KR20210070062A - An adaptive clustering method for massive IoT and the sensor node thereof - Google Patents

Abstract

The present invention relates to an adaptive clustering method for massive IoT and a sensor node thereby. A method for setting up a network protocol for a large-scale sensor node network includes the operations of: setting a plurality of groups to which the large-scale sensor node belongs and group information of each group; setting a slot frame and a plurality of time slots in the slot frame; setting a group member and group leader to perform communication in each time slot; and optimizing the number of group heads for each group. Accordingly, the power consumption of the group leader is optimized in the wireless sensor network, so that the life of the wireless sensor network extends.

Description

An adaptive grouping communication method for massive IoT and a sensor node performed by the method

Various embodiments relate to an adaptive grouping communication method for a large-scale Internet of Things and a sensor node performed by the method.

A plurality of sensor nodes play the role of sensing and information transfer, and are wirelessly connected to each other to collect the sensed information to one sink node or base station, or vice versa, information or commands from the sink node or base station to each node. A network that transmits the data is called a wireless sensor network.

In general, a large number of sensor nodes in a wireless sensor network are deployed to obtain information from a wide area, and for this reason, each sensor node operates with a battery rather than a wired power supply.

Therefore, the lifespan of the wireless sensor network is directly related to the battery life of the sensors, and in order to increase the lifespan of the wireless sensor network, the battery life should be increased, and in order to increase the battery life, the power used by each sensor node should be reduced as much as possible.

One of the biggest causes of power consumption in a sensor node is a signal transmission/reception process for communication with another sensor node, a sink node, or a base station, and power consumption in the transmission/reception process should be reduced.

An object of the present invention is to solve the problem of the large-scale Internet of Things as described above, and communication is performed by a synchronous medium access control (MAC, Medium Access Control) protocol in a wireless sensor network environment, but is consumed through grouping of nodes. An object of the present invention is to provide an adaptive grouping communication method for large-scale Internet of Things that controls power, delay, and throughput according to the type of application.

In addition, an object of the present invention is that nodes are grouped and communicate with the base station by time division multiple access, but the roles are divided into group leader and group member, and the base station communicates with the group leader and the group leader communicates with the base station or group member. It is to provide an adaptive grouping communication method for large-scale Internet of Things, which controls power consumption, delay, and throughput according to the type of application by adjusting the number of group heads according to the network environment.

The technical problems to be achieved in this document are not limited to the technical problems mentioned above, and other technical problems not mentioned can be clearly understood by those of ordinary skill in the art to which the present invention belongs from the description below. There will be.

According to various embodiments of the present disclosure, a method for setting a network protocol for a large-scale sensor node network includes an operation of setting a plurality of groups to which the large-scale sensor node belongs and a grouping of each group, a slot frame and a plurality of times within the slot frame It may include an operation of setting a slot, an operation of setting a group member and group leader to perform communication in each time slot, and an operation of optimizing the number of group heads for each group.

According to various embodiments, the method and apparatus proposed by the present disclosure may reduce the amount of power consumed in each node by lowering a collision probability during signal transmission between nodes.

According to various embodiments, the method and apparatus proposed in the present disclosure can reduce power consumption and minimize interference between groups by grouping nodes based on location to enable intra-group communication even with low power.

According to various embodiments, the method and apparatus proposed by the present disclosure allow one group to perform intra-group communication and the other group to perform out-of-group communication, thereby minimizing interference caused by out-of-group communication on intra-group communication.

Effects obtainable in the present disclosure are not limited to the above-mentioned effects, and other effects not mentioned may be clearly understood by those of ordinary skill in the art to which the present disclosure belongs from the description below. will be.

1 is a diagram illustrating an example of a large-scale IoT environment including large-scale sensor nodes.
2 is a diagram illustrating an example in which time slots are allocated to a group according to a slot frame and the above-described allocation method.
3 is a diagram illustrating an example of an interference region that may be caused during communication.
4 is a flowchart illustrating an example of setting a network protocol in a base station for data communication in a network including large-scale sensor nodes.
5 is a flowchart illustrating an example in which the base station determines the number of group heads of a specific group.
6 is a diagram showing the configuration of a base station.
In connection with the description of the drawings, the same or similar reference numerals may be used for the same or similar components.

