CN102821398B - Method and system for storing sensor network data facing user multiple demands - Google Patents
Method and system for storing sensor network data facing user multiple demands Download PDFInfo
- Publication number
- CN102821398B CN102821398B CN201110153721.2A CN201110153721A CN102821398B CN 102821398 B CN102821398 B CN 102821398B CN 201110153721 A CN201110153721 A CN 201110153721A CN 102821398 B CN102821398 B CN 102821398B
- Authority
- CN
- China
- Prior art keywords
- data
- user
- sensor network
- node
- event
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Abstract
The invention provides a method and a system for storing sensor network data facing user multiple demands. The method for storing sensor network data includes a data storage mode based on data serving as a center is implemented inside a sensor network; and data storage is performed outside the sensor network by using a Sink node, wherein the data storage mode based on data serving as a center also performs a step of building a multi-granularity data storage strategy for a user through an incident rule mechanism before the step of data storage, when the Sink node performs data storage, the Sink node analyzes user inquiry requests to inquire data needed by the user, if the Sink node stores the inquiry data of the user, an inquiry result is returned to the user; and if not, the inquiry requests are sent to the Sink node to perform intelligent analysis to obtain key values of the incident needed by the user, and a data report function of a corresponding storage node is started according to the key value, the data needed by the user are uploaded to the Sink node for storage.
Description
Technical field
The present invention relates to the data managing method of sensor network, be specifically related to sensor network data access method and the system of the many demands of a kind of user oriented.
Background technology
Wireless sensor network (Wireless Sensor Networks, WSN) be the current hot research field, forward position received much concern in the world, there is very wide application prospect, the fields such as military and national defense, industrial or agricultural control, city management, biologic medical, environment measuring, rescue and relief work, deathtrap Long-distance Control can be applied to, be considered to have an immense impact on one of the technology of power to 21 century.
Wireless sensor network forms by being deployed in cheap microsensor nodes a large amount of in monitored area, the network system of the self-organizing of the multi-hop formed by communication, its objective is perception collaboratively, the information of perceived object in acquisition and processing network's coverage area, and (may people be comprised by being positioned at network edge Sink node to user, computer or other equipment) field data information is provided, typical wireless sensor network application structure as shown in Figure 1, wherein Sink node is as the intermediary of connecting sensor network and external network (Internet/3G etc.), inwardly communicate with sensor node with sensor network communication agreement, outwards with existing various wired or wireless mode accessing Internets or other communication networks.Wireless sensor network, except the feature such as mobility, discontinuity, dynamic with traditional ad-hoc network, also has node communication and computing capability is limited, power supply energy and the limited distinguishing feature of storage capacity.Therefore, when designing the application system based on wireless sensor network, Energy design principle is one of key content determining its availability and validity.
Major consideration for wireless sensor network application comprises low power consumption node design, the Routing Protocol of efficiency realizes, efficient network data management scheme etc., and wherein network data management specifically comprises the collection of perception data, storage and these aspects of retrieval; Concerning user, the core of sensor network is perception data, namely emphasize data-centered feature, therefore, designing the data management scheme not only meeting wireless sensor network feature but also meet consumers' demand is an importance of carrying out radio sensing network application.Data management scheme in current wireless sensor application can be divided into two kinds substantially: data management scheme and the outer data management scheme of net in net.
In net, data management scheme emphasizes the collection realizing perception data in wireless sensor network inside, store and retrieval, in this scheme, node in network is by more stored therein by certain machine-processed node selected for the environmental data perceived, these nodes storing perception data are referred to as memory node (memory node self also possesses the function of image data), the inquiry of user to target data is initiated by Sink node, Sink node is broadcasted in a network or clean culture contains the inquiry packet of target data information, forward between inquiry packet node in a network, until arrive the memory node at target data place, target data is mail to Sink node and is delivered subsequently to user by memory node, so far, the collection of perception data, store and inquire about this whole process and all completed by the node cooperation in wireless sensor network.
According to the selection mechanism of memory node, in net, data management scheme can be divided into again based on this locality storage (Local-Storage Based) with based on data grid technology storage (Data-Centric-Storage Based) two kinds of embodiments.In the embodiment stored based on this locality, memory node is own node, and namely the data collected are stored in the memory space of self by node, inquires about and is realized by Sink node broadcast query bag in the whole network, in the embodiment stored based on data grid technology, sensor network inside realizes the data storage and query being similar to DHT mode, perception data in network is endowed key assignments based on event rules, pass through hash function, node in these key assignments and network sets up mapping relations, the node that so these and a certain key assignments establish mapping relations is the memory node of these key assignments data, after the node in network collects data, the memory node that data can be sent to its correspondence according to the key assignments of data stores, inquire about by Sink node in a network unicast packet realize containing the inquiry packet of target data key assignments, based on the program wireless sensor network application structure as shown in Figure 2.
