KR102249993B1 - Apparatus and method for controlling resource reselection in vehicle communication system - Google Patents

Abstract

The present invention provides a device and method for controlling resource reselection in a vehicle communication system, wherein location information and the packet reception rate are received from each of at least one vehicle, a resource maintenance possibility of each vehicle is initialized as to an initial resource maintenance probability in accordance with the number of surrounding vehicles of each vehicle determined from the received location information of each of the at least one vehicle, and thereafter, the resource maintenance probability is updated in response to a change in the packet reception rate of each of the received at least one vehicle to be transmitted to the corresponding vehicle, thereby improving the packet reception rate and the instantaneous data rate.

Description

Resource reselection control device and method of vehicle communication system {APPARATUS AND METHOD FOR CONTROLLING RESOURCE RESELECTION IN VEHICLE COMMUNICATION SYSTEM}

The present invention relates to an apparatus and method for controlling resource reselection in a vehicle communication system, and to an apparatus and method for controlling resource reselection in a vehicle communication system using reinforcement learning.

Typical standards for vehicle networks utilizing a band below 6GHz include Cellular Vehicle to Everything Communication (C-V2X) and IEEE 802.11p. In addition, the existing C-V2X standard includes C-V2X mode 4 (mode 4) based on automatic scheduling and C-V2X mode 3 (mode 3) based on network controlled scheduling.

Of these, in C-V2X mode 4, when each vehicle transmits a certain number of packets, it determines whether to use the existing resource block as it is or to select another resource block based on the sensing information. Therefore, the existing infrastructure (Road Side Unit (RSU), base station, etc.) is not used, so there is no additional latency that occurs while going through the infrastructure, whereas the efficiency of resource selection is lower than the technique using the infrastructure. .

In contrast, in C-V2X mode 3, resource blocks are directly allocated to vehicles based on information sensed by the base station. Therefore, additional latency occurs due to communication through the infrastructure, but there is a benefit of improving the data rate through efficient resource selection.

Meanwhile, the need for a study of MAC (Media Access Control) that can support the demand for a higher data rate is increasing due to the improvement of the level of automatic driving technology and the diversification of in-vehicle applications. Accordingly, research on directional transmission using more than 6 GHz, that is, millimeter wave, is being intensively conducted.

In C-V2X mode 3, in which resources are allocated based on information sensed by the base station, if all vehicles use a directional antenna in a frequency band of 6 GHz or higher, the data rate gain may decrease. However, in the case of C-V2X mode 4, there is a limitation that performance improvement is not significantly improved even when millimeter waves in a frequency band of 6 GHz or higher are used.

Korean Patent Publication No. 10-2019-0114757 (published on Oct. 10, 2019)

An object of the present invention is to provide an apparatus and method for controlling resource reselection of a vehicle communication system capable of improving frequency reselection efficiency.

Another object of the present invention is to provide an apparatus and method for controlling resource reselection of a vehicle communication system capable of improving a packet reception rate and an instantaneous data rate by improving the efficiency of frequency reselection.

The apparatus for controlling resource reselection of a vehicle communication system according to an embodiment of the present invention for achieving the above object receives location information and a packet reception ratio (PRR) from each of at least one vehicle, and receives the received at least The resource maintenance probability of each vehicle is initialized as the initial resource maintenance probability according to the number of surrounding vehicles of each vehicle determined from the location information of each vehicle, and thereafter, resources in response to a change in PRR of each received at least one vehicle The maintenance probability is updated and transmitted to the corresponding vehicle.

The apparatus for controlling resource reselection may obtain a resource maintenance probability by updating a weight in a predetermined manner according to a change in PRR received from each of at least one vehicle and applying the updated weight.

The resource reselection control device calculates the resource retention probability (p _i ) for each of at least one vehicle (i) from the weight (w _i )

(Where t is time, u(t) represents PRR as a utility function, and α is a utility weighting parameter).