Hereinafter, the embodiments disclosed in the present specification will be described in detail with reference to the accompanying drawings, but the same or similar components are assigned the same reference numerals regardless of reference numerals, and redundant description thereof will be omitted.

The suffix 'module' or 'part' for the components used in the following description is given or mixed in consideration of only the ease of writing the specification, and does not have a meaning or role distinct from each other by itself. In addition, a 'module' or 'unit' may mean a software component or a hardware component such as an FPGA (field programmable gate array) or ASIC (application specific integrated circuit), and the 'unit' or 'module' refers to what role carry out the However, 'part' or 'module' is not meant to be limited to software or hardware. A 'unit' or 'module' may be configured to reside on an addressable storage medium or may be configured to reproduce one or more processors. Thus, as an example, 'part' or 'module' refers to components such as software components, object-oriented software components, class components and task components, processes, functions, properties, may include procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables. Components and functionality provided in 'units' or 'modules' may be combined into a smaller number of components and 'units' or 'modules' or additional components and 'units' or 'modules' can be further separated.

The steps of a method or algorithm described in connection with some embodiments of the present invention may be implemented directly in hardware executed by a processor, a software module, or a combination of the two. A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of recording medium known in the art. An exemplary recording medium is coupled to the processor, the processor capable of reading information from, and writing information to, the storage medium. Alternatively, the recording medium may be integral with the processor. The processor and recording medium may reside within an application specific integrated circuit (ASIC). The ASIC may reside within the user terminal.

In describing the embodiments disclosed in the present specification, if it is determined that detailed descriptions of related known technologies may obscure the gist of the embodiments disclosed in the present specification, the detailed description thereof will be omitted. In addition, the accompanying drawings are only for easy understanding of the embodiments disclosed in the present specification, and the technical spirit disclosed herein is not limited by the accompanying drawings, and all changes included in the spirit and scope of the present invention , should be understood to include equivalents or substitutes.

Terms including an ordinal number, such as first, second, etc., may be used to describe various elements, but the elements are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

When it is said that a component is 'connected' or 'connected' to another component, it is understood that it may be directly connected or connected to the other component, but other components may exist in between. it should be On the other hand, when it is mentioned that a certain element is 'directly connected' or 'directly connected' to another element, it should be understood that the other element does not exist in the middle.

sensor network

A plurality of sensor nodes play the role of sensing and information transfer, and are wirelessly connected to each other to collect the sensed information into one sink node or base station, or vice versa, information or information from the sink node or base station to each sensor node. A network that transmits commands is called a wireless sensor network.

In general, sensor nodes in a wireless sensor network are very numerous and want to obtain information from a wide area, so that each sensor operates with a battery rather than a wired power supply. Therefore, the lifespan of the wireless sensor network is directly related to the battery life of the sensors, and in order to increase the battery life, the power consumption of each sensor node should be minimized.

One of the biggest causes of power consumption of a sensor node is a signal transmission/reception process for communication. In order to reduce power consumption in this transmission/reception process, a medium access control (MAC) end of the network uses a method called duty cycling. use.

Duty cycling:

Duty cycling is a technique in which a transceiver or a transceiver or a radio unit including a modem used to transmit or receive a radio signal (or it may be referred to as a communication unit) is turned on for a certain period of time and then turned off for a certain period of time. However, it is mainly used in network applications where energy consumption needs to be reduced because it consumes less power than when the radio is always on.

However, this turning on/off process requires synchronization between the transmitting and receiving parties, which are the parties performing the communication. Otherwise, transmission and reception are not performed. This is because, when the transmitting side transmits a signal to send a packet, if the radio unit of the receiving side is turned off, the signal cannot be received and packet loss occurs.

Synchronization in the process of turning on and off the radio unit at the transmitting side and the receiving side can be performed by the MAC protocol. In the present invention, the MAC protocol called time division multiple access (TDMA) is the basis, and a large number of sensor nodes are required. We propose a new MAC protocol suitable for a massive Internet of Things (IoT) environment.

time-division multiple access (TDMA)

It is a MAC protocol whose purpose is to keep the receiver awake when the transmitter sends a radio signal to transmit a packet. The method used by TDMA is to set the time for both the transmitter and receiver to stay awake.