These two kinds of embodiments of comparative analysis, data management scheme based on data grid technology storage is better than the data management scheme stored based on this locality and shows following two aspects: in embodiment a) based on this locality storage, because user cannot determine the memory node of target data, therefore only have in a network broadcast query bag to obtain data; And based in the embodiment stored centered by DHT mechanism data, user clearly knows the memory node of target data, therefore can obtain target data by the mode of clean culture, the data management scheme Query Cost namely stored based on data grid technology will lower than the data management scheme scheme stored based on this locality; B) by event rules mechanism, many granularity datas service function can be realized, also for user provides data, services more flexibly while reduction energy consumption, as passed through definition event rules, different stage merging related data and gives different key assignments is stored on different nodes, and namely user obtains these varigrained data by key assignments.
Net storage and inquiry that outer data management scheme then emphasizes to realize in sensor network outside perception data, in this scheme, after node in network collects data, send data to Sink node uniformly, these perception datas are sent to fixing memory device storage by sink node subsequently, the inquiry of user is sent to Sink node, and Sink node is going to memory device obtain target data and Query Result is returned to user after resolving.The outer data management embodiment of typical net be by direct for memory device carry in Sink node, due in such an implementation, the storage of data and inquiry are all the edges at sensor network, therefore, the outer data management scheme of this net is referred to as the data management storing (Edge-Storage Based) based on edge, based on the program wireless sensor network application structure as shown in Figure 3.
By above to the description of data management scheme of wireless sensor network application, be summarized as follows:
In net, data management scheme maximally utilises the limited communication of node, calculating and storage capacity,
Node achieves the collection to perception data, storage and inquiry work; DHT working method and event rules mechanism is utilized based on data management scheme in the net that data grid technology stores, reduce the communication cost of network to greatest extent, and providing many granularity datas service function for user, this is that the sensor network application carrying out the many demands of user oriented provides and supports guarantee flexibly.
Net the core content that outer data management scheme then simplifies sensor network as much as possible, by perception data
Store and inquire about and realize in sensor network outside, because the calculating of Sink node and memory device, storage capacity are unrestricted, therefore, in Sink node and memory device, can realize the service functions such as the data mining of complicated energy consumption, historical data management, this is that the sensor network application carrying out the many demands of user oriented provides larger development space.
But we are not difficult to find that they all exist some respective defects:
For data management scheme in net, because node storage capacity is limited, the memory space of node is very soon by depleted, new data will cover old data, and therefore memory node always preserves perception data that ought be interior for the previous period, and namely historical data is all lost, in many sensor network application, user is by predicting future to the analysis mining of historical data, and obviously, this data management scheme is difficult to the application scenarios being applicable to historical data be had to demand.
For the outer data management scheme of net, mainly contain the deficiency of following aspect: no matter sensor network a) combined at increasing transducer and actuator is in ageing or intrinsic demand in applying, the outer data management of this net is all not suitable for, as based in the fire alarm monitoring and measuring application of sensor network, different sensor nodes (node has temperature respectively, humidity, the transducers such as smokescope) gather temperature in monitored area incessantly, humidity, the environmental datas such as smokescope are also stored in corresponding memory point, the actuator node being connected with alert sound device then obtains the last look of these data incessantly from memory point, once the display of these data there occurs the condition of a fire, alert sound device sounds, obviously, in this type of application, data query and the minimum artificial interference of low delay just may provide more believable fire alarm Monitoring Service, and the outer data management of net is difficult to respond the demand of netting interior user in real time because its data are stored in network-external, b) because the initial data of collection is directly sent in External memory equipment by all the sensors node in network, but the data may inquired about by user in reality are by considerably less, namely this scheme will cause a large amount of redundant datas, and network node transmits the network limited energy of these redundant datas meeting consuming sensor, obviously, this is not suitable for the sensor network applied system design principle of efficiency.
In sum, in existing wireless sensor network application, mainly have employed the collection of scheme realization to perception data of data management or the outer data management of net in net, the operations such as storage and inquiry, for the user of data, services single demand, these data management scheme can meet the demand of user preferably, but, along with popularizing of sensor network application, user presents variation trend gradually to the data, services demand that sensor network provides, the demand to different grain size data may be comprised, to the demand etc. of historical data, obviously, previous data management scheme cannot adapt to the scene application under the many demands of this user data services.
Summary of the invention
The object of the invention is to, for the wireless sensor network that overcomes prior art is to the scheme adopting data management or the outer data management of net in single net in the access of image data, and tend to along with user presents variation gradually to the data, services demand that sensor network provides, the scene application under only adopting the technical scheme of a certain data store strategy cannot adapt to the many demands of this user data services.Therefore, on the one hand the present invention is to the strategy adding many granularities in net based on data-centric storage mode, and the data gathered by each sensor node are carried out the preliminary treatment of many granularities and performed data-centric storage in net again; On the other hand, intelligent interaction module is invention increases when carrying out data storage based on the outer Sink node of net, the reception storage mode of existing passive type is improved to the interactive net external memory based on user's request, namely the present invention proposes sensor network data access method and the system of the many demands of a kind of user oriented.
