KR20090059291A - Method and system controlling pwoer of sense network according to event - Google Patents

Abstract

A method and a system controlling the power of a sense network according to an event are provided to increase the power efficiency of a sensor network by controlling the operation of sensor node according to the events classified by an application layer. An event classification unit(101) analyzes the events occurring on an application layer to classify the events, and the events are classified to at least one category of an event type, a data size and a data duplication and a real-time characteristic. An operation setting unit(102) sets up the operations of a sensor node that senses the event according to the classified event. A node control unit(103) controls the sensor node based on the setup operation, and manages the power of the sensor network related to the sensor node by controlling the active and sleep modes of the sensor node according to the reporting time of the predicted event.

Description

Sensor network power management method and sensor network power management system according to an event {METHOD AND SYSTEM CONTROLLING PWOER OF SENSE NETWORK ACCORDING TO EVENT}

The present invention relates to a sensor network power management method and a sensor network power management system according to an event that increases the power efficiency of a sensor network by classifying events occurring in an application layer and controlling the operation (protocol) of sensor nodes according to the classification. .

The present invention is derived from the research conducted as part of the IT growth engine technology development project of the Ministry of Information and Communication and the Ministry of Information and Communication Research and Development. Development ] .

Sensor networks have periods of use ranging from months to years, depending on the state of the application layer. Sensor nodes capable of communication and computation included in the sensor network can form ad-hoc networks and collect and process data using their wireless communication functions.

These sensor nodes generally use a limited source of energy, such as batteries, and energy consumption is mostly associated with inter-node communication.

Traditional sensor network power management methods have been implemented with a focus on network protocols, which operate independently of other layers such as the application layer.

Therefore, the existing sensor network power management method using only limited information that does not include information of other layers has a limitation in increasing energy consumption efficiency.

The present invention has been made to improve the prior art as described above, by classifying events occurring in the application layer and controlling the operation (protocol) of the sensor node according to the classification, the sensor according to the event to increase the power efficiency of the sensor network An object of the present invention is to provide a network power management method and a sensor network power management system.

It is also an object of the present invention to provide a sensor network power management method and a sensor network power management system according to an event that can reduce unnecessary operation of sensor nodes of a sensor network and increase transmission efficiency.

Sensor network power management method according to the event to achieve the above object is to classify the type of the event by analyzing the event occurred in the application layer, setting the operation (protocol) of the sensor node for detecting the event according to the classified type And controlling the sensor node with a set operation.

In addition, as a technical configuration of the present invention for achieving the above object, the event classification means for classifying the types of events by analyzing the events occurring in the application layer, the operation (protocol) of the sensor node for detecting the events according to the classified type Operation setting means for setting a value; And node control means for controlling the sensor node in a set operation.

According to the present invention, a sensor network power management method and a sensor network power management system according to an event that increases the power efficiency of a sensor network by classifying events occurring in an application layer and controlling the operation (protocol) of the sensor nodes according to the classification. Can provide.

In addition, according to the present invention, it is possible to provide a sensor network power management method and a sensor network power management system according to an event that can reduce unnecessary operation of sensor nodes of a sensor network and increase transmission efficiency.

Hereinafter, a sensor network power management method and a sensor network power management system according to an event according to the present invention will be described in detail with reference to the accompanying drawings.

An event often referred to herein may be defined as the occurrence of data that should be sent to the sink.

1 is a configuration diagram of a sensor network power management system according to an inventor's event.

Referring to FIG. 1, the sensor network power management system 100 according to an event includes an event classification means 101, an operation setting means 102, and a node control means 103.

Event classification means 101 performs a function of classifying the event by analyzing the event occurred in the application layer.

The event classification means 101 may classify an event occurring at the application layer according to at least one of an event type, data size, data redundancy, or real-time characteristics.

The event type is a factor used to classify events into scheduled and non-scheduled events. A scheduled event means that the event occurs at a fixed time, and is a periodic event that occurs periodically at various intervals, such as hourly, daily, weekly, or monthly, and is scheduled on a specific day / hour. It can be divided into non-periodic events. Unscheduled events are cases in which an event is unknown at which point it will occur. There are two types of events: source-initiated and sink-initiated. An event caused by a source occurs when an asynchronous event occurs in the source, and an event caused by a sink requests a report to a specific or unspecified source through a query or the like.

The nature of the data is a factor used to classify events using two factors: the size of the data and the data redundancy. The characteristics of the data can be classified into a burst mode requiring a large amount of data and a single mode requiring a small amount of data. This classification depends on the type of data required by the application layer of the sensor network, and if it requires simple numerical information such as temperature or humidity for the area, it is classified as single mode and if it requires information such as video, voice, etc. Classified as burst mode.