Due to this, when the transmitting side has information to send to the receiving side, it puts it in a packet and waits until the specified time arrives, then turns on the radio to transmit the packet when the specified time arrives, and the receiving side can receive the packet at the specified time.

In general, in the time division multiple access method, since the transmitting side and the receiving side are allocated one time slot for communication among a plurality of divided time slots set within a large time period called a slot frame, the receiving side is In the slot frame period, the radio unit is turned on only for the time slot and the packet is received, and whenever a packet to be transmitted is generated, the transmitter transmits the packet within the time slot according to the scheduled time slot.

Massive IoT

Recently, the number of communicable devices per unit area has greatly increased due to the rapid increase in IoT devices and smartphones, and the scale of the wireless sensor network may be large enough to include up to 10,000 nodes. Such a large-scale IoT network is called Massive IoT.

Massive IoT nodes generally assume a very low transmission rate, but nevertheless, since the number of nodes is very large, when a general MAC protocol is used, the throughput is very low, the packet loss rate, and the delay time are very There are difficulties in increasing.

exponential backoff

If two or more transmitters located at a distance from the interference transmit packets at the same time, a collision occurs and retransmission is required. If the transmitters wait for the same amount of time (backoff) and then transmit packets again at the same time, a collision will occur again. There are several ways to do this, and one of them is exponential backoff. Exponential backoff means that the window size is set to 1, 2, 4, 8, 16,?? This is a method in which an integer random number is generated from a minimum of 1 to a maximum (window size-1), and the unit time for the random number is waited for and then retransmitted. As the collision repeatedly occurs, the random number generation range increases, so the probability of re-collision decreases exponentially.

1 is a diagram illustrating an example of a large-scale IoT environment including large-scale sensor nodes.

Referring to FIG. 1 , large-scale sensor nodes are scattered around the base station 100 , and the sensor nodes may wish to communicate with other sensor nodes or with an external device (eg, a control server) through the base station 100 . have.

In this case, each sensor node is highly likely to be operated depending on the battery. In this case, power consumption must be minimized to prolong the lifespan of the sensor node.

The present invention proposes a method for minimizing the power used by each sensor node for communication in a large-scale IoT environment as shown in FIG. 1 .

To this end, the present invention proposes a method of communicating by grouping sensor nodes.

Initialization method

The method proposed in this method initializes large-scale sensor nodes in the following order.

1. Nodes may be grouped based on location so that the total amount of traffic of each group (210, 220, 230, 240) is similar. According to an embodiment, in the case of sensor nodes, since the size and amount of packets transmitted by each sensor node are highly likely to be constant, each group may be grouped to include the same number of sensor nodes. Also, nodes that are close to each other may be set as one group.

2. Any node in each group (210, 220, 230, 240) can be designated as the group leader. According to an embodiment, one arbitrarily selected one of the sensors in the group may be designated as the group leader.

3. A value obtained by multiplying the number of groups by the maximum consumption time of packet transmission/reception can be designated as the length of a slot frame. According to another embodiment, the length of the slot frame may be designated as a value obtained by multiplying the number of groups by a predetermined time.

4. Slot frames can be equally divided by the number of groups. That is, the slot frame may have as many time slots as the number of groups. Then, one of the time slots of the slot frame is allocated to the group leader of each group in a round-robin manner.

5. The time slot allocated to the group leader of a specific group is also allocated to the group members of the group most distant from the specific group.

2 is a diagram illustrating an example in which time slots are allocated to a group according to a slot frame and the above-described allocation method.

Referring to FIG. 2 , the number of groups may be six. Accordingly, the slot frame 300 may include six time slots 310 to 315 . According to an embodiment, the length of each time slot may be a maximum consumption time required for packet transmission and reception.

Each time slot is assigned to the group leader of a specific group in a round-robin manner, and at the same time assigned to the group members of the group furthest away from the specific group. Referring to FIG. 2 , the first time slot 310 is allocated to the group leader G1h of the G1 group and at the same time allocated to the group member G4 of the G4 group, and the second time slot 311 is assigned to the group leader G2h of the G2 group. is allocated to the group member G5 of the G5 group while being allocated, the third time slot 312 is allocated to the group head G3h of the G3 group and is allocated to the group member G6 of the G6 group at the same time, the fourth time slot 313 ) is assigned to the group head G4h of group G4 and at the same time assigned to the group member G1 of the G1 group, and the fifth time slot 314 is assigned to the group head G5h of the G5 group and at the same time the group member of the G2 group ( G2), and the sixth time slot 315 may be allocated to the group leader G6h of the G6 group and at the same time allocated to the group member G3 of the G3 group.

how it works

As described above, after initialization that determines a group of sensor nodes and an initial one group head, the base station 100 solves an optimization problem to be described later for each group. The optimization problem here is to determine the number of group heads that should exist in each group.

Step 1: If the number of current group heads is less than the optimal value, the number of group heads is increased by 1, and if the current number of group heads is larger than the optimal value, the number of group heads is decreased by 1. If the number of group heads is equal to the optimal value, leave it as it is.

Step 2: If the number of group heads changes, group members (sensor nodes) assigned to each group head are rearranged. When assigning group members to each group head, the group members are assigned to each group head so that the sum of the average traffic of the group members is as uniform as possible. Then, the above optimization problem is solved again and again until the number of group heads is equal to the optimal value.

optimization problem

In general, an application in which the MAC protocol is used can be replaced with an optimization problem that maximizes the lifetime of the network (or sensor node) under the constraints of delay and throughput. When using the above protocol, it can be replaced with an optimization problem described later. By solving this optimization problem, it is possible to determine the optimal number of group heads for each group.

3 is a diagram illustrating an example of an interference region that may be caused during communication.

Referring to FIG. 3 , a dotted circle indicates an interference area caused by communication between the group leader of the G1 group and the base station 100, and a solid line shows an example of an interference area caused by communication between the group leader and group member in each group did it Since the group leader and the group member are close in distance and can perform communication with lower power, the area that can cause interference is also limited, but the group leader uses a fairly high power to communicate with the base station 100 . Therefore, the area that can cause interference is also widened.

Referring to FIG. 2 , one time slot is allocated so that the group leader of one group communicates with the base station 100 and at the same time communicates with the group member of another distant group. Therefore, when a plurality of group heads in one group (eg, the G1 group in the time slot 310 ) try to transmit a packet to the base station 100 , a collision may occur and packet transmission may fail. Also, when at least two group members among group members of one group (eg, the G4 group in the time slot 310) try to transmit a packet, a collision may occur and packet transmission may fail. However, as shown in FIG. 3, since the interference area due to packet transmission by the group head of the G1 group does not reach the G4 group, packet transmission by the group head of the G1 group and packet transmission by the group member of the G4 group in the same time slot Even if there is, a collision may not occur with each other.

If the packet is normally transmitted, the receiving node or base station 100 may transmit an ACK packet to inform the group member or the group leader that the packet has been normally received. However, when a collision occurs, the receiving node or the base station 100 cannot transmit the ACK packet because the normal packet reception has not occurred, and the node that transmitted the packet fails to receive the ACK packet within a certain time and thus fails to transmit the packet. Able to know.

A node recognizing that packet transmission has failed may retransmit the packet through exponential backoff, and the unit time of exponential backoff may be a slot frame time. According to an embodiment, if the generated random number is 3, back-off by three slot frame time and retransmission may be attempted.

The notation of constants or variables used in the base station to solve the optimization problem of determining the number of group heads optimized for each group is as follows.

N _G : Number of groups.

L: Slot frame length. The unit is seconds.

S: time slot length. The unit is seconds, and it can be obtained as _{S=L/N G.}

R: Baud rate. The unit is byte/second.

B: Packet length. The unit is bytes.

p _c (n,R): probability of packet loss due to contention of n nodes at data rate R

M _i : The number of group heads in the i-th group.

P _R : Received power or standby power.

P _h : Transmission power during communication between the base station and the group head.

Pi: Transmission power during intra-group communication and between group members and group heads. smaller than P _h.

The optimization problem to be solved by the base station is to minimize the power consumption of the group leader under the constraints of a given delay time and throughput.

Here, the limiting condition for the delay time is as follows [Equation 1].