The resource reselection control device analyzes the maximum value and the minimum value from the number of surrounding vehicles of each vehicle determined from the received location information of each of the at least one vehicle, and As the number approaches the maximum value, the initial resource retention probability may be set to be smaller, and as the number of nearby vehicles approaches the minimum value, the number may be set to be closer to a maximum value within a predetermined range.

_{The resource reselection control device calculates an initial resource retention probability (p 0} ) _{from the number of surrounding vehicles (n i} ) for each of at least one vehicle (i) and the determined maximum value (N) and the minimum value (n). expression

(Here, d _th is a reference distance for discriminating surrounding vehicles around the vehicle i).

Each of the at least one vehicle may be updated in the resource reselection control apparatus to change a resource reselection probability according to a resource maintenance probability, and determine whether to reselect a resource according to the changed resource reselection probability.

The resource reselection control device may be implemented as an infrastructure of a vehicle communication system.

A method for controlling resource reselection of a vehicle communication system according to another embodiment of the present invention for achieving the above object receives location information and packet reception rate (PRR) from each of at least one vehicle, Initializing a resource maintenance probability of each vehicle as an initial resource maintenance probability according to the number of surrounding vehicles of each vehicle determined from the location information; And thereafter, updating the resource maintenance probability in response to a change in the PRR of each of the received at least one vehicle.

Accordingly, in the apparatus and method for controlling resource reselection of a vehicle communication system according to an embodiment of the present invention, the infrastructure performs reinforcement learning based on information uploaded by each vehicle on a network consisting of a plurality of vehicles using the infrastructure and the directional antenna. Network parameters are updated through the network parameters, and the efficiency of frequency reselection can be improved by varying the resource reselection probability of each vehicle based on the updated network parameters. The rate can be improved.

1 is a diagram for explaining a concept in which each vehicle reselects a resource in C-V2X mode 4;
FIG. 2 is a diagram illustrating a concept of updating a resource reselection probability so that each vehicle can reselect a resource by a resource reselection control apparatus in a vehicle communication system according to an embodiment of the present invention.
FIG. 3 is a diagram illustrating a process of initializing a resource reselection probability based on the location of each vehicle by the resource reselection control apparatus of FIG. 2.
4 illustrates a method for controlling resource reselection in a vehicle communication system according to an embodiment of the present invention.
5 shows packet reception rate performance according to the resource reselection control method of the vehicle communication system of the present invention.

In order to fully understand the present invention, operational advantages of the present invention, and objects achieved by the implementation of the present invention, reference should be made to the accompanying drawings illustrating preferred embodiments of the present invention and the contents described in the accompanying drawings.

Hereinafter, the present invention will be described in detail by describing a preferred embodiment of the present invention with reference to the accompanying drawings. However, the present invention may be implemented in various different forms, and is not limited to the described embodiments. In addition, in order to clearly describe the present invention, parts irrelevant to the description are omitted, and the same reference numerals in the drawings indicate the same members.

Throughout the specification, when a certain part "includes" a certain component, it means that other components may be further included, rather than excluding other components unless specifically stated to the contrary. In addition, terms such as "... unit", "... group", "module", and "block" described in the specification mean a unit that processes at least one function or operation, which is hardware, software, or hardware. And software.

1 is a diagram for explaining a concept in which each vehicle reselects a resource in C-V2X mode 4;

In general, in C-V2X mode 4, resource reservation is made in a two-dimensional space of the time axis and the frequency axis as shown in FIG. 1, and in the time axis, the resource is divided into a plurality of sub-frames, and in the frequency axis. Resources are divided into a number of sub-channels. In addition, each block divided by a plurality of sub-frames and a plurality of sub-channels is a resource R, and may be composed of a plurality of resource blocks, respectively.

In C-V2X mode 4, since there is no separate coordinator for performing resource scheduling, there is a possibility of a collision when each vehicle transmits data. Accordingly, each vehicle repeatedly selects a resource by performing a process of sensing, selection, and reselection to minimize the occurrence of data collision.

