ , as shown in FIG. The conversion characteristics g ₁ , g ₂ , and g ₃ of the conversion table may be varied according to the number n of interference source groups considered for calculating the scores S ₁ and S ₂ . When the conversion characteristics g ₁ , g ₂ , and g ₃ of the conversion table are made different according to the number n of the interference source groups, a number of conversion tables corresponding to the number of the interference source groups are required, and many storage tables 40 are stored in the storage unit 40. Storage capacity is required.

Therefore, in order to save the storage area, only the conversion table 44 of the conversion characteristic g ₁ corresponding to a predetermined number (for example, n = 1) of interference source groups is stored in the storage unit 40 as shown in FIG. Keep it. Regardless of whether the number of interference source groups is a predetermined number (n = 1) or not, the scores S ₁ and S ₂ are converted into communication quality values using the conversion table 44 of the conversion characteristics g _1. Is done. The storage unit 40, converts the communication quality values obtained by using the conversion table 44 characteristic g _1, the interference source modification information 46 for correcting in accordance with the number n of the group is set. The processing unit 30 can correct the communication quality value obtained by the conversion to an appropriate value corresponding to the number of interference source groups by correcting the communication quality value based on the correction information corresponding to the number of interference source groups.

The correction information 46 may correct the scores S ₁ and S ₂ . Further, the correction information 46 may correct the score or the communication quality value for other purposes.
For example, when the interference source is a mobile router, the mobile router moves. Therefore, even if interference occurs in a short time, there is little influence as long-term interference. Therefore, when the identifier of the interference source group is an identifier indicating a mobile base station (mobile access point) such as a mobile router, the scores S ₁ and S ₂ may be modified so that the influence of interference is estimated to be small. In this case, the correction information 46 is based on the type of the base station (access point) such as whether the identifier of the interference source group is a mobile base station (mobile access point) or a fixed base station (fixed access point). The scores S ₁ and S ₂ are set to be corrected.

In addition, the correction information 46 sets the scores S ₁ and S ₂ so that the influence of interference is estimated to be small for the interference source group determined that the influence on the interference is short-term or small from past communication results. It may be corrected.
Further, the correction information 46 may correct the scores S ₁ and S ₂ so that the interference source group that is determined to have a large influence on interference is greatly estimated. As an interference source group having a large influence on interference, an interference source group including a wireless LAN access point installed in an office near the communication apparatus 1 can be cited. Such an interference source uses a specific channel for a long period of time, and the interference to the channel is large. If an identifier of an interference source group having a large influence on interference is set in the storage unit 40 in advance, and the scores S ₁ and S ₂ are corrected so as to greatly estimate the influence of interference for the preset interference source group. Good.

As described above, when calculating the estimated values of the interference level (scores S ₁ and S ₂ ), the interference source group type (mobile station or fixed station type from various viewpoints, and the degree of influence on interference) More appropriate calculation can be performed by considering the type and the type according to whether or not the interference source group is set in advance.

[2.3.3 Interference countermeasure processing]
The processing unit 30 performs an interference countermeasure process 36 based on the evaluation results for all the evaluation target channels by the evaluation process 34. As the interference countermeasure process 36, for example, a channel switching (channel selection) process for switching the communication channel of the communication apparatus 1 to the switching destination channel candidate selected as the channel having the best communication quality from among all the evaluation target channels. Do. The selection of the switching destination channel may be performed based on the score S _1, S _2, may be performed based on the communication quality value obtained based on the score S _1, S _2.

The interference countermeasure process is performed based on the scores S ₁ and S ₂ obtained by the evaluation process 34 and the time-series data 48 in which changes over time of the scores S ₁ and S ₂ estimated in the past are recorded. The interference countermeasure to be selected may be selected, and the selected interference countermeasure may be executed.
Specifically, the score S estimated in the past based on the situation when the scores S ₁ and S ₂ obtained by the evaluation process 34 are obtained (the date and time when the scores S ₁ and S ₂ are obtained). _1, with reference to the sequence data 48 when recording the time course of S _2, interference is determined whether the subsequent long-term (step S31 in FIG. 16).
If the interference is long-term, channel switching is selected as a countermeasure against interference (step S32). If the interference is short-term interference, the channel switching is not performed and the low-rate communication or continuous frame transmission is used (step S32). S33). Step S33 may be a countermeasure against interference by increasing the transmission power. Further, step S33 may be a countermeasure against interference by switching channels with a short term and returning to the original channel when the term is over.

The selection of interference countermeasures using the time-series data 48 is not limited to being performed based on the scores S ₁ and S ₂ , and may be performed based on the observed values p ₁ , p ₂ , and p _3. .
For example, as time series data 48, as shown in FIG. 17, data indicating a change with time of the channel occupation rate for each channel is held. In FIG. 17, channel 6 (CH6) has a higher channel occupancy rate and is more likely to receive interference in the long term than channel 1 (CH1). In this case, channel 1 (CH1) is preferentially selected as the switching destination channel.