Here, B may be any value set according to an actual application or by a base station.

The limiting conditions for the throughput are as follows [Equation 2].

Here, A may be any value set according to an actual application or by a base station.

The power consumption of the group leader is as follows [Equation 3].

_{The base station 100 may obtain the number of group heads M i} in which [Equation 3] can be minimized while satisfying [Equation 1] and [Equation 2] for each group. Since it is common that the number of group heads obtained according to the above equations is a real number, it is possible to obtain an integer value by rounding. The base station 100 may add or remove group heads of each group according to step 1 of the above-described operation method by comparing the integer values obtained by solving the above-described optimization problem for each group with the current number of group heads of each group.

By having such an optimization problem, the power consumption of the group leader, the node that uses the most power in the wireless sensor network, can be minimized and the overall network lifespan can be maximized.

The results of the above [Equation 1] to [Equation 3] can be obtained as follows.

로 결정되고, 패킷당 전송 시간(t_pk)은 B/R(second/packet)로 결정된다.If the average packet arrival rate of the nodes is μ (packets/second) and the number of nodes in the group is N _i , the number of group members assigned to each group head of the i-th group is

is determined, and the transmission time per packet (t _pk ) is determined by B/R (second/packet).

)은 할당된 그룹원과 자기 자신의 평균 패킷 도착률의 합인

로 결정될 수 있다. 그러면 i번째 그룹의 그룹장이 1초중에서 패킷들을 보내는데 걸리는 시간은

로 결정되고, i번째 그룹의 그룹장의 송신 소모 전력은

이고, i번째 그룹의 그룹장의 수신 소모 전력은

이다. 그리고 i번째 그룹의 그룹장의 충돌에 의한 재전송 소모 전력은

이다.Since the group leader has to deliver his/her own packet to the base station 100 with the group members below the group leader, the average packet arrival rate (

) is the sum of the assigned group members and their own average packet arrival rate.

can be determined as Then, the time it takes for the group leader of the i-th group to send packets out of 1 second is

is determined, and the transmission power consumption of the group head of the i-th group is

and the reception power consumption of the group head of the i-th group is

to be. And the power consumption of retransmission due to the collision of group heads of the i-th group is

to be.

이고,

이고,

로 구해질 수 있다.therefore,

ego,

ego,

can be saved with

4 is a flowchart illustrating an example of setting a network protocol in the base station 100 for data communication in a network including large-scale sensor nodes.

Referring to FIG. 4 , in operation 410 , the base station 100 may group large-scale sensor nodes. The base station 100 may group nodes that are close to each other to form one group while the total amount of traffic transmitted from the sensors of each group to the base station is similar. In addition, any one of the group members (sensor nodes) included in one group may be set as the group leader.

In operation 420 , the base station 100 may set a slot frame for transmitting a packet from the large-scale sensor node to the base station 100 . According to an embodiment, the base station 100 may determine the length of the slot frame as a value obtained by multiplying the number of groups set in operation 410 by the maximum time required for packet transmission and reception. And the slot frame may include the same number of time slots as the number of groups.

In operation 430, the base station 100 may set a group member and group leader to transmit packets for each time slot. As shown in FIG. 2 , the group leader belonging to one group may transmit a packet to the base station 100 in one time slot, and a group member belonging to another group may transmit a packet to the group leader at the same time. Accordingly, the base station 100 may set a group to which a group leader to transmit a packet belongs and a group to which a group member to transmit a packet for a specific time slot belongs. Referring to FIG. 2 , in a time slot 310 , a group leader belonging to the G1 group and a group member belonging to the G4 group may be configured to transmit packets.

In operation 440, the base station 100 may optimize the number of group heads for each group. According to various embodiments, a specific group may have a plurality of group heads. In operation 410, one group head is arbitrarily set, but a plurality of group heads may be set based on the optimization result in operation 440 .

5 is a flowchart illustrating an example in which the base station 100 determines the number of group heads of a specific group.

Referring to FIG. 5 , in operation 510 , the base station 100 may allocate group members to group heads. According to an embodiment, the base station 100 may allocate the same number of group members to each group head. For example, if there are 104 nodes in one group, 4 nodes act as group heads, and 100 nodes act as group members, 25 group members can be assigned to each group leader. If there is one group leader initially, all group members can be assigned to the group leader.