Redundancy of data refers to whether data to be reported at the time of an event has redundancy, a factor used to classify events in relation to locality. Here, locality can be divided into temporal locality and spatial locality.

First, temporal locality refers to an attribute that can omit or abbreviate the reporting of current data that is not significantly different from previous data in subsequent reporting of events.

Spatial locality refers to sharing events with neighboring nodes that are close in distance so that reporting can be omitted, except for one of the nodes that detected the event.

That is, it is possible to classify events according to whether the application layer has such locality, and this can determine whether data aggregation of the sensor network is performed.

Real-time characteristics are factors used to classify events into non real-time events and real-time events according to the latency tolerance of the report.

In the case of surveillance applications that are flexible in reporting latency, they can be divided into non-real-time types.

On the other hand, if a report needs to be completed within a limited time such as a fire alarm, it may be a real-time type. The real-time nature of these events can affect the duty cycle of the sensor network MAC protocol because it affects the active time of the sensor to detect the event.

In addition, there are many factors that classify sensor network events, but the factors mentioned above have major influences on the operation and power management of routing and MAC protocols.

TABLE 1

Table 1 shows the classification factors of the events and their notation. Using this notation, an event can be represented as a bundle of {event type, data size, data redundancy, and real-time characteristics}.

The operation setting means 102 performs a function of setting an operation of a sensor node that detects an event according to the classified event.

The operation setting means 102 may set the operation of the sensor node to at least one of a power management operation according to a schedule, a power management operation by a source, or a power management operation by a sink.

The node control means 103 performs a function of controlling the sensor node by the set operation.

Specifically, the node control means 103 may control the activity or dormancy of the sensor node according to the reporting time point of the scheduled event, thereby managing the power of the sensor network associated with the sensor node.

In addition, the node control means 103 receives the activity or dormant signal transmitted by the source including the sensor node that has detected the event and controls the operation of the sensor nodes on the transmission path, thereby powering up the sensor network associated with the sensor node. Can manage

In addition, the node control unit 103 may manage the power of the sensor network associated with the sensor node by receiving the activity or sleep signal transmitted by the sink transmitting the query and controlling the operation of the sensor nodes on the transmission path.

Hereinafter, an embodiment of a sensor network power management system according to an inventor's event will be described with reference to FIG. 2.

TABLE 2

Table 2 categorizes events that occur at known major sensor network application layers using classification factors.

For example, a fire alarm can be viewed as an unscheduled event type, single mode, or real time type. Of course, a fire alarm might include a scheduled event type that periodically reports local temperature information, depending on the situation.

As such, events occurring in the application layer can be viewed as a collection of several event types that have different data characteristics and real-time characteristics for each event type, rather than having only one event type, data characteristics, and real-time characteristics.

The state of the application layer depends on the type of events that the application layer handles.

FIG. 2 is a diagram illustrating a state transition according to an event type of an application layer of a sensor network including a sensor network power management system according to the present invention.

In the node deployment phase, the state of the application starts at the initialization state 206 and after the initialization process is completed, the application layer enters the standby state 205. Depending on the cause of the event, the application layer transitions to the relevant state. At this time, a new event may occur when one event is being processed, and in this case, each event is processed sequentially. In other words, after the first event is completed, the application layer returns to the waiting state 205 and immediately changes to processing the second event.

The state of the application layer depends on the type of events that occurred in the application layer.

The protocol of the sensor network must be tailored to the performance and energy requirements of the application during the network design phase. This process is called protocol customization. The network protocol is set in consideration of the characteristics of the application layer as well as the network size, data size, and maximum allowable time of each event type.

Events 204 sensed by the sensor node are classified into bundles of {event type, data size, data redundancy, real-time characteristics} as shown in Table 2. The type of traffic transmitted in the network varies depending on the event, and the packet transmission of the node and the energy consumption pattern accordingly vary. Depending on the event type, the power management method of the node needs to be changed.

The events occurring in the sensor network are classified into three types as described above, and the sensor node that detects the event controls the operation of the node by applying a power management scheme according to the type of the classified event. For example, the detailed operation of some power management schemes is described below.

1. Scheduled Event (201)

The sensor network power management system according to the inventor's event knows the reporting period of the scheduled event 201, so that the activity and non-work schedule of the node can be adjusted accordingly. This is called scheduled-driven power control. The sensor network power management system according to the event knows the scheduled time of occurrence of both periodic and non-periodic events in advance, so power management can be applied by applying this method.

In the scheduled event type 201, the node does not need to operate other than the time of performing the report, so the sensor network power management system according to the event causes the node to be active only at a preset reporting time.

If the application layer of the sensor network has only a predetermined event type, the sensor network power management system according to the event can eliminate energy consumption by periodic activity intervals through this method.

In this case, the sensor network power management system must ensure that all nodes present in the transmission path are in an active state in addition to the node reporting the scheduled event 201. When multiple nodes report, it is necessary to adjust the activity schedule of the node in consideration of the worst transmission delay.

If all nodes in the network are global events that perform reporting, the sensor network power management system may allow each node to start transmitting packets at the same time. This is called parallel transmission. This management method can increase the amount of transmission in the entire network, but the transmission delay occurs due to contention for channel acquisition.

In contrast, the sensor network power management system may perform reporting after each node waits for all its child nodes to finish reporting in the routing tree.

The sensor network power management system manages the information about the child nodes of each node in a table form. By displaying the completed child nodes in this table, all the child nodes check whether they have completed the report.

The sensor network power management system allows all child nodes to perform their own reporting after completing the reporting, so that reporting is performed sequentially from the end nodes located at the bottom of the routing tree. This method is called global coordination and is a method of controlling the transmission order in the sensor network.

In this case, the sensor network power management system can reduce the total number of transmissions because each node can transmit the report of its child nodes and its own report in a single message. In addition, in the case of collectable reporting, the sensor network power management system may reduce the size of the reported message.

2. Unscheduled Incident by Source (202)

The unscheduled event 202 by the source reports the sensed data at the source node so that each node cannot predict the occurrence of the event. In addition, the sensor network power management system does not know how many source nodes are reporting.

Thus, the sensor network power management system cannot use global coordination and scheduled power control techniques. The sensor network power management system cannot predict the occurrence of an event, allowing each node to determine whether an event occurs while reducing energy consumption through periodic activity and sleep.

In the event of an event by the source, the sensor network power management system may apply source-driven power control by the source. This method informs all nodes on the transmission path destined for the sink by the node that has detected the unscheduled event 202 through the wakeup signal, and the message transmission is completed through the sleep signal. This is a way to inform.

The sensor network power management system ensures that nodes that receive an activity signal are in a fully wakeup state that does not have a sleep state while delivering a message. This approach, taken by the sensor network power management system, allows for short message transmission times and large intervals of activity during standby when the data reported is large. The sensor network power management system allows the reported nodes to send dormant signals, allowing the nodes on the transmission path to return to standby.

In this case, when a plurality of nodes detects an unscheduled event 202 by a source, the plurality of nodes may report. In addition, nodes that occur at the same time and are simultaneously included in the reporting path of the two events may receive multiple activity signals.

As such, when a node receives a number of activity signals, the sensor network power management system receives reports from other source nodes if the sensor network power management system changes the node state in response to sleep signals sent from a particular source node. You will not be able to.

Thus, the sensor network power management system may allow each node to remain active until it has received a dormant signal the same number of times that it has received the activity signal.

The activity and dormant signals are sent only to the nodes on the transmission path. The RTS packet used in the IEEE 802.11 DCF or S-MAC protocol has an address that can designate the next node on the transmission path and can be used as an activity signal. . In other words, the node specified in the RTS packet belongs to the transmission path and thus operates in full activity after receiving the first RTS packet. In addition, a fragment flag included in a general data packet header may be used to indicate that the last packet of the message is used as a dormant signal.

3. Unscheduled incident by sink (203)

The unscheduled event 203 by the sink is reported by a query and, as with the unscheduled event by the source, each node cannot predict when the event will occur, so the sensor network power management system is responsible for periodic activity. Sleep helps to determine if an event occurs while reducing energy consumption.

In such a case, the method taken by the sensor network power management system for effective data transfer is sink-driven power control. This method differs only in that the source is controlled by the source in the source control method, and the sink transmits activity and sleep signals.

That is, the sink node informs the sensor nodes of the occurrence of an event and the completion of the report.

In the event of an unscheduled event 203 by the sink, the sensor network power management system may allow the query itself to be used as an activity signal and the nodes receiving the query to remain in full activity.

On the other hand, when a report for the sink is completed from the network, the sensor network power management system performs a sleep signal transmission. At this time, unlike the event 202 caused by the source, the completion of reporting may be confirmed by the sink node. Unlike the power control method by the source, the sensor network power management system allows the sink node to send a sleep signal to each sensor node to notify completion of the report.

Therefore, according to the present invention, it is possible to provide a sensor network power management system according to an event that increases the power efficiency of the sensor network by classifying events occurring in the application layer and controlling the operation (protocol) of the sensor nodes according to the classification. .

3 is a flowchart of a method for managing a power supply network of a sensor according to an inventor's event.

In step 301, the event classification means of the sensor network power management system according to the inventor event analyzes the event that occurred and classifies the type of the event. In this step, the event classification means classifies events occurring in the application layer by type according to event type, data size, data redundancy, and real-time characteristics.

In step 302, the operation setting means of the sensor network power management system according to the inventor's event sets the operation of the sensor node detecting the event according to the classified type. In this step, the operation setting means sets the operation of the sensor node that detects an event according to a classified type to at least one of a power management operation, a power management operation by a source, and a power management operation by a sink.

In step 303, the node control means of the sensor network power management system according to the inventor event controls the sensor node in a set operation. This step is a process in which the node control means controls the sensor node.

If the set operation is associated with power management, step 304 may control activity and sleep of the sensor node according to a reporting time point of a scheduled event, thereby managing power of the sensor network.

In addition, in step 304, when the set operation is associated with power management, the node control means receives an activity and a dormant signal transmitted by a source that detects an event, and controls the operation of sensor nodes on a transmission path to power up the sensor network. It can also be managed.

Also, in step 304, when the set operation is associated with power management, the node control means receives the activity and dormancy signals transmitted by the sink transmitting the query and controls the operation of the sensor nodes on the transmission path to power on the sensor network. It can also be managed.

Therefore, according to the present invention, by classifying the events occurring in the application layer and controlling the operation (protocol) of the sensor nodes according to the classification, it is possible to provide a sensor network power management method according to the event that increases the power efficiency of the sensor network. .

According to the present invention, by classifying events occurring in the application layer and controlling the operation (protocol) of the sensor node according to the classification, the power efficiency of the sensor network is increased.

In addition, according to the present invention, it is possible to reduce unnecessary operation of the sensor nodes of the sensor network and to increase transmission efficiency.

While specific embodiments of the present invention have been described so far, various modifications are possible without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined not only by the claims below, but also by those equivalent to the claims.

1 is a configuration diagram of a sensor network power management system according to an inventor's event.

FIG. 2 is a diagram illustrating a state transition according to an event type of an application layer of a sensor network including a sensor network power management system according to the present invention.

3 is a flowchart of a method for managing a power supply network of a sensor according to an inventor's event.

<Explanation of symbols for the main parts of the drawings>

100: power management system

101: means of classifying a case

102: operation setting means

103: node control means

Claims (12)

Classifying the event by analyzing the event that occurred in the application layer; Setting an action of the sensor node to detect the event according to the categorized event; And Controlling the sensor node with the set operation Power management method in a sensor network comprising a. The method of claim 1, The step of classifying events, Classifying the event according to at least one of event type, data size, data redundancy or real-time characteristics Power management method in a sensor network comprising a. The method of claim 1, Setting an operation of the sensor node, Setting an operation of the sensor node to at least one of a power management operation according to a schedule, a power management operation by a source, or a power management operation by a sink. Power management method in a sensor network comprising a. The method of claim 1, The step of controlling the sensor node, Managing power of a sensor network associated with the sensor node by controlling activity or sleep of the sensor node according to a reporting time point of a scheduled event Power management method in a sensor network comprising a. The method of claim 1, The step of controlling the sensor node, Managing the power of a sensor network associated with the sensor node by receiving an activity or sleep signal transmitted by a source including the sensor node detecting an event and controlling the operation of the sensor nodes on a transmission path; Power management method in a sensor network comprising a. The method of claim 1, The step of controlling the sensor node, Managing the power of the sensor network associated with the sensor node by receiving an activity or sleep signal transmitted by the sink transmitting the query and controlling the operation of the sensor nodes on the transmission path; Power management method in a sensor network comprising a. Event classification means for distinguishing the events by analyzing the events occurring at the application layer; Operation setting means for determining an operation of a sensor node sensing the event according to the distinguished event; And Node control means for controlling the sensor node with the determined operation Power management system in the sensor network comprising a. The method of claim 7, wherein The event classification means, Power management system in a sensor network, characterized in that the event is distinguished according to at least one of event type, data size, data redundancy, or real-time characteristics. The method of claim 7, wherein The operation setting means, And determining an operation of the sensor node by at least one of a power management operation according to a schedule, a power management operation by a source, and a power management operation by a sink. The method of claim 7, wherein The node control means, Power management of the sensor network associated with the sensor node by controlling activity or dormancy of the sensor node according to a reporting time point of a scheduled event. The method of claim 1, The node control means, In the sensor network, characterized in that to control the operation of the sensor node associated with the sensor node by receiving the activity or sleep signal transmitted from the source including the sensor node that detected the event and controlling the operation of the sensor node on the transmission path Power management system. The method of claim 1, The node control means, The power management system of the sensor network, characterized in that for receiving the activity or dormant signal transmitted by the sink transmitting the query and controlling the operation of the sensor nodes on the transmission path, power management of the sensor network associated with the sensor node.

Images