The present invention is directed to the scene that existing data management of sensor network scheme cannot adapt under the many demands of user data services and apply this defect, on the basis of data management and the outer data management of net in comprehensive network, propose the data management of sensor network scheme of the many demands of a kind of user oriented, the program combines in net and stores based on data grid technology and the outer data management scheme stored based on edge of net, according to the different pieces of information demand for services of user, provide data management flexibly.
For achieving the above object, the invention provides the sensor network data access method of the many demands of a kind of user oriented, the method selects the date storage method of sensor network nodes based on user's request, and comprise data storage and query method, described date storage method comprises:
Inner at sensor network, implement the data storage method based on data grid technology; Outside at sensor network, adopt Sink node to carry out data storage;
Wherein,
The described data storage method based on data grid technology also performed by event rules mechanism as user sets up the step of many granularity datas storage policy before carrying out data storing steps; Described Sink node is carried out data memory time base in the demand data of query processing to user of user and is done analyses and prediction and make corresponding response, realizes adaptive history data store;
Described data enquire method is: described Sink node resolves the inquiry request of user, and based on the requesting query user requested data after parsing, if described Sink node stores the data query of user, Query Result is returned to user by interface; Otherwise, the key assignments of required data query is sent to described sink node and carries out the key assignments that intellectual analysis process obtains event needed for user, and the result according to analyzing and processing is sent the order bag opening the memory node data reporting functions that described key assignments maps, and being issued to corresponding memory node by sensor network interface, memory node receives this order and unwraps to open its data reporting functions and start to upload its data and store to described Sink node.
In technique scheme, described many granularity datas storage policy is:
First, according to the task of user, take out different primary event, the perception data of sensor network interior joint collection is associated with the event belonging to it, and the perception data belonging to each event is given a unified key value Key; Secondly, from primary event, taking out high level event, being associated with high-level event belonging to it by memory node merging mary event data that perception data obtains, and the mary event data belonging to high-level event is given a unified key assignments key; Repeat with this, until reach highest level event; Finally; by above-mentioned event rules mechanism; perception data in network and event data are all endowed key assignments; memory node in these key assignments and network is set up mapping relations by Hash mechanism further that be similar to DHT; after node collects perception data; according to its key assignments perception data is sent on mapped memory node and stores, when memory node merges after perception data obtains event data, also according to its key assignments event data is sent on mapped memory node and stores.Finally, sensor network data is organized into a kind of many Granularity Structures, from low to high, presents the fusion process from perception data to event data.
In technique scheme, described adaptive history data store comprises following steps further:
Step 301, open or close the step of the data reporting functions of the memory node of described sensor network, during netinit, described Sink node is opened to described sensor network broadcast or closes the order bag of memory node data reporting functions, and the memory node in described sensor network is unwrapped according to this order and opened or close its data reporting functions;
Step 302, network working stage, described sensor network interface receives the packet of described sensor network, is sent in the data memory module that described sink node comprises and stores after parsing;
Step 303, the step of inquiring user demand information, user sends the solicited message of certain category information that the described sensor network of inquiry gathers, this solicited message is converted to the intelligible querying command of data memory module of described Sink node by the query processing module of described sink node, described data memory module accepts described querying command and retrieves corresponding data, and described intelligent object also receives described querying command for carrying out analyzing and processing to user data demand simultaneously;
Step 304, when not storing the data that user will inquire about in described data memory module, then inquires about failure, enters next step; Otherwise, then in data memory module described in described user feedback, deposit the data of inquiry, enter last step;
Step 305, described intelligent object analyzes described querying command, be some characteristic parameters by described command analysis, described characteristic parameter comprises the data key values information of user's inquiry request information, then according to this data key values information generate open described data key values map the order bag of the data reporting functions of memory node, described order bag is issued to the memory node of sensor network by sensor network interface, described memory node uploads the data of described user inquiry to the data memory module of described Sink node, waits for that user sends inquiry request next time.
Described characteristic parameter also comprises the enquiry frequency information of user to certain category information.
Described intelligent object also performs the step of the data reporting functions of the memory node that closedown has been opened, according to user to the enquiry frequency information of certain category information, if when a certain key assignments is not queried within a very long time, described intelligent object then perform close this data key values map the order of the data reporting functions of memory node.
Described in step 301 during netinit memory node data reporting functions default setting for opening.
User described in step 303 sends inquiry request information by Internet/3G interface.