In operation 520 , the base station 100 may optimize the number of group heads based on the group member allocation in operation 510 . According to an embodiment, the optimized number of group heads (M _i ) may be determined based on [Equation 1] to [Equation 3].

In operation 530, the number of optimized group lengths (M _i ) obtained in operation 520 may be compared with the current number of group lengths.

In operation 540, if the result of the comparison in operation 530 is the same, the operation is terminated while maintaining the current number of group heads. If not, operation 550 may be performed.

In operation 550, the base station 100 is the number of optimization geurupjang (M _i) is increased by one the number of current geurupjang greater than the current number of geurupjang of and, the number of optimization geurupjang (M _i) is less than the number of the current geurupjang If so, the current number of group leaders is decreased by 1. Then, operations 510 to 540 may be performed again.

The base station 100 may determine the optimized current number of group heads in each group by repeating operations 510 to 550 until the _{optimized number of group heads (M i ) and the current number of group heads are the same.}

6 is a diagram illustrating the configuration of the base station 100 .

Referring to FIG. 6 , the base station 100 may include a communication unit 110 and a processor 120 .

The communication unit 110 may perform a communication function of transmitting and receiving packets to and from the sensor nodes, and preferably may communicate with the sensor nodes selected as a group leader.

The processor 120 may perform an operation according to the MAC protocol proposed by the present invention. According to an embodiment, there may be a plurality of processors 120 . According to an embodiment, the processor 120 may perform an operation according to the flowchart illustrated in FIG. 4 and/or an operation according to the flowchart illustrated in FIG. 5 .

The above-described operation according to the MAC protocol proposed in the present invention may be dynamically performed by the processor 120 of the base station, but according to another embodiment, an external server performs the corresponding operation and the base station receives the result from the external server , can also be performed according to the results.

As described above, the reason for grouping nodes in an IoT environment including large-scale sensor nodes and giving each group a different time slot for communication is the number of nodes competing for packet transmission within the same time compared to when not doing so. This is to reduce the probability of packet transmission collision by reducing .

In addition, since a uniform time slot is used, the probability of collision can be reduced by grouping so that the number of nodes in each group is as uniform as possible.

In the case of grouping in this way, two types of communication should be considered, intragroup communication and out-of-group communication. In-group communication is communication between the group leader and group members, and out-of-group communication is communication between the group leader and the base station.

In the present invention, since the nodes are grouped based on the location of the nodes, intra-group communication is possible with low power, thereby reducing power consumption and minimizing interference between groups.

Because out-of-group communication may interfere with intra-group communication of a nearby group that occurs during the same time slot, in order to minimize this, two groups are grouped as far apart as possible during the same time slot so that one group conducts intra-group communication and the other group performs intra-group communication. By allowing communication, the influence of interference is minimized.

According to the protocol for packet transmission on the network of the present invention described above, all nodes are classified by group and the group head of each group is determined. The role of the group leader is to receive packets from group members that are nodes other than the group leader in the group and deliver them to the base station. There is at least one group head within a group, and the maximum is when all nodes in the group are group heads.

Group members can be set so that packets can be sent to the base station through two-hop communication by being distributed as evenly as possible to the group heads. In the case of a plurality of group heads, if group members are assigned to a plurality of group heads based on the location of the group head and group members, the number of group members who have to contend for packet transmission within the same time slot to the group head is reduced, thereby reducing the probability of collision. . On the other hand, since the plurality of group heads must compete in the same time slot to transmit packets to the base station, the probability of collision increases as the number of group heads increases. Therefore, it is necessary to select the appropriate number of group heads according to the target value of the network environment and application. By replacing this problem with an optimization problem and solving it in the manner described above, the optimal number of group heads maximizing the network lifetime can be determined according to the type of application and the network environment.

Claims (1)

A method for setting a network protocol for a large-scale sensor node network, the method comprising:
setting a plurality of groups to which the large-scale sensor node belongs and group information of each group;
setting a slot frame and a plurality of time slots in the slot frame;
setting a group member and group leader to perform communication in each time slot; and
A method that includes optimizing the number of group heads for each group.

Images