KR101078348B1 - Apparatus and method for storing data using R-tree - Google Patents

Abstract

Disclosed are a data storage device and method using an area tree. The data storage device may include a sensing unit for acquiring data in a predetermined region, a data region for storing data obtained by the sensing unit, and an index of an area tree having the plurality of blocks included in the data region as a terminal node. And a controller configured to store the data acquired by the sensing unit by using the index region, sensing position information of the data obtained by the sensing unit, and the region tree.

Description

Apparatus and method for storing data using R-tree}

Embodiments of the present invention relate to sensing data storage and retrieval techniques in a mobile environment.

Mobile or stationary robots can be useful when exploring unknown areas or scouting disaster or military areas. The robots collect various types of information in a corresponding area according to the type of sensor they have, and transmit a sensed information by forming a network such as a sensor network with surrounding robots.

However, devices such as robots generally have limited energy and data storage spaces such as usable power sources. Therefore, there is a need for a method for storing and indexing sensing data acquired in real time by a data sensing device such as a robot having insufficient storage space in a limited storage space.

Embodiments of the present invention provide a method for efficiently storing and indexing data sensed in real time using an area tree, thereby providing a means for effectively utilizing data storage space and coping with spatiotemporal query processing. .

According to an aspect of the present invention, there is provided a data storage device including: a sensing unit configured to acquire data in a predetermined area; A data area including a plurality of blocks and storing data acquired by the sensing unit; An index area including an index of an area tree type having the plurality of blocks included in the data area as a terminal node; And selecting a block to store the sensed data from among the blocks in the data area by using the sensing position information of the data acquired by the sensing unit and the region tree, and storing the data obtained by the sensing unit in the selected block. It includes a control unit.

On the other hand, the data storage method according to an embodiment of the present invention for solving the above problems, in the data storage device, the step of acquiring data for a predetermined predetermined area; Selecting, at the data storage device, a terminal node to store the sensed data from among the terminal nodes of the region tree by using the sensing position information and the region tree of the acquired data; And storing the obtained data in a record indicated by a current data pointer of the selected terminal node in the data storage device.

Other aspects, features, and advantages other than the foregoing will be apparent from the following drawings, claims, and detailed description of the invention.

Embodiments of the present invention can be effectively used for data storage and storage space management for mobile devices and indexing data for space-time query processing by storing and indexing sensing data using an area tree.

Hereinafter, specific embodiments of the present invention will be described with reference to the drawings. However, this is only an example and the present invention is not limited thereto.

In describing the present invention, when it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted. The following terms are defined in consideration of the functions of the present invention, and may be changed according to the intention or custom of the user, the operator, and the like. Therefore, the definition should be based on the contents throughout this specification.

The technical spirit of the present invention is determined by the claims, and the following embodiments are merely means for efficiently explaining the technical spirit of the present invention to those skilled in the art.

1 is a diagram illustrating a configuration of a data storage device 100 using an area tree according to an embodiment of the present invention.

As shown, the data storage device 100 using an area tree according to an embodiment of the present invention includes a sensing unit 102, a data storage unit 104, and a control unit 106.

The sensing unit 102 obtains information or data about an area where the data storage device 100 is located. The data acquired by the sensing unit 102 may be, for example, a temperature, a humidity, an illuminance, etc. of a corresponding area, which may vary depending on the type of sensor provided in the data storage device 100.

The data storage unit 104 stores the data obtained by the sensing unit 102. The data storage unit 104 may be configured as a data storage device or a medium such as a magnetic disk or a semiconductor memory, and is internally divided into a plurality of blocks. The size of the block may vary depending on the type of data storage device or media constituting the data storage unit 104, but generally, the block may have a size of 1024 bytes, 2048 bytes, 4096 bytes, or the like. In 104), each block has the same size.

The data storage unit 104 is further divided into a data area 108 and an index area 110. The data area 108 is an area where data acquired by the sensing unit 102 is actually stored, and the index area 110 is an area for storing data in the data area 108 and retrieving the stored data as needed. It is configured to index data using an area tree.

The control unit 106 controls the data area 108 and the index area 110 in the data storage unit 104 to store data obtained by the sensing unit 104 or the data area 108 using the index area 110. Retrieve data stored in). In detail, when the sensing unit 102 senses data, the controller 106 selects a terminal node to insert the data by traversing the area tree using the data acquisition position information of the sensed data. The sensed data is stored in the terminal node.

Area of tree  Configuration and features

As described above, in the embodiment of the present invention, the control unit 106 stores and manages the sensing data using the area tree. Hereinafter, the configuration and features of the area tree will be described.

2 is a diagram illustrating an example of an area for sensing data by the data storage device 100 according to an exemplary embodiment of the present invention, and FIG. 3 is a diagram illustrating an area in which the data storage device 100 moves as shown in FIG. 2. It is a figure which shows an example of the formed area tree.

The area tree in the embodiment of the present invention is a tree in which a conventional area tree (R-tree) used for a multidimensional data index or the like is expanded into a storage space efficient structure. The difference between the area tree in the embodiment of the present invention and the conventional area tree is that in the case of the area tree, when the space utilization of the terminal node exceeds the threshold, the terminal node space is reused without dividing the terminal node. That is, in the embodiment of the present invention, the area tree sequentially stores the data sensed by the sensing unit 102 in the terminal node on the basis of the sensing time (timestamp). If an overflow occurs, the region tree is divided. If necessary, the oldest data is overwritten without partitioning if the size of the partitioned area is less than or equal to the threshold.

Each node in the region tree and the data to be indexed are represented by a minimum bounding region (MBR) or a minimum bounding box (MBB) (hereinafter collectively referred to as MBR), and the region tree is represented by a containment relationship between the MBR or MBB.

In FIG. 2, numbers 1 to 7 are numbers of sensing regions numbered in the order of sensing, and each rectangle is an MBR of each node of the region tree shown in FIG. Nodes of the area tree are classified into non-terminal nodes and terminal nodes according to the presence or absence of child nodes. The non-terminal node is also referred to as an internal node where a child node exists and corresponds to R, a, and b nodes in FIG. 3. The non-terminal node stores the minimum rectangle including the MBR of the child nodes and the address of the child node. Node R among the non-terminal nodes is a root node, and an area indicated by the root node becomes all of the sensing area. Meanwhile, the terminal node is a node in which a child node does not exist and data is directly stored, and nodes 108-1 to 108-7 correspond to this in FIG. 3. The terminal node includes sensing data and a current data pointer. The current data pointer is a pointer indicating a location where new data is to be stored in the terminal node.

The level of the area tree is configured such that the level increases with the root level as 1 toward the child node. That is, in the area tree of FIG. 3, the level of node R, which is the root node, is 1, the level of nodes a, b is 2, and the level of nodes 108-1 to 108-7, which are terminal nodes, is 3. The height H of the area tree is equal to the level of the terminal node.

개의 엔트리를 포함해야 한다. 데이터 저장부(104)의 블록의 크기를 B라 하고, 상기 영역트리의 엔트리 크기를 S라 하면 상기 M 값은 다음의 수학식 1을 이용하여 계산될 수 있다.At least two entries must exist in the root node of the area tree, and all terminal nodes are configured to exist at the same level. In addition, all nodes except the root node of the region tree may include a maximum of M entries, and internal nodes except the root node and the terminal node are at least

Must contain three entries. If the block size of the data storage unit 104 is B and the entry size of the area tree is S, the M value may be calculated using Equation 1 below.

Storage Space Division of Data Storage Unit 104

In an embodiment of the present invention, the data storage unit 104 is divided into a data region 108 and an index region 110, and non-terminal nodes (internal nodes) of the region tree are stored in the index region 110. Are stored in the data area 108, respectively. Hereinafter, the division of the storage space of the data storage unit 104 to efficiently use the data storage unit 104 and maximize the data storage capacity will be described.

When the total number of blocks of the data storage unit 104 is N _B , the number of internal node blocks of the area tree stored in the index area 110 is I _B , and the terminal node blocks of the area tree stored in the data area 108 are represented. When the number (data block number) is L _B , the relationship between N _B , I _B , and L _B is expressed by Equation 2 below. It is assumed here that the node size (bytes) and the block size (bytes) are the same. Thus, the number of nodes and the number of blocks mean the same thing.

이므로, 상기 영역트리의 최소 팬아웃은

이 되며, 이 경우 영역트리의 최대 높이는 다음의 수학식 3과 같이 계산된다.The height of the area tree is maximum when the fanout of each node is minimum, that is, when the number of entries of each internal node is minimum. The minimum number of entries for each internal node in the region tree is

Since the minimum fanout of the area tree is

In this case, the maximum height of the area tree is calculated as in Equation 3 below.

)^H-1이므로, (

)^H-1 ≤ L_B인 최대 H를 구한 것이다.In the above equation, when the height of the area tree is H (H≥1), the minimum number of terminal nodes is (

) ^H-1, so (

Is the maximum H where ^H-1 ≤ L _B.

On the other hand, when the number of blocks of the terminal node is L _B and the height of the area tree is H, the maximum number of blocks I _B of the internal node has a value in the following range.

If the block number I _B of the internal node having the range value of Equation 4 is correctly calculated, it is as follows.

First, at level L = H-1, the number of internal nodes I _{H -1} is

to be. This level H - because the number of entries to be L _B are actually stored in the internal nodes 1. In a similar way, the number of internal nodes I _{H -2} at level L = H-2 is

Becomes

Using this, the equation for calculating the maximum number of blocks of internal nodes at each level of the region tree is expressed as a recurrence relation as follows.

Generalizing this, the number I _L of internal nodes at the level L (1 = 1 = H-1) of the region tree is expressed by Equation 5 below.

The number of internal nodes of each level of the area tree can be known by Equation 5 above, and the sum of all of the internal nodes (number of blocks) of the area tree can be known as follows.

The value of I _{H -1} in a recursive relationship of the equation (5) is is determined by L _B, L _B Can be substituted with the formula of N _B and I _B by Equation 2 (that is, L _B = N _B -I _B ). Therefore, if the total number of blocks of the data storage unit 104 is given by N _B , the number of internal node blocks I _B and data of the area tree of the area tree stored in the index area 110 by Equation 2, Equation 5 and Equation 6 are given. Since the number of terminal node blocks (number of data blocks) L _B of the area tree stored in the area 108 can be obtained, the limited space can be efficiently utilized.

Data file structure

4 is a diagram showing an example of a data storage record in the data area 108 according to an embodiment of the present invention.

Each block constituting the data area 108 is divided into a plurality of records. The record is a basic unit in which data sensed by the sensing unit 102 is stored. In the illustrated embodiment, the record 400 consists of 24 bytes in total length, the first 4 bytes 402 is the data acquisition time (timestamp), and the next 8 bytes 404, 406 are the data acquisition coordinates (x, y). ), The remaining 12 bytes (408, 410, 412) are composed of the sensing data. In the present embodiment, the temperature, illuminance, and humidity information of the data storage device 100 at the sensed location are stored by 4 bytes in the 12 bytes (408, 410, 412), respectively.

Insert new record

5 is a flowchart illustrating a record insertion method 500 in a data storage device 100 using an area tree according to an embodiment of the present invention.

When new data is sensed by the sensing unit 102, the control unit 106 first traverses the region tree from the root node and selects a terminal node into which the newly sensed data is to be inserted (502).

The newly acquired data has a terminal node to be stored according to a data acquisition position, that is, x and y coordinate values of the area where the data is sensed. For example, if there is a node including the data sensing position among the nodes of the region tree, the data may be inserted into the node. Alternatively, when there is no node including the data acquisition location among the nodes of the area tree, a node having the smallest area extension when including the data sensing location may be selected as a terminal node to store the newly sensed data.

When the terminal node is selected, it is determined whether there is a space for inserting a new record in the next selected terminal node (504). In this case, if there is a space for inserting a new record in the selected terminal node, the sensed data is inserted in the empty space (506). Since each terminal node includes a current data pointer as described above, newly acquired data may be inserted into a storage location indicated by the current data pointer. After the data is inserted, since the storage location of the last data in the block is changed, the current data pointer is updated to point to the record in which newly acquired data is stored (508). If the sensed data is the last record of the terminal node, the current data pointer is updated to point to the first record of the terminal node.

Thereafter, it is determined whether the MBR of each node of the region tree needs to be adjusted according to the storage position of the inserted record (510), and if necessary, the MBR of each node is adjusted (512). In the case where it is necessary to adjust the MBR by the inserted record, there is no node including the sensing position of the sensed data among the nodes of the region tree, so that the MBR must be extended by the inserted record. Such adjustment of the MBR may be performed starting from the terminal node to the root node.

On the other hand, if there is no space to insert a new record in the selected terminal node in step 504, it is next determined whether or not the selected terminal node can be divided (514). If it is possible to divide the selected terminal node, a new terminal node is created in the data area 108 (516), and the record stored in the selected terminal node and the newly sensed data are distributed to the newly generated terminal node ( 518). Thereafter, the MBR of the selected terminal node and the ancestor node of the generated terminal node is adjusted to reflect the change of the MBR of the terminal nodes (520).

If the selected terminal node is not partitionable in step 514, the sensed data is overwritten with the oldest record among the records stored in the selected terminal node (522). That is, in the embodiment of the present invention, when the terminal node no longer has space to store data and the terminal node cannot be divided, the record allocated to the terminal node is cyclically reused.

There are two cases in which the selected terminal node is not partitionable.

(1) When the size of the MBR region of the selected terminal node is less than or equal to a predetermined value. That is, the size of the MBR region of the selected terminal node is too small to divide.

(2) When the height of the area tree becomes H + 1 after splitting of the terminal node (where H is the maximum height of the area tree). That is, the number of terminal nodes constituting the area tree becomes L _B , and there are no terminal node blocks to allocate.

6 to 9 are diagrams for illustrating a terminal node division and record insertion process according to an embodiment of the present invention.

First, assuming that the data storage device 100 senses data while the data storage device 100 moves within the area as shown in FIG. 6, the numbers shown in the rectangles of FIG. Is the number of. FIG. 7 is an area tree generated by the data storage device 100 while moving the area of FIG. 6.

If the data storage device 100 senses data at the fifteenth position shown in the circle in FIG. 6, the controller 106 selects a terminal node (108-15 in FIG. 9) to store the sensing data of the fifteenth position. Allocate the sensing data. Since the fifteenth position belongs to the region c, the MBR of the fifteenth position must be inserted into the node c, which is a non-terminal node of the region tree. However, as shown in FIG. 6, the node c has no space to insert data anymore. Therefore, partition node c into nodes c and e, and divide the existing data stored in node c into nodes c and e according to the location of MBR. Similarly, since the root node of FIG. 6 also overflows according to the division of the node c, the root node of FIG. 6 is divided into nodes F and G, and a new root node is created to point to the divided nodes F and G. 8 and 9 illustrate region division and region tree when the sensing data of the fifteenth region is inserted in this manner.

Deletion of records

As described above, in the embodiment of the present invention, the data storage device 100 does not have a space for inserting a new record in the terminal node to store the sensing data, and when the terminal node is not partitionable, the data storage device 100 is the most among the records of the terminal node. Configured to overwrite the sensing data with an old record. That is, the existing data is replaced by new data over time, so there is no separate data deletion routine in the embodiment of the present invention. Therefore, according to an exemplary embodiment of the present invention, since a separate operation and energy consumption are not required to delete data, energy can be efficiently used.

Retrieval of data

10 is a flowchart illustrating a data retrieval method 1000 according to an embodiment of the present invention.

First, an area query to be searched is generated (1002). The area query may be a query for searching for data of a specific time in a specific area among the areas searched by the data storage device 100.

Next, the area tree is searched from the root node using the coordinate values (x, y values) of the area query to find a terminal node including the query area (1004).

Finally, it searches for whether the record to be found in the region query exists by sequentially searching the records stored in the terminal node (1006).

Meanwhile, an embodiment of the present invention may include a computer readable recording medium including a program for performing the methods described herein on a computer. The computer-readable recording medium may include program instructions, local data files, local data structures, etc. alone or in combination. The media may be those specially designed and constructed for the purposes of the present invention, or they may be of the kind well-known and available to those skilled in the computer software arts. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical recording media such as CD-ROMs, DVDs, magnetic-optical media such as floppy disks, and ROM, RAM, flash memory, and the like. Hardware devices specifically configured to store and execute program instructions are included. Examples of program instructions may include high-level language code that can be executed by a computer using an interpreter as well as machine code such as produced by a compiler.

Although the present invention has been described in detail with reference to exemplary embodiments above, those skilled in the art to which the present invention pertains can make various modifications to the above-described embodiments without departing from the scope of the present invention. Will understand.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined by the claims below and equivalents thereof.

1 is a diagram illustrating a configuration of a data storage device 100 using an area tree according to an embodiment of the present invention.

2 is a diagram illustrating an example of an area for sensing data by the data storage device 100 according to an exemplary embodiment of the present invention.

3 is a diagram illustrating an example of an area tree formed while the data storage device 100 moves the area shown in FIG. 2.

4 is a diagram showing an example of a data storage record in the data area 108 according to an embodiment of the present invention.

5 is a flowchart illustrating a record insertion method 500 in a data storage device 100 using an area tree according to an embodiment of the present invention.

6 to 9 are diagrams for illustrating a terminal node division and record insertion process according to an embodiment of the present invention.

10 is a flowchart illustrating a data retrieval method 1000 according to an embodiment of the present invention.

Claims (13)

A sensing unit for sensing data in a predetermined area; A data area including a plurality of blocks and storing data sensed by the sensing unit; An index area including an index of an area tree type having the plurality of blocks included in the data area as a terminal node; And Selecting a block to store the data sensed by the sensing unit from among the blocks in the data area by using the sensing position information of the data sensed by the sensing unit and the area tree, and the data sensed by the sensing unit in the selected block A control unit for storing; Data storage device comprising a. The method of claim 1, Each block of the data area includes a current data pointer indicating a location where new data is stored, and the controller stores the data sensed by the sensing unit at a location indicated by the current data pointer of the selected block. Data storage device. 3. The method of claim 2, The control unit stores the data sensed by the sensing unit and updates the current data pointer to point to a next record of the record in which the data is stored among the records in the selected block. The method of claim 3, If the data sensed by the sensing unit is stored in the last record of the selected block, the control unit updates the current data pointer such that the current data pointer points to the first record of the selected block. The method of claim 1, The maximum height of the area tree is the following equation

Where H is the maximum height of the area tree, M is the minimum number of entries of each internal node of the area tree, and L _B is the number of terminal node blocks of the area tree stored in the data area. Calculated by the data storage device. The method of claim 5, The number of internal nodes at each level of the region tree is

Where I _L is the number of internal nodes at level L (1 = 1 = H-1) of the area tree, and M is the minimum number of entries of each internal node in the area tree. Calculated by the data storage device. A data storage method of a data storage device, Sensing data on a predetermined area in the data storage device; Selecting, by the data storage device, a terminal node to store the sensed data from among the terminal nodes of the region tree by using the sensing position information and the region tree of the sensed data; In the data storage device, storing the sensed data in a record indicated by a current data pointer of the selected terminal node; Data storage method comprising a. The method of claim 7, wherein After performing the data storing step, further comprising updating, at the data storage device, the current data pointer such that the current data pointer points to the next record of the record in which the data is stored. The method of claim 8, When the sensed data is stored in the last record of the selected terminal node, the data storage device updates the current data pointer such that the current data pointer points to the first record of the selected terminal node. . The method of claim 7, wherein The data storage step, Determining whether a new record insertion space exists in the selected terminal node; Determining whether the selected terminal node is splittable when there is no new record insertion space in the selected terminal node; Generating a new terminal node and distributing a record stored in the selected terminal node and the sensed data to the generated terminal node when the selected terminal node is partitionable; Adjusting a minimum bounding region of the selected terminal node and ancestor nodes of the generated terminal node; Comprising a data storage method. The method of claim 10, And when the selected terminal node is not partitionable, overwrite the sensed data on the oldest record among the records stored in the selected terminal node. The method of claim 11, Whether or not the selected terminal node is split may include whether or not the size of the minimum limit area of the selected terminal area is equal to or less than a predetermined value or the height of the area tree after division of the terminal node increases the maximum height of the area tree. The method of storing data, determined by whether or not to exceed. A computer-readable recording medium having recorded thereon a program for performing the method according to any one of claims 7 to 12 on a computer.

Images