[3. Addendum]
The above-described embodiments and modifications should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

1 Communication device 10 Communication device (receiver)
20 Computer (communication processing device)
30 processing unit 32 frame monitoring process 34 evaluation process (estimation process)
36 Interference countermeasure processing 40 Storage unit 42 Observation value database 43 Conversion table 44 Correction information 48 Time series data 100 Mobile station (client terminal)
101 Interference source group 101a Wireless LAN base station 101b Client terminal 102 Interference source group 102a Wireless LAN base station 102b Client terminal 102c Client terminal 103 Interference source group 103a Non-wireless LAN device

Claims (17)

A receiver for receiving the signal;
A processing unit;
With
The processor is
A monitoring process for obtaining an observation value of an index related to interference for each interference source group based on the signal received by the receiver;
An interference level estimation process in the estimation target channel;
Is configured to do
The estimation process includes
Selecting one or more of the interferer groups;
A process of calculating an estimated value of an interference level in the estimation target channel from the observed value for each of the selected one or a plurality of the interference source groups;
Including a communication device. Each of the one or more interference source groups selected by the process of selecting one or more interference source groups is an interference source group that uses any of a plurality of calculation target channels.
The communication apparatus according to claim 1, wherein the plurality of calculation target channels include one or a plurality of channels in the vicinity of the estimation target channel. The communication apparatus according to claim 2, wherein the index includes an inter-channel distance between the estimation target channel and a channel used by each of the selected one or a plurality of the interference source groups. The communication apparatus according to any one of claims 1 to 3, wherein the index includes an intensity of an interference signal from the interference source group. The communication apparatus according to claim 1, wherein the index includes a channel occupation rate of the interference source group. The processing unit further performs interference countermeasure processing based on the result of the estimation processing,
The estimated value of the interference level in the estimation target channel is calculated for each of the plurality of estimation target channels,
The interference countermeasure process includes a process of selecting a communication channel of the communication apparatus from a plurality of the estimation target channels based on the estimated value of the interference level calculated for each of the plurality of estimation target channels. The communication device according to any one of claims 1 to 5. A storage unit;
The storage unit stores a conversion table for converting the estimated value of the interference level into a communication quality value,
The estimated value of the interference level calculated for each of the plurality of estimation target channels is converted into a communication quality value using the conversion table,
The communication apparatus according to claim 6, wherein the interference countermeasure process includes a process of selecting a communication channel of the communication apparatus from a plurality of the estimation target channels based on the communication quality value. The storage unit further stores correction information,
The conversion table is configured to convert the estimated value of the interference level when the number of the interference source groups is a predetermined number into a communication quality value;
The correction information is for correcting the communication quality value converted by the conversion table to a value according to the number of the interference source groups,
The communication quality value converted by the conversion table is corrected based on the correction information when the number of the interference source groups selected by the process of selecting the interference source group is other than the predetermined number. 8. The communication device according to 7. The interference countermeasure process includes a process of selecting an interference countermeasure to be performed based on the estimated value of the interference level in the estimation target channel and a temporal change of the estimated value calculated in the past. The communication device according to any one of claims 6 to 8. The communication apparatus according to claim 1, wherein the calculation of the estimated value of the interference level in the estimation target channel is performed by further considering the type of the interference source group. The communication apparatus according to claim 10, wherein the type of the interference source group includes a type of an access point included in the interference source group. The communication apparatus according to claim 11 or 12, wherein the type of the interference source group includes a type corresponding to whether or not the interference source group is a preset interference source group. The communication apparatus according to claim 1, wherein the number of the interference source groups selected in the process of selecting one or a plurality of the interference source groups is equal to or less than a preset upper limit value. The selection of the interference source group in the process of selecting one or a plurality of the interference source groups is performed based on an intensity of an interference signal from the interference source group. The communication device described. The selection of the interference source group in the process of selecting one or a plurality of the interference source groups is performed based on a channel occupation rate of the interference source group. Communication device. A computer program for causing a computer to execute processing in a communication apparatus including a receiver for receiving a signal and a computer,
The processing is as follows:
A monitoring process in which the computer obtains an observation value of an index related to interference for each interference source group based on the signal received by the receiver;
An estimation process in which the computer estimates an interference level in an estimation target channel;
Including
The estimation process includes
A process in which a computer selects one or more of the interference source groups;
A process in which a computer calculates an estimated value of an interference level in the estimation target channel from the observed value for each of the selected one or a plurality of the interference source groups;
Including computer programs. Based on the received signal, obtaining an observed value of the interference indicator for each interference source groove;
Selecting one or more of the interferer groups;
Calculating an estimated value of the interference level in the estimation target channel from the observed value for each of the selected one or a plurality of the interference source groups;
Including a channel evaluation method.