In C-V2X mode 4, each vehicle first senses the resource use state of another vehicle before selecting a resource to transmit data. This is to prevent collision of transmitted data by analyzing the resource (R) used by other vehicles. Each vehicle senses each resource (R) and determines that the resource (R) is determined as being used by another vehicle is Set to in use, so that the resource is not selected.

During the sensing process, each vehicle monitors the usage status of each resource (R) during the time period (generally 1000ms) specified by the sensing window based on the received signal strength indicator (RSSI). do. In FIG. 1, since sub-frames are divided into 1 ms time intervals, 1000 sub-frames may be included in the sensing window. Therefore, the vehicle monitors the resource usage status of each sub-channel in each of the previous 1000 sub-frames based on RSSI.

As a result of monitoring, the resource R whose RSSI is equal to or greater than a predetermined threshold value is determined as a resource being used by another vehicle. That is, available and unusable resources are determined based on the RSSI measured for each resource (R).

After the sensing process, in the selection process, the vehicle selects a resource for transmitting data. In the selection process, the vehicle selects one of the resources R determined as available resources based on the RSSI within a time period designated by the selection window (here, for example, 100 ms). Here, the selection window is set to designate a subframe in which the vehicle can reserve a resource. If multiple vehicles frequently attempt to reserve resources, frequent collisions occur, and in general, a resource reservation interval (RRI) is designated to prevent this. The RRI may be set to 20 ms, 50 ms, and 100 ms, and the selection window has a size corresponding to the RRR. In FIG. 1, it is assumed that the RRI is 100 ms, and the selection window is set to a size including 100 subframes. In addition, the vehicle may select only one resource in the time interval corrected by the selection window.

At this time, the vehicle calculates the average RSSI (Average RSSI) of the resources (R) at intervals corresponding to the RRI based on the result of monitoring each resource (R) in the RRI unit of the sensing window during the sensing process, and the calculated average The resource corresponding to the lower 20% of the RSSI is selected as a selectable candidate resource, and one of the selectable candidate resources is selected. For example, the vehicle may select a resource with the minimum average RSSI.

At this time, the vehicle may also set a resource reselection counter (hereinafter referred to as RC) indicating the number of times to use the selected resource. Each vehicle can set the RC in a predetermined manner. For example, when the RRI is 100 ms, the vehicle can randomly select one of [5, 15], and when the RRI is 50, it can randomly select one of [10, 30].

And when the packet is transmitted the number of times specified by RC, the vehicle performs a reselection process. During the reselection process, the vehicle reselects resources according to a resource reselection probability (1-p) based on a preset resource keeping probability (p), or reuses the currently used resource as it is. Determine whether to do it. The resource retention probability (p) is usually set to a value between 0 and 0.8 (p=[0, 0.8]), and accordingly, the resource reselection probability (1-p) can be determined to a value between 0 and 20%. have.

If it is determined that the vehicle reselects a resource based on the resource reselection probability (1-p), the vehicle re-selects the resource based on the resource usage status monitored during the sensing window period in the sensing process in the same way as the selection process. Choose.

However, in the existing C-V2X mode 4, it was difficult to maintain or reselect resources adaptively to the surrounding environment as the resource reselection probability (1-p) of each vehicle is preset and fixed. However, in the vehicle communication system according to the present embodiment described below, each vehicle communicates with the infrastructure, and the infrastructure adjusts the resource reselection probability (1-p) of each vehicle. It allows resources to be maintained or reselected adaptively to the environment

The concept in which each vehicle reselects a resource in the C-V2X mode 4 is described above because each vehicle basically reselects a resource based on the C-V2X mode 4 in the vehicle communication system according to the present embodiment. .

FIG. 2 is a diagram for explaining the concept of updating a resource reselection probability so that each vehicle can reselect a resource by a resource reselection control apparatus in a vehicle communication system according to an embodiment of the present invention, and FIG. A diagram for explaining a process of initializing a resource reselection probability based on the location of each vehicle by the resource reselection control apparatus of.

Here, it is assumed that each vehicle has already selected a resource by performing an initial sensing process and a selection process for the resource, and is using the selected resource.

Referring to FIG. 2, in the vehicle communication system according to the present embodiment, each vehicle A is a localization scheme such as time-of-arrival (ToA) and time-of-arrival (TDoA). Use to acquire your own location information. Each vehicle can determine its own location information by detecting its surroundings from time to time using various sensors. Even before, in C-V2X mode 4, the vehicle uses sensors to detect its own location information and surrounding environment. Therefore, it is assumed here that each vehicle is always detecting its own location information.

In addition, when the vehicle transmits the packet the number of times specified by the RC and performs the reselection process, the packet reception ratio (hereinafter referred to as PRR) from the previous selection window to the present is measured.

When the PRR is measured, the vehicle uplink transmits the measured PRR to the resource reselection control device (IS) along with its location information. Here, the resource reselection control device IS is an infrastructure provided to the vehicle communication system, and may be a roadside unit (RSU) or a base station.

As described above, in the case of the C-V2X mode 3 in the vehicle communication system, the infrastructure is used, while the C-V2X mode 4 does not use the infrastructure. However, in this embodiment, each vehicle performs communication based on the C-V2X mode 4, but the infrastructure functions as a resource reselection control device (IS) to adjust the resource reselection probability (1-p) of each vehicle. To be able to. This is because the infrastructure functioning as the resource reselection control device (IS) changes the resource reselection probability (1-p) of each vehicle adaptively to the vehicle's surrounding environment, that is, the road environment, thereby improving the PRR of each vehicle. It is for sake.

When the location information and the PRR are applied from the vehicle, the resource reselection control device IS initializes the resource reselection probability (1-p) of each vehicle in a predetermined manner based on the location information of the vehicle.

Thereafter, resource reselection probability (1-p) of each vehicle is adjusted by updating the resource retention probability (p) for each vehicle using the reinforcement learning method.

Resource re-selection control unit (IS), as illustrated in Figure 3 first, the other vehicle is located within, based on the vehicle position information sent from the plurality of vehicle groups based on a specified distance around the location of each vehicle (d _th) Count the number. The resource reselection control device (IS) counts the number of other vehicles (n _{i ) located within a reference distance (d th} ) with respect to each location of a plurality of vehicles (i) to which the location information is transmitted, and the maximum value ( Determine N = max(n _i )) and minimum value (n = min(n _i )).

Then, the resource maintenance probability (p) of each vehicle (i) is initialized in a known manner using the determined maximum value (N) and the minimum value (n) to obtain _{an initial resource maintenance probability (p 0 ).}

The resource reselection control apparatus IS may obtain an _{initial resource maintenance probability p 0 according to Equation 1, for example.}

According to Equation 1, the resource reselection control device (IS) _zeroes the initial resource maintenance probability (p 0) according to the number of other vehicles (n _{i ) located within the reference distance (d th) from the vehicle (i).} Normalize to a value between 0.8 (p ₀ = [0, 0.8]).

That is, as the number of other vehicles (n _{i ) located within the reference distance (d th} ) from the i-th vehicle (i) approaches the maximum value (N), the initial resource retention probability (p ₀ ) is set smaller, and other vehicles As the number of n _i approaches the minimum value n, the predetermined initial resource retention probability p ₀ is set to be closer to the maximum retention probability (here, as an example, 0.8).

Then, the initial resource retention probability (p ₀ ) set for each of the plurality of vehicles i is transmitted to the corresponding vehicle, thereby initializing the resource reselection probability (1-p) of each vehicle.

Therefore, as there are more other vehicles around each vehicle, the resource retention probability (p) decreases, and the resource reselection probability (1-p) obtained from the resource retention probability (p) increases. In addition, as there are fewer other vehicles around each vehicle, the resource retention probability (p) increases, and the resource reselection probability (1-p) decreases.

This is to increase the resource reselection efficiency of each vehicle even at the beginning of learning by allowing vehicles with a high probability of causing or receiving interference because there are many vehicles around them to reselect resources more inconveniently.

Therefore, it can be seen that the resource reselection control apparatus IS initializes the resource reselection probability (1-p) by recognizing the location of the vehicle during the initialization process.

Thereafter, the resource reselection control apparatus IS updates the resource maintenance probability p of each vehicle based on the PRR value for each vehicle in a reinforcement learning method.

)을 획득하는 것을 목적으로 한다.In this embodiment, the resource reselection control apparatus IS updates the resource retention probability p in order to improve the PRR of each vehicle. That is, the resource reselection control apparatus (IS) maximizes the PRR representing the ratio of the received packet to the transmitted packet according to Equation 2 (

It aims to acquire ).

)을 직접 획득하는 것은 현실적으로 매우 어렵다.However, since the road environment is always changing, the resource reselection control device (IS) maximizes the PRR represented by Equation 2 (

) Is very difficult in reality.

Accordingly, the resource reselection control apparatus IS according to the present embodiment updates the resource retention probability p using the reinforcement learning algorithm according to Equation 3, so that the PRR of each vehicle can be improved.

Referring to Equation 3, the resource reselection control apparatus IS _{updates the resource retention probability p i} for each vehicle i based on the weight w _i (t). Here, the next weight (w _i (t+1)) is determined by the current weight (w _i (t)) and the utility function (u(t)), and the utility function (u(t)) is set to PRR. In other words, the next weight (w _i (t+1)) is the resource for the _{current weight (w i} (t)) and the utility weight parameter (α) and weight (w _{i) pre-specified in the difference between the current PRR and the previous PRR.} It is obtained based on the rate of change of the retention probability (p _{i ).}

Resource re-selection control unit (IS) is to update the resource holding probability (p _i) on the basis of the equation (3), and transmitting downlink to a vehicle (i) corresponding to keep updated resource probability (p _i), each vehicle The resource reselection probability (1-p) of (i) is updated.

And each vehicle (i) determines whether to reselect a resource based on the updated resource reselection probability (1-p), and when it is determined that the resource is reselected, a selection window is set, and a sensing window is performed in the sensing process. Based on the result of monitoring each resource (R) in RRI units of, a resource corresponding to one of the resources corresponding to the lower 20% of the average RSSI in the sensing window is selected within the set selection window. Then, the data is transmitted to another vehicle using the selected resource.

4 illustrates a method for controlling resource reselection in a vehicle communication system according to an embodiment of the present invention.

Referring to FIGS. 2 and 3, the method of controlling resource reselection of FIG. 4 is first obtained by receiving location information and PRR from each of at least one vehicle (S10). As described above, each of the at least one vehicle may determine its own location information using various sensors, and measure a PRR from the previous selection window to the present. In addition, the resource reselection control apparatus (IS) receives and acquires the location information and PRR transmitted uplink from each vehicle.

When the location information of each of the at least one vehicle is acquired, the resource retention probability (p) of each vehicle is initialized _{to the initial resource retention probability (p 0) according to the number of surrounding vehicles of each vehicle determined based on the acquired location information.} maintaining the initial resource probability (p ₀₎ of the downlink transmission to the vehicles keep each vehicle is transmitted initial resource probability (p ₀₎ an initial resource reselection probability of a resource reselection probability (1-p) in accordance with (1 - p ₀ ) to be changed (S20).

In order to change the resource reselection probability (1-p) of each vehicle to the initial resource reselection probability (1-p ₀ ), the resource reselection control device (IS) first focuses on each of at least one vehicle (i). The number of surrounding vehicles (n _i ) within a predetermined reference distance (d _th ) is counted (S21). The maximum value (N = max(n _i )) and the minimum value (n = min(n _{i )) of the number of surrounding vehicles (n i} ) counted for each of the at least one vehicle are determined (S22).

When the maximum value (N) and minimum value (n) are determined, based on the determined maximum value (N) and minimum value (n) and the number of surrounding vehicles (n _i ) counted for each of at least one vehicle, each vehicle is The initial resource maintenance probability (p ₀ ) for is obtained in a predetermined manner as shown in Equation 1 (S23). At this time, the initial resource retention probability (p ₀ ) can be obtained by normalizing to have a value in a predetermined range ([0, 0.8]), and other vehicles located within a _{reference distance (d th) around each of at least one vehicle} As the number of (n _i ) approaches the maximum value (N), the initial resource retention probability (p ₀ ) is set smaller, and the closer the number of other vehicles (n _i ) to the minimum value (n), the maximum within a specified range. It can be set to be close to the value.

And, by transmitting the set initial resource retention probability (p ₀ ) for each vehicle to the corresponding vehicle, each vehicle changes the resource reselection probability (1-p) to the initial resource reselection probability (1-p ₀ ). (S24).

When the initial resource retention probability (p ₀ ) is obtained, the resource retention probability (p) is then updated in a reinforcement learning method to improve the PRR of each vehicle (S30).

In order to update the initial resource retention probability p ₀ , the resource reselection control apparatus IS receives and obtains a PRR from each vehicle (S31). At this time, the location information of each vehicle may be authorized together. Then, based on the change in the received PRR, the weight (w _i (t)) is calculated in a known manner, and the resource retention probability (p) is updated using the _{calculated weight (w i (t)) (} S32). When the resource retention probability (p) is updated, the updated resource retention probability (p) is transmitted to the corresponding vehicle, and each vehicle changes the resource reselection probability (1-p) according to the updated resource retention probability (p). Let it be (S24).

5 shows packet reception rate performance according to the resource reselection control method of the vehicle communication system of the present invention.

Figure 5 is a result of measuring the PRR of the vehicle according to the progress of time, (a) shows a case in which the vehicle location recognition-based resource maintenance probability initialization process and the resource maintenance probability update process by reinforcement learning are performed according to the present embodiment. Show. (b) shows a case in which the resource maintenance probability is initialized at random, and then the resource maintenance probability update process by reinforcement learning is performed. And (c) shows a case in which a directional antenna and a beamforming technique are applied to the existing C-V2X mode 4.

In FIG. 5, both (a) to (c) appear in a form in which the PRR gradually increases with time, but when the resource retention probability update process by reinforcement learning is performed as shown in (a) and (b), the PRR is It can be seen that it appears even higher. In addition, in the case of performing the process of initializing the resource maintenance probability based on vehicle location recognition as shown in (a), it can be seen that the PRR is further improved than the case of randomly initializing the resource maintenance probability.

This improves the reselection efficiency of the resource corresponding to the frequency, and can improve the instantaneous data rate together with the PRR of each vehicle.

The method according to the present invention can be implemented as a computer program stored in a medium for execution on a computer. Here, the computer-readable medium may be any available medium that can be accessed by a computer, and may also include all computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, and ROM (Read Dedicated memory), RAM (random access memory), CD (compact disk)-ROM, DVD (digital video disk)-ROM, magnetic tape, floppy disk, optical data storage device, and the like.

The present invention has been described with reference to the embodiments shown in the drawings, but these are merely exemplary, and those of ordinary skill in the art will understand that various modifications and equivalent other embodiments are possible therefrom.

Therefore, the true technical protection scope of the present invention should be determined by the technical spirit of the appended claims.

Claims (14)

Resource maintenance probability of each vehicle according to the number of neighboring vehicles of each vehicle determined from the location information of each of the received at least one vehicle by receiving location information and packet reception ratio (PRR) from each of at least one vehicle Is initialized to the initial resource retention probability,
Thereafter, the resource retention probability is updated in response to a change in the PRR of each of the received at least one vehicle and transmitted to the corresponding vehicle,
The weight is updated in a predetermined manner according to the change of the PRR received from each of the at least one vehicle, and the updated weight is applied to obtain a resource maintenance probability,
Equation _{of the resource maintenance probability (p i} ) for each of at least one vehicle (i) from the weight (w _i)

(Where t is time, u(t) represents PRR as a utility function of reinforcement learning, and α is a predefined utility weighting parameter)
Resource reselection control device of the vehicle communication system obtained according to.
delete delete The method of claim 1, wherein the resource reselection control device
Analyzing the maximum and minimum values from the number of surrounding vehicles of each vehicle determined from the received location information of each of at least one vehicle, and as the number of surrounding vehicles located within a reference distance around each of at least one vehicle approaches the maximum value A resource reselection control apparatus of a vehicle communication system configured to set the initial resource retention probability to be small, and set so that the number of nearby vehicles approaches the minimum value, the closer to the maximum value within a predetermined range. The method of claim 4, wherein the resource reselection control device
Equation _{of the initial resource retention probability (p 0 ) from the number of surrounding vehicles (n i} ) for each of at least one vehicle (i) and the determined maximum value (N) and the minimum value (n)

(Where d _th is the reference distance for determining the surrounding vehicle around the vehicle (i), and n _i is the number of other vehicles located within the reference distance around the vehicle (i))
Resource reselection control device of the vehicle communication system obtained according to. The method of claim 1, wherein each of the at least one vehicle
A resource reselection control device of a vehicle communication system that is updated in the resource reselection control device to change a resource reselection probability according to a resource maintenance probability, and to determine whether to reselect a resource according to the changed resource reselection probability. The method of claim 1, wherein the resource reselection control device
Resource reselection control device implemented as an infrastructure of a vehicle communication system. By receiving location information and packet reception rate (hereinafter referred to as PRR) from each of at least one vehicle, the initial resource maintenance probability of each vehicle is maintained according to the number of surrounding vehicles of each vehicle determined from the location information of each of the received at least one vehicle Initializing with probability; And
Thereafter, including the step of updating the resource maintenance probability in response to a change in the PRR of each of the received at least one vehicle,
The updating step
Updating a weight in a predetermined manner according to a change in PRR received from each of the at least one vehicle; And
Comprising the step of obtaining the updated resource retention probability by applying the updated weight,
Updating the weights
Equation _{of the weight (w i} ) for each of at least one vehicle (i)

(Where t is time, u(t) represents PRR as a utility function of reinforcement learning, and α is a predefined utility weighting parameter)
Acquired according to, and
The step of obtaining the updated resource maintenance probability
Equation of resource retention probability (p _i)

Resource reselection control method of the vehicle communication system obtained according to the.
delete delete delete The method of claim 8, wherein the initializing step
Analyzing a maximum value and a minimum value from the number of surrounding vehicles of each vehicle determined from the received location information of each of the at least one vehicle; And
The initial resource retention probability is set smaller as the number of surrounding vehicles located within a reference distance from each of at least one vehicle approaches the maximum value, and the closer the number of surrounding vehicles approaches the minimum value, the closer to the maximum value within a predetermined range. Resource reselection control method of a vehicle communication system comprising the step of setting to be. The method of claim 12, wherein the setting step
Equation _{of the initial resource retention probability (p 0 ) from the number of surrounding vehicles (n i} ) for each of at least one vehicle (i) and the determined maximum value (N) and the minimum value (n)

(Where d _th is the reference distance for determining the surrounding vehicle around the vehicle (i), and n _i is the number of other vehicles located within the reference distance around the vehicle (i))
Resource reselection control method of the vehicle communication system set according to. The method of claim 8, wherein each of the at least one vehicle
Changing a resource reselection probability according to the updated and transmitted resource maintenance probability; And
Resource reselection control method of a vehicle communication system comprising the step of determining whether to reselect resources according to the changed resource reselection probability.

Images