Based on above method, the present invention also provides the sensor network data storage system of the many demands of a kind of user oriented, this system carries out the mode of data storage based on the Sink node outside the data storage method of data grid technology and net, comprise: data perception node in sensor network, event data storage node in sensor network, sink memory node, is characterized in that
In described net, sensing node comprises:
Perception data key assignments generation module, for giving the key assignments of event belonging to it by the perception data of node collection;
Sending module, for sending the perception data gathered to event data storage node in described net;
In described net, event data storage node is divided into some levels from high in the end according to the rank of stored event, in described net, event data storage node specifically comprises: unlatching or closedown arrange module, carry out the unlatching of data reporting functions or the state of closedown for arranging memory node to described Sink node;
Event data key assignments generation module, the event data for being obtained by fusion perception data gives the key assignments of high-level event belonging to it;
Sending module, for sending event data to event data storage node in described more senior net;
Receiver module, for receiving the perception data that described sensing node gathers or the event data accepting event data storage node transmission in more low-level net;
Described Sink node comprises intelligent object, for doing analyses and prediction based on the demand data of query processing to user of user and sending the order of data reporting functions of opening or closing some sensor network memory node, realize adaptive historical data management and store.
In technique scheme, described Sink node also comprises:
Data memory module, the data that the memory node that the inquiry request information for receiving user carries out data query or reception unlatching reports store;
Query processing and sending module, for described request information is converted to the intelligible querying command of described data memory module, mail to described data storage cell and described intelligent cell;
Sensor network interface module, for connecting described Sink node and described sensor network; With
Subscriber Interface Module SIM, for receiving the data inquiry request information that user sends.
Described intelligent object comprises further:
Buffer memory pretreatment unit: for intercepting each inquiry request of user, and extract relevant query metadata from described inquiry request; Described data query is obtained after filtration to some characteristic parameter of associated subscriber demand data with statistical disposition;
Analyses and prediction unit: the characteristic parameter receiving the associated subscriber demand data that described buffer memory pretreatment unit exports, and based on these parameters, adopt model analysis method, dope the different demands of user to data, and based on predicting the outcome, export the action parameter opening that memory node reporting functions;
Action executing and memory cell: receive the action parameter that described analyses and prediction unit exports, described action parameter is converted into sensor network order bag be sent to described sensor network interface and then be issued to sensor network, closes or turn-on data reporting functions with the memory node controlling to be correlated with;
Wherein, described analyses and prediction unit opens the action parameter of the reporting functions of memory node according to the key assignments information of user's inquiry request; Whether described analyses and prediction unit closes the action parameter of the reporting functions of memory node according to the judgement of user query frequency information.
The step of the data reporting functions of described closedown memory node, according to user to the enquiry frequency information of certain category information, if when a certain key assignments data are not queried within a very long time, then perform close this data key values map the order of the data reporting functions of memory node.Described user interface section comprises: Internet interface or 3G interface.
The invention has the advantages that, adopt the strategy to adding many granularities in net based on data-centric storage mode on the one hand, the data gathered by each sensor node are carried out the preliminary treatment of many granularities and are performed data-centric storage in net again, like this while meeting the varigrained demand data of user, also can reduce energy ezpenditure during convergence further; On the other hand, intelligent interaction module is invention increases when carrying out data storage based on the outer Sink node of net, the reception storage mode of existing passive type is improved to the interactive net external memory based on user's request, achieve the management to historical data flexible and efficiently, greatly improve the flexibility of user data services quality, save the memory space of Sink node simultaneously.
Accompanying drawing explanation
Fig. 1 is the typical radio sensor network application structure of prior art;
Fig. 2 is the wireless sensor network data managerial structure stored based on data grid technology of prior art;
Fig. 3 is the wireless sensor network data managerial structure stored based on edge of prior art;
Fig. 4 is the wireless sensor network data managerial structure of the many demands of user oriented of the present invention;
Fig. 5 is the many Granularity Structures of data in the Smart Home based on sensor network of the present invention application;
Fig. 6 is the functional module schematic diagram of sink node of the present invention;
Fig. 7 is intelligent object function unit figure of the present invention;
Fig. 8 is the flow chart that system of the present invention performs data storage;
Fig. 9 is the flow chart that system of the present invention performs user's inquiry.
Embodiment
Below in conjunction with drawings and the specific embodiments, the invention will be further described.
Specific embodiment of the invention scheme is: inner at sensor network, implements the data management stored based on data grid technology, by event rules Mechanism establishing many granularity datas Managed Solution; Outside at sensor network, strengthen the data managing capacity stored based on edge, Sink node increases intelligent object, to provide intelligent subscriber demand analysis and feedback function, in the present invention, this Sink node combining larger storage capacity and intellectual analysis and feedback function is called Proxy, by the storage of Proxy and intelligent subscriber demand analysis and feedback, set up adaptive history data management scheme; The wireless sensor network application structure of the many demands of user oriented that the present invention proposes as shown in Figure 4.
Based on many granularity data management of event rules in net:
In data grid technology storage scheme, the node in sensor network is identified as sensing node or memory node based on its function; All nodes in network can gather perception data, based on certain rule, they are stored in a network on a part of node, namely this part node is except possessing the function of collection perception data, also be responsible for storage self or the data from other nodes, be designated memory node, other the node gathering perception data function that only possesses then is designated sensing node.
Based in many granularity datas Managed Solution of event rules, first, according to the task of user, take out different primary event, the perception data that sensor network interior joint (comprising sensing node and memory node) gathers is associated with the event belonging to it, and the perception data belonging to each event is given a unified key value Key, secondly, from primary event, taking out high level event, being associated with high-level event belonging to it by memory node merging mary event data that perception data obtains, and the mary event data belonging to high-level event is given a unified key assignments key, repeat with this, until reach highest level event, such as based in the Smart Home application system of wireless sensor network, according to the task of user, take out primary event and have fire monitoring, to leap in through the window monitoring, to force open the door monitoring, signal light control, humidifier control etc., associate the perception data of these primary event, perception data as fire monitoring event correlation comprises temperature and smokescope, key assignments " condition of a fire " is given by these perception datas, leap in through the window and monitor the perception data involving vibrations of event correlation, perception of sound amount, these perception datas are given key assignments " broken window ", the perception data of monitoring event correlation of forcing open the door comprises a magnetic strength amount of knowing, give key assignments by these perception datas " to force open the door ", other intrusion detection event of high one-level is taken out further from leap in through the window monitoring and these two primary event of monitoring of forcing open the door, the data of associated are then the broken window event data after merging based on perception data and event data (event data generally shows as the boolean's offset judging whether a certain event occurs) of forcing open the door, key assignments " invasion " is given by these event datas, other dangerous situation of high one-level monitoring event is taken out again further from intrusion detection and these two events of fire monitoring, and be that corresponding data give key assignments " dangerous situation ", repeat with this
By event rules mechanism recited above; the event data that the perception data of node collection and fusion perception data obtain all is endowed key assignments key; memory node in these key assignments and network is set up mapping relations by Hash mechanism further that be similar to DHT; after node collects perception data; according to its key assignments perception data is sent on mapped memory node and stores; when memory node merges after perception data obtains event data, also according to its key assignments event data is sent on mapped memory node and stores.
Finally, sensor network data is organized into a kind of many Granularity Structures, from low to high, present the fusion process from perception data to event data, for in the application of above-mentioned Smart Home, provide its data organizational structure's example as shown in Figure 5, these varigrained data store on memory nodes different in a network uniformly, for Sink node provides unified access mode.
Net outer adaptive history data management:
Outside at sensor network, the present invention is based on the storage of Proxy and intelligent subscriber demand analysis and feedback function, realize adaptive history data management, the functional block diagram of Proxy as shown in Figure 6.
On the basis of the sink node of prior art, by adding intelligent object, intelligent object does analyses and prediction based on the demand data of query processing to user of user and makes corresponding response, thus realizes adaptive historical data management, and the function unit figure of intelligent object as shown in Figure 7.
The function of described intelligent object comprises: buffer memory pretreatment unit, analyses and prediction unit and motor unit, and the major function of each unit is described below:
Buffer memory pretreatment unit: each inquiry of user is queried after processing module does respective handling and can be intercepted by this unit, therefrom extract relevant query metadata, after filtration, some characteristic parameters of associated subscriber demand data are obtained after the process such as statistics, as the key assignments, enquiry frequency etc. of data query, and send these parameters to its subsequent cell.
Analyses and prediction unit: the characteristic parameter receiving the associated subscriber demand data that buffer memory pretreatment unit exports, and based on these parameters, adopt model analysis method, dope the different demands of user to data, and based on predicting the outcome, output action parameter.
Motor unit: the output of receiving and analyzing predicting unit, is converted into sensor network order bag and is sent to sensor network interface and then is issued to sensor network, with the data reporting functions of the memory node opened or in closure sensor network by action parameter.
After adding intelligent object, the Proxy workflow possessing storage and intelligent subscriber demand analysis and feedback function is as described below:
During netinit, Proxy opens to sensor network broadcast or closes the order bag of memory node data reporting functions, and the memory node in network is unwrapped according to this order and opened or close its data reporting functions (under default situations, the data reporting functions of memory node is in opening);
Step 301, network working stage, sensor network interface receives the packet from sensor network, is sent in data memory module and stores after parsing;
Step 302, user's inquiry is by after Internet/3G interface, the intelligible querying command of data memory module is converted in query processing module, be sent in data memory module on the one hand and retrieve corresponding data, being sent to intelligent object does the process of user data demand intellectual analysis on the other hand.
Step 303, when the data that user will inquire about are not in data memory module, inquires about unsuccessfully; Because intelligent object obtains the associated metadata of current failure inquiry equally, after its analyses and prediction, can generate open this data key values map the order bag of the data reporting functions of memory node, this order bag is issued to sensor network by sensor network interface, corresponding memory node opens its data reporting functions and reported data, and the inquiry next time of user can obtain these data.
Step 304, the characteristic parameter of user data demand is cached with in buffer memory pretreatment unit in intelligent object, these parameters comprise the key assignments of data, enquiry frequency etc., analyses and prediction unit can make corresponding control action according to user automatically to the enquiry frequency of a certain key assignments data, as when finding that a certain key assignments data are not queried within a very long time, intelligent object will close this data key values map the data reporting functions of memory node.
For storage means provided by the invention and system, if the technological means of employing is prior art when user carries out data query, does not do at this and launch explanation.
Based on explanation above, the data Stored Procedure of present system as shown in Figure 8, the event data that the perception data of the memory node reception sensing node transmission in sensor network or other memory nodes send also stores, judge whether its data reporting functions is in opening on the one hand, if, the perception data then stored or event data send Proxy node to by sensor network interface and are stored in the data memory module of Proxy further, otherwise do not upload and continue to wait for the arrival of new data; Judge whether the event of higher level on the other hand, if have, then merge the event data that the perception data that receives or event data obtain higher level, and sent to mapped memory node according to its key assignments, otherwise, do not process and continue to wait for that new data arrive.
Based on explanation above, the data query flow process of present system as shown in Figure 9, the inquiry of user is sent to query processing module by the Internet/3G interface of Proxy, query processing module is sent to inquiring user desired data in data memory module after resolving inquiry, data memory module returns the result of inquiry, if successful inquiring, then the result data obtained is returned to user by Internet/3G interface by query processing module; Otherwise, the key assignments of required data query is sent to intelligent object, the order bag opening the memory node data reporting functions that this key assignments maps is produced after intelligent object execution analysis, and being issued to corresponding memory node by sensor network interface, memory node receives this order and unwraps and open its data reporting functions and start to upload its data in the data memory module of Proxy.
The present invention is directed to the sensor network network application scene of the user data diversification of demand, propose the scheme of data management in the interior data management of comprehensive utilization net and net, achieve in netting the adaptive history data management based on Proxy storage and intelligent subscriber demand analysis and feedback outside managing based on many granularity datas of event rules mechanism and net, for user provides data, services flexibly.
It should be noted that, embodiment of the present invention of above introduction and and unrestricted.It will be understood by those of skill in the art that any amendment to technical solution of the present invention or the equivalent alternative spirit and scope not departing from technical solution of the present invention, it all should be encompassed in right of the present invention.
Claims (12)
1. a sensor network data access method for the many demands of user oriented, the method selects the date storage method of sensor network nodes based on user's request, comprises data storage and query method,
Described date storage method comprises:
Inner at sensor network, implement the data storage method based on data grid technology;
Outside at sensor network, adopt Sink node to carry out data storage;
Wherein,
The described data storage method based on data grid technology also performed by event rules mechanism as user sets up the step of many granularity datas storage policy before carrying out data storing steps;
Described Sink node is carried out data memory time base in the demand data of query processing to user of user and is done analyses and prediction and make corresponding response, realizes adaptive history data store;
Described data enquire method is:
Described Sink node resolves the inquiry request of user, and based on the requesting query user requested data after parsing, if described Sink node stores the data query of user, Query Result is returned to user by interface; Otherwise, the key assignments of required data query is sent to described Sink node and carries out the key assignments that intellectual analysis process obtains event needed for user, and the result according to analyzing and processing is sent the order bag opening the memory node data reporting functions that described key assignments maps, and being issued to corresponding memory node by sensor network interface, memory node receives this order and unwraps to open its data reporting functions and start to upload its data and store to described Sink node.
2. the sensor network data access method of the many demands of user oriented according to claim 1, is characterized in that, described many granularity datas storage policy is:
First, according to the task of user, take out different primary event, the perception data of sensor network interior joint collection is associated with the event belonging to it, and the perception data belonging to each event is given a unified key value Key;
Secondly, from primary event, taking out high level event, being associated with high-level event belonging to it by memory node merging mary event data that perception data obtains, and the mary event data belonging to high-level event is given a unified key assignments key; Repeat with this, until reach highest level event;
Finally; by above-mentioned event rules mechanism; perception data in network and event data are all endowed key assignments; memory node in these key assignments and network is set up mapping relations by Hash mechanism further that be similar to DHT; after node collects perception data; according to its key assignments perception data is sent on mapped memory node and stores, when memory node merges after perception data obtains event data, also according to its key assignments event data is sent on mapped memory node and stores.
3. the sensor network data access method of the many demands of user oriented according to claim 1, is characterized in that, described adaptive history data store comprises following steps further:
Step 301, open or close the step of the data reporting functions of the memory node of described sensor network, during netinit, described Sink node is opened to described sensor network broadcast or closes the order bag of memory node data reporting functions, and the memory node in described sensor network is unwrapped according to this order and opened or close its data reporting functions;
Step 302, network working stage, described sensor network interface receives the packet of described sensor network, is sent in the data memory module that described sink node comprises and stores after parsing;
Step 303, the step of inquiring user demand information, user sends the solicited message of certain category information that the described sensor network of inquiry gathers, this solicited message is converted to the intelligible querying command of data memory module of described Sink node by the query processing module of described sink node, described data memory module accepts described querying command and retrieves corresponding data, and described intelligent object also receives described querying command for carrying out analyzing and processing to user data demand simultaneously;
Step 304, when not storing the data that user will inquire about in described data memory module, then inquires about failure, enters next step; Otherwise, then in data memory module described in described user feedback, deposit the data of inquiry, enter last step;
Step 305, described intelligent object analyzes described querying command, be some characteristic parameters by described command analysis, described characteristic parameter comprises the data key values information of user's inquiry request information, then according to this data key values information generate open described data key values map the order bag of the data reporting functions of memory node, described order bag is issued to the memory node of sensor network by sensor network interface, described memory node uploads the data of described user inquiry to the data memory module of described Sink node, waits for that user sends inquiry request next time.
4. the sensor network data access method of the many demands of user oriented according to claim 3, is characterized in that, described characteristic parameter also comprises the enquiry frequency information of user to certain category information.
5. the sensor network data access method of the many demands of the user oriented according to claim 3 or 4, it is characterized in that, described intelligent object also performs the step of the data reporting functions of the memory node that closedown has been opened, according to user to the enquiry frequency information of certain category information, if when a certain key assignments is not queried within a very long time, described intelligent object then perform close this data key values map the order of the data reporting functions of memory node.
6. the sensor network data access method of the many demands of user oriented according to claim 3, is characterized in that, described in step 301 during netinit memory node data reporting functions default setting for opening.
7. the sensor network data access method of the many demands of user oriented according to claim 3, is characterized in that, the user described in step 303 sends inquiry request information by Internet/3G interface.
8. the sensor network data access system of the many demands of user oriented, this system carries out the mode of data storage based on the Sink node outside the data storage method of data grid technology and net, comprise: data perception node in sensor network, event data storage node in sensor network, sink memory node, it is characterized in that
In described net, sensing node comprises:
Perception data key assignments generation module, for giving the key assignments of event belonging to it by the perception data of node collection;
Sending module, for sending the perception data gathered to event data storage node in described net;
In described net, event data storage node is divided into some levels from high to low according to the rank of stored event, in described net, event data storage node specifically comprises: unlatching or closedown arrange module, carry out the unlatching of data reporting functions or the state of closedown for arranging memory node to described Sink node;
Event data key assignments generation module, the event data for being obtained by fusion perception data gives the key assignments of high-level event belonging to it;
Sending module, for sending event data to event data storage node in more senior net;
Receiver module, for receiving the perception data that described sensing node gathers or the event data accepting event data storage node transmission in more low-level net;
Described Sink node comprises intelligent object, for doing analyses and prediction based on the demand data of query processing to user of user and sending the order of data reporting functions of opening or closing some sensor network memory node, realize adaptive historical data management and store.
9. the sensor network data access system of the many demands of user oriented according to claim 8, is characterized in that, described Sink node also comprises:
Data memory module, the data that the memory node that the inquiry request information for receiving user carries out data query or reception unlatching reports store;
Query processing and sending module, for described request information is converted to the intelligible querying command of described data memory module, mail to described data storage cell and described intelligent cell;
Sensor network interface module, for connecting described Sink node and described sensor network; With
Subscriber Interface Module SIM, for receiving the data inquiry request information that user sends.
10. the sensor network data access system of many demands of user oriented according to claim 8 or claim 9, it is characterized in that, described intelligent object comprises further:
Buffer memory pretreatment unit: for intercepting each inquiry request of user, and extract relevant query metadata from described inquiry request; Described data query is obtained after filtration to some characteristic parameter of associated subscriber demand data with statistical disposition;
Analyses and prediction unit: the characteristic parameter receiving the associated subscriber demand data that described buffer memory pretreatment unit exports, and based on these parameters, adopt model analysis method, dope the different demands of user to data, and based on predicting the outcome, export the action parameter opening that memory node reporting functions;
Action executing and memory cell: receive the action parameter that described analyses and prediction unit exports, described action parameter is converted into sensor network order bag be sent to described sensor network interface and then be issued to sensor network, closes or turn-on data reporting functions with the memory node controlling to be correlated with;
Wherein, described analyses and prediction unit opens the action parameter of the reporting functions of memory node according to the key assignments information of user's inquiry request; Whether described analyses and prediction unit closes the action parameter of the reporting functions of memory node according to the judgement of user query frequency information.
The sensor network data access system of the many demands of 11. user oriented according to claim 10, it is characterized in that, the step of the data reporting functions of described closedown memory node, according to user to the enquiry frequency information of certain category information, if when a certain key assignments data are not queried within a very long time, then perform close this data key values map the order of the data reporting functions of memory node.
The sensor network data access system of the many demands of 12. user oriented according to claim 9, it is characterized in that, described user interface section comprises: Internet interface or 3G interface.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110153721.2A CN102821398B (en) | 2011-06-09 | 2011-06-09 | Method and system for storing sensor network data facing user multiple demands |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110153721.2A CN102821398B (en) | 2011-06-09 | 2011-06-09 | Method and system for storing sensor network data facing user multiple demands |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102821398A CN102821398A (en) | 2012-12-12 |
| CN102821398B true CN102821398B (en) | 2015-03-18 |
Family
ID=47305193
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110153721.2A Expired - Fee Related CN102821398B (en) | 2011-06-09 | 2011-06-09 | Method and system for storing sensor network data facing user multiple demands |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102821398B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103346975B (en) * | 2013-06-14 | 2016-07-06 | 清华大学 | Data based on wireless sensor network forward and storage method |
| CN103458021B (en) * | 2013-08-29 | 2016-07-06 | 中国科学院软件研究所 | A kind of event driven with the adaptive sensor method for organizing of state |
| CN106293490A (en) * | 2015-05-12 | 2017-01-04 | 中兴通讯股份有限公司 | Data storage, the method read, Apparatus and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101631361A (en) * | 2009-08-13 | 2010-01-20 | 上海交通大学 | Method for optimizing network lifetime of wireless LAN |
| CN101715201A (en) * | 2009-11-12 | 2010-05-26 | 北京邮电大学 | Method for acquiring information by sink node in wireless sensor network when moving |
| CN101959218A (en) * | 2009-10-25 | 2011-01-26 | 苏州大学 | Method for detecting event region based on splay tree |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FI118291B (en) * | 2004-12-22 | 2007-09-14 | Timo D Haemaelaeinen | Energy efficient wireless sensor network, node devices for the same and method of arranging, the communications in a wireless sensor network |
-
2011
- 2011-06-09 CN CN201110153721.2A patent/CN102821398B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101631361A (en) * | 2009-08-13 | 2010-01-20 | 上海交通大学 | Method for optimizing network lifetime of wireless LAN |
| CN101959218A (en) * | 2009-10-25 | 2011-01-26 | 苏州大学 | Method for detecting event region based on splay tree |
| CN101715201A (en) * | 2009-11-12 | 2010-05-26 | 北京邮电大学 | Method for acquiring information by sink node in wireless sensor network when moving |
Non-Patent Citations (1)
| Title |
|---|
| 无线传感器网络中数据存储与访问研究进展;蔚赵春, 周水庚, 关佶红;《电子学报》;20081031;第36卷(第10期);第2001-2010页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102821398A (en) | 2012-12-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111163181B (en) | Lightweight intelligent agricultural heterogeneous Internet of things management system | |
| CN110994798A (en) | Substation equipment monitoring system | |
| CN104052759A (en) | System for realizing add-and-play technology of internet of things | |
| CN102625486B (en) | Agent node, sensor network, and method for acquiring sensor network data | |
| CN107027342A (en) | Network range expander with many RF radio support for multiple network interfaces | |
| CN111741073B (en) | Electric power data transmission system based on 5G communication network | |
| CN110191148A (en) | A kind of statistical function distribution execution method and system towards edge calculations | |
| Qian et al. | An edge-computing paradigm for internet of things over power line communication networks | |
| CN102821398B (en) | Method and system for storing sensor network data facing user multiple demands | |
| CN103458021B (en) | A kind of event driven with the adaptive sensor method for organizing of state | |
| CN106817256A (en) | A kind of distributed system network resource operation management reliability method for improving | |
| CN109547524A (en) | User behavior storage method, device, equipment and storage medium based on Physical Network | |
| CN104570061A (en) | Seismic exploration wireless monitoring node and monitoring method and system based on same | |
| CN103942910B (en) | A kind of machine room fire based on IPv6 gives warning in advance method and system | |
| CN112421775A (en) | Local energy consumption control system based on power Internet of things and control method thereof | |
| CN103796343B (en) | M2M gateway devices and its application process | |
| CN112232437A (en) | Internet of things terminal data analysis method and system | |
| Tsai et al. | Reducing energy consumption by data aggregation in M2M networks | |
| Said et al. | DORS: A data overhead reduction scheme for hybrid networks in smart cities | |
| CN102970391B (en) | Inquiry of the domain name processing method, server and system | |
| CN111193779A (en) | Internet of things control method and system | |
| Gashi et al. | The integration of Wireless Sensor Networks, Mobile Networks and Cloud Engineering for a decision support system-A Systematic Literature Review | |
| CN104618357A (en) | Improved KNX/EIB (Konnex/European Installation Bus) protocol stack-based intelligent home management system | |
| CN116193295A (en) | Multi-element Internet of things data fusion and unified data processing method | |
| Kutseva | Adaptation of seven-layered IOT architecture for energy efficiency management in Smart House |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150318 Termination date: 20180609 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |