CN104036117A - Self-adaptive learning method and system based on sensor network - Google Patents
Self-adaptive learning method and system based on sensor network Download PDFInfo
- Publication number
- CN104036117A CN104036117A CN201410203656.3A CN201410203656A CN104036117A CN 104036117 A CN104036117 A CN 104036117A CN 201410203656 A CN201410203656 A CN 201410203656A CN 104036117 A CN104036117 A CN 104036117A
- Authority
- CN
- China
- Prior art keywords
- result
- propelling movement
- behavioral data
- data
- sensor network
- 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.)
- Granted
Links
Landscapes
- User Interface Of Digital Computer (AREA)
- Telephonic Communication Services (AREA)
- Manipulator (AREA)
- Selective Calling Equipment (AREA)
Abstract
The invention relates to a self-adaptive learning method and system based on a sensor network. The method comprises the steps that S1, a sensor is used for collecting behavior data; S2, the collected behavior data are classified and stored; S3, frequency distribution statistics is carried out on the stored behavior data to obtain a push result; S4, a scene scheme is formed through the push result; S5, the scene scheme is pushed, a feedback result obtained after pushing is received, and the feedback result is analyzed. The method and system have the advantages of being self-adaptive, intelligent and automatic, and complex operations of people on various devices are eliminated.
Description
Technical field
The present invention relates to wireless communication technology field, particularly, relate to a kind of adaptive learning method and system thereof based on sensor network.
Background technology
Wireless sensor network is considered to one of technology of tool application prospect of 21 century, has obtained showing great attention to of numerous government organs, scientific research group, commercial company.Along with the quick progress of hardware chip, wireless communication technology, sensor is simple, convenient with it, practical feature causes rapidly the very big concern of academia, business circles, and just towards miniaturization, diversified future development.Along with people more and more pay close attention to production, traffic, office and living environment, numerous sensing equipments starts to be applied to widely social industry-by-industry, and with the closely bound up life of people in the middle of.The concrete application scenarios such as intelligent industrial, intelligent building, Smart Home, intelligent vehicle swarms out, and facilitates greatly people's work behavior.Specifically, after small one by one sensor utilizes wireless mode connection networking, people can pass through the intelligent and portable equipment such as mobile phone, panel computer, realize the long-range control to various device, thereby have improved people's quality of life.
Along with the rapid progress of science and technology, the various equipment in social each field is more and more tending towards variation, complicated.But the control of current most of equipment must artificially participate in.Wanting everyone can grasp loaded down with trivial details and waste time and energy the skilled control of every equipment, is also unpractical.
Summary of the invention
The invention provides a kind of adaptive learning method and system thereof based on sensor network, there is the feature of self-adaptation, intellectuality and robotization, freed the loaded down with trivial details manipulation of people to various device.
For this purpose, the present invention proposes a kind of adaptive learning method based on sensor network, it is characterized in that, described method comprises step: S1, uses sensor to gather behavioral data; S2, classifies and stores gathered behavioral data; S3, carries out frequency distribution statistics by the behavioral data of storage, obtains pushing result; S4, forms scene scheme by described propelling movement result; S5, pushes described scene scheme, and receives the feedback result after pushing, and described feedback result is analyzed.
Wherein, described step S1 specifically comprises: use sensor to gather behavioral data, and described data are formed to specific tlv triple { device
i, timestamp
j, action
k, wherein, i, j, k is natural number, device
ibe i terminal name, timestamp
jbe j timestamp, action
kbe k behavior.
Wherein, described step S2 specifically comprises: described behavioral data is stored according to default very first time section, and the behavioral data that exceeded for the second time period is deleted.
Wherein, described step S3 specifically comprises: S31, in stored tlv triple, selecting any one is not empty device
i, obtain comprising device
ifirst set; S32, in described the first set, according to behavior action
kcarry out frequency distribution statistics, select action
kpeaked tlv triple, form second set; S33, in described the second set, according to timestamp timestamp
jcarry out frequency distribution statistics, select timestamp
jmaximal value, forms default propelling movement group; S34, judges described default propelling movement group and the feedback result receiving in described step S5; S35 if described default propelling movement group is conflicted mutually with described feedback result, gets rid of above-mentioned institute default propelling movement group in described behavioral data, repeats above-mentioned steps, until obtain the propelling movement result of not conflicting with described feedback result; S36, if described default propelling movement group do not conflict with described feedback result, directly using described default propelling movement group as push result.
Wherein, described step S5 comprises: be pushed to each corresponding terminal by pushing result by sensor, described terminal receives described propelling movement result, in special time, adjusts, form new behavioral data, and using described new behavioral data as feedback result.
Wherein, described step S5 also comprises: described propelling movement result is directly sent to each whole user terminal, exceed specific number of times if described user terminal is held adverse opinion to described propelling movement result, cancel the propelling movement to described propelling movement result, repeating step S3, obtains new propelling movement result.
According to another aspect of the present invention, provide a kind of adaptive and learning system based on sensor network, it is characterized in that, the described decorum comprises: data acquisition module, for using sensor to gather behavioral data; Data memory module, for classifying and store gathered behavioral data; Data analysis module, for the behavioral data of storage is carried out to frequency distribution statistics, obtains pushing result; Scene scheme forms module, and described propelling movement result is formed to scene scheme; Data-pushing and feedback module, for described scene scheme is pushed, and receive the feedback result after pushing, and described feedback result is analyzed.
Known by above-described embodiment, use adaptive learning method and the system thereof based on sensor network of the present invention, by using sensor, Intellisense, storage and analysis are carried out in data behavior to user terminal, choose the behavioral data that frequency is the highest certain scene scheme is provided in advance, and then the various terminals that sensor is housed are carried out to Based Intelligent Control, and the feedback information of terminal is made to response.Method and system of the present invention has self-adaptation, intellectuality and robotization a little, has freed greatly the loaded down with trivial details manipulation of people to various device.
Brief description of the drawings
Can more clearly understand the features and advantages of the present invention by reference to accompanying drawing, accompanying drawing is schematically to should not be construed as the present invention is carried out to any restriction, in the accompanying drawings:
Fig. 1 shows the process flow diagram of the adaptive learning method based on sensor network of the present invention;
Fig. 2 shows the process flow diagram of the step S3 of the adaptive learning method based on sensor network of the present invention;
Fig. 3 shows the structured flowchart of the adaptive and learning system based on sensor network of the present invention;
Fig. 4 shows the topological schematic diagram of an embodiment of the adaptive and learning system based on sensor network of the present invention.
Embodiment
Below in conjunction with accompanying drawing, embodiments of the present invention is described in detail.
Fig. 1 shows the process flow diagram of the adaptive learning method based on sensor network of the present invention.
With reference to Fig. 1, the adaptive learning method based on sensor network of the embodiment of the present invention comprises step:
S1, uses sensor to gather behavioral data.
In the present embodiment, gather the behavioral data of each terminal by being arranged on sensor on equipment, as information such as the settings of the switch of equipment, performance.Data mode can be set as: device name, timestamp and behavior, and three is formed to tlv triple { device
i, timestamp
j, action
k, wherein, i, j, k is natural number, device
ibe i device name, timestamp
jbe j timestamp, action
kbe k behavior.
S2, classifies and stores gathered behavioral data, described behavioral data is stored according to default very first time section, and the behavioral data that exceeded for the second time period is deleted.
In the present embodiment, very first time section is set as one week, and the second time period was set as 1 year, meanwhile, for very first time section and the second time period, can set according to concrete needs, as also very first time section being set as to one month, the second time period was set as six months etc.
Simultaneously, also the behavioral data of collection can be categorized into the data in very first time section, data in the second time period and the data in the 3rd time period, in the present embodiment, very first time section can be set as one week, and the second time period can be set as one month, and the 3rd time period can be set as 1 year, and very first time section and the data in the second time period are stored respectively, the data that exceeded for the 3rd time period are deleted.
S3, carries out frequency distribution statistics by the behavioral data of storage, obtains pushing result.
Fig. 2 shows the process flow diagram of the step S3 of the adaptive learning method based on sensor network of the present invention.
With reference to Fig. 2, step S3 specifically comprises:
S31, in the tlv triple of behavioral data, selecting any one is not empty device
i, as selected device
3set as the first set;
S32, in the first set, according to device
3behavior action
kcarry out frequency distribution statistics, select action
kmaximal value, as action
7, then according to action
7timestamp timestamp
jform the second set, as { device
3, timestamp
j, action7};
S33, in the second set, according to timestamp timestamp
jcarry out frequency distribution statistics, select timestamp
jmaximal value, as timestamp
2, form default propelling movement group { device
3, timestamp
2, action
7, repeat above-mentioned steps, thereby obtain the default propelling movement group of each equipment;
S34, judges described default propelling movement group and the feedback result receiving;
S35 if described default propelling movement group is conflicted mutually with described feedback result, gets rid of above-mentioned institute default propelling movement group in described behavioral data, repeats above-mentioned steps, until obtain the propelling movement result of not conflicting with described feedback result;
S36, if described default propelling movement group do not conflict with described feedback result, directly using described default propelling movement group as push result.
S4, forms scene scheme by described propelling movement result.
S5, pushes described scene scheme, and receives the feedback result after pushing, and described feedback result is analyzed.
Scene scheme is sent to each terminal accordingly, and add up the reaction of each terminal to scene scheme, exceed specific number of times if described terminal is held adverse opinion to described propelling movement result, cancel the propelling movement to described propelling movement result, repeating step S3, obtains new propelling movement result.
In the present embodiment, the number of times of adverse opinion is generally no more than three times.
For by the terminal of sensor control, this terminal is receiving after scene scheme, within the specific time, (in as one minute) is adjusted, data after sensor record adjustment also send these data as behavioral data, repeat step S1, meanwhile, using the behavior data as feedback result, judge with the default propelling movement group in step S34.
Fig. 3 shows the structured flowchart of the adaptive and learning system based on sensor network of the present invention.
In another embodiment of the present invention, provide a kind of adaptive and learning system based on sensor network, as shown in Figure 3, this system comprises:
Data acquisition module 110, for using sensor to gather behavioral data;
Data memory module 120, for classifying and store gathered behavioral data;
Data analysis module 130, for the behavioral data of storage is carried out to frequency distribution statistics, obtains pushing result;
Scene scheme forms module 140, and described propelling movement result is formed to scene scheme;
Data-pushing and feedback module 150, for described scene scheme is pushed, and receive the feedback result after pushing, and described feedback result is analyzed.
Fig. 4 shows the topological schematic diagram of an embodiment of the adaptive and learning system based on sensor network of the present invention
In yet another embodiment of the present invention, provide a kind of adaptive and learning system based on sensor network, this system specifically comprises: high in the clouds 210, intelligent gateway 220, sensor 230 and user terminal 240.
When adaptive and learning system in use based on sensor network is learnt, sensor 230 gathers the behavioral data of user terminal 240, forms tlv triple.
Then the data unification of collection is converged to intelligent gateway 220 by sensor 230.Intelligent gateway 220 possesses multiple communication interface, can either carry out that information interaction by communication with sensor 230, high in the clouds 210 that again can connecting Internet, can also directly connect the intelligent and portable equipment 250 that user uses, as mobile phone, panel computer etc.
After the behavioral data that intelligent gateway 220 gathers each sensor 230 gathers, be uploaded to high in the clouds 210 and store.
High in the clouds 210 is carried out frequency distribution statistics by very first time section (as a week) and the second time period by data respectively according to timestamp.
And the data that will exceed the 3rd time period (as 1 year) are deleted.
Data are being carried out after frequency distribution statistics, high in the clouds 210 will push result and form scene scheme, and send to intelligent gateway 220, intelligent gateway 220 sends to respectively corresponding sensor 230 according to the position of each user terminal 240 by scene scheme, and sensor 230 is controlled each user terminal 240 according to timestamp and behavior.
Meanwhile, high in the clouds 210 also can directly send to scene scheme user's intelligent and portable equipment 250, as mobile phone, panel computer etc.
For the scene scheme being pushed on user's intelligent and portable equipment 250, if user holds the number of times that adverse opinion exceedes setting, the propelling movement of this scene scheme of immediate cancel, re-starts frequency distribution statistics, forms new propelling movement result.
The user terminal 240 of controlling for sensor 230, if user (as a minute) in special time adjusts, sensor 230 records also send this behavioral data immediately.
Known by above-described embodiment, use adaptive learning method and the system thereof based on sensor network of the present invention, by using sensor, Intellisense, storage and analysis are carried out in data behavior to user terminal, choose the behavioral data that frequency is the highest certain scene scheme is provided in advance, and then the various terminals that sensor is housed are carried out to Based Intelligent Control, and the feedback information of terminal is made to response.Method and system of the present invention has self-adaptation, intellectuality and robotization a little, has freed greatly the loaded down with trivial details manipulation of people to various device.
Although described by reference to the accompanying drawings embodiments of the present invention, but those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, such amendment and modification all fall into by within claims limited range.
Claims (7)
1. the adaptive learning method based on sensor network, is characterized in that, described method comprises step:
S1, uses sensor to gather behavioral data;
S2, classifies and stores gathered behavioral data;
S3, carries out frequency distribution statistics by the behavioral data of storage, obtains pushing result;
S4, forms scene scheme by described propelling movement result;
S5, pushes described scene scheme, and receives the feedback result after pushing, and described feedback result is analyzed.
2. the adaptive learning method based on sensor network according to claim 1, is characterized in that, described step S1 specifically comprises:
Use sensor to gather behavioral data, and described data are formed to specific tlv triple { device
i, timestamp
j, action
k, wherein, i, j, k is natural number, device
ibe i terminal name, timestamp
jbe j timestamp, action
kbe k behavior.
3. the adaptive learning method based on sensor network according to claim 1, is characterized in that, described step S2 specifically comprises:
Described behavioral data is stored according to default very first time section, and the behavioral data that exceeded for the second time period is deleted.
4. the adaptive learning method based on sensor network according to claim 2, is characterized in that, described step S3 specifically comprises:
S31, in the tlv triple of stored behavioral data, selecting any one is not empty device
i, obtain comprising device
ifirst set;
S32, in described the first set, according to behavior action
kcarry out frequency distribution statistics, select action
kpeaked tlv triple, form second set;
S33, in described the second set, according to timestamp timestamp
jcarry out frequency distribution statistics, select timestamp
jmaximal value, forms default propelling movement group;
S34, judges described default propelling movement group and the feedback result receiving in described step S5;
S35 if described default propelling movement group is conflicted mutually with described feedback result, gets rid of above-mentioned institute default propelling movement group in described behavioral data, repeats above-mentioned steps, until obtain the propelling movement result of not conflicting with described feedback result;
S36, if described default propelling movement group do not conflict with described feedback result, directly using described default propelling movement group as push result.
5. the adaptive learning method based on sensor network according to claim 1, is characterized in that, above-mentioned steps S5 comprises:
Be pushed to each corresponding terminal by pushing result by sensor, described terminal receives described propelling movement result, in special time, adjusts, and forms new behavioral data, and using described new behavioral data as feedback result.
6. the adaptive learning method based on sensor network according to claim 5, is characterized in that, described step S5 also comprises:
Described propelling movement result is directly sent to each user terminal, exceed specific number of times if described user terminal is held adverse opinion to described propelling movement result, cancel the propelling movement to described propelling movement result, repeating step S3, obtains new propelling movement result.
7. the adaptive and learning system based on sensor network, is characterized in that, the described decorum comprises:
Data acquisition module, for using sensor to gather behavioral data;
Data memory module, for classifying and store gathered behavioral data;
Data analysis module, for the behavioral data of storage is carried out to frequency distribution statistics, obtains pushing result;
Scene scheme forms module, and described propelling movement result is formed to scene scheme;
Data-pushing and feedback module, for described scene scheme is pushed, and receive the feedback result after pushing, and described feedback result is analyzed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410203656.3A CN104036117B (en) | 2014-05-14 | 2014-05-14 | Self-adaptive learning method based on sensor network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410203656.3A CN104036117B (en) | 2014-05-14 | 2014-05-14 | Self-adaptive learning method based on sensor network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104036117A true CN104036117A (en) | 2014-09-10 |
| CN104036117B CN104036117B (en) | 2017-05-03 |
Family
ID=51466887
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410203656.3A Expired - Fee Related CN104036117B (en) | 2014-05-14 | 2014-05-14 | Self-adaptive learning method based on sensor network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104036117B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080126533A1 (en) * | 2006-11-06 | 2008-05-29 | Microsoft Corporation | Feedback based access and control of federated sensors |
| CN102387207A (en) * | 2011-10-21 | 2012-03-21 | 华为技术有限公司 | Push method and system based on user feedback information |
| CN103529770A (en) * | 2013-08-08 | 2014-01-22 | 山东大学 | Application of intelligent household appliance control system based on conditioned reflex mechanism |
-
2014
- 2014-05-14 CN CN201410203656.3A patent/CN104036117B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080126533A1 (en) * | 2006-11-06 | 2008-05-29 | Microsoft Corporation | Feedback based access and control of federated sensors |
| CN102387207A (en) * | 2011-10-21 | 2012-03-21 | 华为技术有限公司 | Push method and system based on user feedback information |
| CN103529770A (en) * | 2013-08-08 | 2014-01-22 | 山东大学 | Application of intelligent household appliance control system based on conditioned reflex mechanism |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104036117B (en) | 2017-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Abd-Elmagid et al. | A reinforcement learning framework for optimizing age of information in RF-powered communication systems | |
| Gunduz et al. | Designing intelligent energy harvesting communication systems | |
| CN105450654B (en) | Smart home development platform and its business development method based on middleware Technology | |
| CN102088754B (en) | Network access behavior-based access control method and system for wireless local area network | |
| CN103744411A (en) | Control method related to ZigBee technology for realizing smart home | |
| CN105116766B (en) | A kind of sound box parameter collocation method, mobile terminal, server and system | |
| Kamienski et al. | Context-aware energy efficiency management for smart buildings | |
| CN103489346A (en) | Experimental teaching method, equipment and system of internet of things | |
| Stojmenovic et al. | Inaugural issue of ‘cyber-physical systems’ | |
| CN107870571A (en) | Control the method, apparatus and smart machine of smart machine | |
| Raja Gopal et al. | Intelligent edge based smart farming with LoRa and IoT | |
| CN103607395A (en) | Touch sense data transmission method and apparatus | |
| CN107004003A (en) | model parameter fusion method and device | |
| CN103500059A (en) | Method, system and device for drawing through mobile internet | |
| CN105245571A (en) | Data acquisition method, apparatus and system | |
| Ashraf et al. | Device discovery and configuration scheme for Internet of Things | |
| CN104036117A (en) | Self-adaptive learning method and system based on sensor network | |
| CN111767473A (en) | Object selection method and computer-readable storage medium | |
| Li et al. | SEBAR: Social energy based routing scheme for mobile social delay tolerant networks | |
| CN105871583A (en) | User data obtaining method and device | |
| Asonye et al. | Analysis of personal area networks for zigbee environment using random early detection-active queue management model | |
| CN108683547A (en) | A kind of wireless sensor network configuration method of software definition | |
| CN107086957A (en) | Ad Hoc Routing Protocol verification methods based on BeagleBone Black | |
| CN111968361B (en) | Infrared repeated code learning method and infrared remote control system applied to Internet of things | |
| CN102721148A (en) | Air conditioner control method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190510 Address after: Room 106, Tianzuo Road, Xingcheng City, Mount Camera, Xigang Office, Qixia District, Nanjing, Jiangsu Province Patentee after: NANJING WANGHE INTELLIGENT TECHNOLOGY CO., LTD. Address before: 100083 Block B 608, Caizhi International Building, 18 Zhongguancun East Road, Haidian District, Beijing Patentee before: Beijing Wang He Time Technology Co., Ltd |
|
| TR01 | Transfer of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170503 Termination date: 20210514 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |