CN106686533A - Wireless body area network data transmission method based on reinforced learning algorithm - Google Patents
Wireless body area network data transmission method based on reinforced learning algorithm Download PDFInfo
- Publication number
- CN106686533A CN106686533A CN201710011775.2A CN201710011775A CN106686533A CN 106686533 A CN106686533 A CN 106686533A CN 201710011775 A CN201710011775 A CN 201710011775A CN 106686533 A CN106686533 A CN 106686533A
- Authority
- CN
- China
- Prior art keywords
- node
- data
- area network
- body area
- aggregation
- 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
- 230000005540 biological transmission Effects 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000002776 aggregation Effects 0.000 claims description 59
- 238000004220 aggregation Methods 0.000 claims description 59
- 238000005728 strengthening Methods 0.000 claims description 9
- 230000002708 enhancing effect Effects 0.000 claims description 6
- 238000012546 transfer Methods 0.000 claims description 6
- 230000000295 complement effect Effects 0.000 claims description 2
- 238000000605 extraction Methods 0.000 claims 1
- 230000008569 process Effects 0.000 description 11
- 230000008859 change Effects 0.000 description 8
- 230000036541 health Effects 0.000 description 7
- 239000008280 blood Substances 0.000 description 5
- 210000004369 blood Anatomy 0.000 description 5
- 230000003862 health status Effects 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 5
- 230000009471 action Effects 0.000 description 3
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 3
- 230000036772 blood pressure Effects 0.000 description 3
- 201000010099 disease Diseases 0.000 description 3
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 229910052760 oxygen Inorganic materials 0.000 description 3
- 239000001301 oxygen Substances 0.000 description 3
- 239000013589 supplement Substances 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000008103 glucose Substances 0.000 description 2
- 206010020772 Hypertension Diseases 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 210000003423 ankle Anatomy 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000000386 athletic effect Effects 0.000 description 1
- 208000029078 coronary artery disease Diseases 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 201000001421 hyperglycemia Diseases 0.000 description 1
- 238000001727 in vivo Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 210000002414 leg Anatomy 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 210000000707 wrist Anatomy 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/70—Services for machine-to-machine communication [M2M] or machine type communication [MTC]
Abstract
The invention relates to a wireless body area network data transmission method based on a reinforced learning algorithm. The method aims at characteristics of a wireless body area network and is a hybrid data transmission method. The wireless body area network data transmission method comprises the steps of dividing nodes to different grades according to heterogeneous characteristics of different node energy levels, different data services and different workloads; calculating channel states of nodes and a convergent node according to a Markov channel model, determining whether a relay node is required in data transmission; if channel state is good, directly transmitting node data to a convergent node, and otherwise, searching an optimal path between the nodes and the convergent node by means of the reinforced learning algorithm based on node grade dividing, and continuously optimizing path selection in a data transmission process. The wireless body area network data transmission method can ensure efficient and reliable physiological data related in the wireless body area network. The wireless body area network data transmission method can be applied in multiple fields for ensuring accuracy of data received by a monitoring center.
Description
Technical field
The present invention relates to it is a kind of using the method that learning algorithm solves wireless body area network data transmission problems is strengthened, for profit
The medical health field being monitored with wireless body area network, to ensure that Surveillance center receives the accuracy of data one kind side is provided
Method.
Background technology
The 6th censuses result of State Statistics Bureau shows that China 60 years old and above population account for 13.26%, than the 5th time
Censuses rise 2.93%.In the old man of 60 years old and the above, old man's proportion of health is 44%, basic health it is old
People accounts for 39%, and old man that is unhealthy but can taking care of oneself accounts for 14%, it is impossible to which the old man of self-care accounts for 3%.As can be seen here, 56% old man needs
To be concerned the situation of body the moment.Even for the 44% of current health old man, their body is it can also happen that disease
Become.In addition, (not yet announcing the investigation knot of 2015 according to State Statistics Bureau's Sample Survey result of 2004 to 2014
Really), the ratio of 60 years old, 80 years old and more than 90 years old population is constantly rising.As can be seen here, social aging accelerates.For
For old people, their health status more merit attention.With advancing age, many functions of the old person start to decline
Weak, resistance is also worse and worse.Many old men are with diseases such as hypertension, hyperglycemia, coronary heart disease.The health of old people is to children
For be the part most important thing thing.But, need work and life as we of children, it is impossible to accompany always in old man
At one's side.The health of father and mother often becomes the thing that children care for the most.Particularly with old solitary people, their health status are more
It is troubling, old people's dead tragedy being just found for many days at home even occur in some places.Accordingly, it would be desirable to a kind of skill
Art is remotely monitoring the health status of old people.In addition, for the patient of hospital, it is also desirable to which such a technology is making medical care people
Member monitors at any time their health, understands their demand.
For the problems referred to above, a kind of emerging network technology has occurred --- wireless body area network.Wireless body area network can be right
The health status of human body are continued to monitor.The body surface of human body or internal is placed sensoring devices in, to blood pressure, blood glucose and pulse
It is acquired Deng sign data and transmits to aggregation node.Aggregation node finally sends data to Surveillance center of hospital, relatives
Mobile phone or Community Service Center.In wireless body area network, because sensor node is attached to human body, the movement of people or appearance
The change of gesture can all cause network topology to change.In addition, for the sensor node in wireless body area network, especially in vivo
Node, is difficult to supplement energy.Node energy exhausts, and can cause link failure, can also cause the change of network topology.It is unstable
Network topology easily cause the loss of packet.And in wireless body area network, the deployment of sensor node does not possess redundancy,
This requires that each packet will transmitting.Packet loss can cause the inaccurate of gathered data, and then affect the prison to human body
Effect is surveyed, What is more, the life of entail dangers to people.Therefore, for wireless body area network, how will while effectively utilizes energy
Data are reliably transmitted to aggregation node becomes crucial.
There are the following problems for existing data transmission method in wireless body area network：(1) pass with regard to the data of wireless body area network
Transmission method, existing scheme is broadly divided into single-hop transmission and the class of multi-hop transmission two.However, there is drawback in both schemes.It is right
In the first scheme, the data for collecting are transmitted directly to aggregation node by sensor node.If sensor node distance is converged
Farther out, single-hop transmission mode can then consume more energy to poly- node.In addition, with the movement or the change of posture of human body,
Channel status Jing between node and aggregation node often changes.If the channel shape between sensor node and aggregation node
State is poor, and directly data being issued into aggregation node can then cause the loss of packet.For second scheme, each sensor section
When point collects data, aggregation node is sent data to by the forwarding of multiple nodes.The scale of wireless body area network after all
Less, if adopting multi-hop transmission mode merely, unnecessary data forwarding will certainly be brought, this will bring and postpone and more
Energy expenditure.(2) isomerism of node is not accounted in existing data transmission scheme.Wireless body area network interior joint is present
Isomerism, is mainly reflected in：Each node energy level is different, and present position is different, and the complexity for supplementing energy is different, to body
It is difficult that interior nodes supplement energy equivalence；The data service of each node collection is different, with different importances；The work of each node
Amount is different, and some need to continue to monitor data, and some only need to gathered data in a short period of time.
The content of the invention
The purpose of the present invention is for problem present in existing wireless body area network data transmission method, there is provided Yi Zhonggao
The data transmission method of effect, to realize from source node reliably transmitting data to convergence section on the premise of effectively utilizes energy
Point.
In order to achieve the above object, the technical scheme is that there is provided it is a kind of based on strengthen learning algorithm without wire body
Area network data transmission method, it is characterised in that comprise the following steps：
Step 1, for the energy level of sensing node, data service this isomery different with workload in wireless body area network
Property feature, is different grades by node division.According to deployed position divided rank A of sensing node, the grade of body surface node
It is higher, as far as possible using body surface node for data forwarding；According to importance divided rank B of node, the grade of important node is relatively low, makes
Important node undertakes less forwarding task；According to workload divided rank C of node, very short time can complete the section of acquisition tasks
Point it is higher ranked, making to undertake forwarding task this kind of node more.
Step 2, current sensor node are collected after data, are judged and the channel status between aggregation node, if channel
In good condition, then current sensor node directly sends the data to aggregation node, this DTD；Otherwise, enter
Step 3；
Step 3, using current sensor node as source node；
Step 4, the via node for selecting to lead to aggregation node for source node using enhancing learning algorithm；
Step 5, data are reached after the selected via node of step 4, are judged between current hop node and aggregation node
Channel status, if channel status is good, current hop node directly sends the data to aggregation node, this data transfer
Terminate；Otherwise, using current hop node as source node, return to step 4.
Preferably, in the step 2 or the step 5, using Markov channels model current sensor section is judged
Point or the channel status between current hop node and aggregation node.
Preferably, with the shape of data message when the current sensor node or the current hop node send data
Formula is transmitted, and is sending node by the current sensor node or the current hop node definition that send data, then
The field of data message includes position, the dump energy of sending node, the optimum of sending node to aggregation node of sending node
The quality in path, the mark of next-hop node and load data.
Preferably, in the step 5, when current hop node listens to the data message for sending over, extract first
Information in data message, will be inserted or updated in local neighbor table about the information of sending node, then, current hop section
Point judges whether itself is next-hop node indicated in data message, if it is not, then directly abandoning data message；It is no
Then, the channel status between current hop node and aggregation node is judged, if channel status is good, current hop node is direct
Aggregation node is sent the data to, otherwise, using current hop node as source node, return to step 4.
Preferably, using the via node for strengthening learning algorithm and leading to aggregation node for source node selection in the step 4
Comprise the following steps：
Step 4.1, source node calculate the quality in the path that aggregation node is reached by each neighbor node, select path matter
Maximum neighbor node is measured as next-hop node, wherein, source node i reaches the current road of aggregation node by adjacent node j
Path quality is Qnew(i, j), then have：
Qnew(i, j)=(1- α) Qold(i, j)+α (L (i, j)+Rj), in formula, α is learning rate, Qold(i, j) is source node i
The path quality history value of aggregation node is reached by adjacent node j, L (i, j) is from source node i to the direct of adjacent node j
Link-quality, RjFor the quality of the adjacent node j that locally stored to the optimal path of aggregation node；
If step 4.2, source node i select adjacent node j as next-hop forward node, R is updatedi, RiRepresent source node
The quality of the optimal path of i to aggregation node：
Ri=Qnew(i, j)；
Step 4.3, source node i updated the data message before data message is sent, and the position of itself is believed
Quality R of breath, dump energy, the optimal path to aggregation node for currently being foundiWith selected next-hop node, i.e. phase
Neighbors j is substituted into the head of data message, and then, source node i sends the data message after renewal.
Preferably, the L (i, j) is calculated by below equation：
In formula, EjRepresent the dump energy of node j, Aj、BjAnd CjRepresent respectively and press
The grade of the adjacent node j divided according to node deployment position, node importance and node workload, DijThen represent source node i
To the distance of adjacent node j.
The present invention in combination with artificial intelligence field, provides wireless body area network for wireless body area network a kind of based on enhancing
Hybrid-type data transmission method of study, it is possible to achieve reliably pass data from source node while effectively utilizes energy
Aggregation node, and then the accuracy of guarantee Surveillance center information collected are transported to, is that data analysiss lay the foundation, such that it is able to effective
Ground prevention disease simultaneously rescues in time emergency situations.Scheme in the present invention can be used to build healthy platform, real-time monitoring human body
Health status, apply also for other multiple fields such as military exercises, athletic training, amusement be shared.
Description of the drawings
Fig. 1 collects data transfer flow process figure during data for sensor node；
Fig. 2 listens to data transfer flow process figure during data for sensor node；
Fig. 3 is using the flow chart for strengthening learning algorithm selection via node.
Specific embodiment
With reference to specific embodiment, the present invention is expanded on further.It should be understood that these embodiments are merely to illustrate the present invention
Rather than restriction the scope of the present invention.In addition, it is to be understood that after the content for having read instruction of the present invention, people in the art
Member can make various changes or modifications to the present invention, and these equivalent form of values equally fall within the application appended claims and limited
Scope.
The present invention relates to two kinds of node, one kind is sensor node, and one kind is aggregation node.By sensor node
Ankle, leg, wrist and arm of human body etc. are deployed in, the sign datas such as blood pressure, blood glucose and pulse are acquired and are passed
Transport to aggregation node.Using mobile phone as aggregation node in the present invention, aggregation node sends data to monitoring eventually through outer net
Center.For the data transmission problems inside wireless body area network, the present invention proposes a kind of based on the data biography for strengthening learning algorithm
Transmission method, specifically includes following steps：
(1) it is different grades by node division.In wireless body area network, there is isomerism in sensor node：Each node
Energy level is different, and present position difference causes the complexity of supplementary energy different, supplements energy equivalence to internal node and is stranded
It is difficult；The data service of each node collection is different, comprising electrocardiogram, blood pressure and blood oxygen etc., with different importances；Each node
Workload is different, and some need to continue to monitor data, for example：Electrocardio node of graph needs one day or several days continuous monitorings are Electrocardiographic
Dynamic situation, and some only need to gathered data in a short period of time, for example：Blood oxygen node can be completed in the very short time
Conventional blood oxygen inspection.For this feature of node isomerism, the present invention sets up grade classification mechanism.According to the portion of sensing node
Administration position divided rank A, body surface node it is higher ranked, as far as possible using body surface node for data forwarding；According to the importance of node
Divided rank B, the grade of important node is relatively low, makes important node undertake forwarding task less；Divided according to the workload of node etc.
Level C, very short time can complete the higher ranked of the node of acquisition tasks, make to undertake forwarding task this kind of node more.
(2) channel status between present node and aggregation node is judged according to Markov channels model.Wireless body area
The channel model commonly used in net is path loss model and power delay profile model, and the former is used for describing in signals transmission
Power attenuation, the latter is used for portraying the characteristic of multipath channel.But, both models lack to dynamics of channels variation characteristic
Description in real time.For this problem, the present invention is carried using IEEE 802.15.6 working groups in wireless body area network Channel Modeling document
The scheme for going out, using Markov channels model.Markov process of the model based on finite state, can believe according to history
The rule that road state changes, thus it is speculated that the trend that future channel status change.Therefore, the model can characterize the dynamic change of channel
Change.Concrete modeling process is as follows：
(2.1) channel of wireless body area network has been divided into into state and bad state two states, under good state, can be successful
Transmission data.
(2.2) using universally recognized two states Markov channels model in the industry, P is as follows for its state-transition matrix：
In formula (1), state set S={ 0,1 }, 0 represents bad state, and 1 has represented state；PijRepresent that channel is turned by i states
For j shape probability of states, formula (2) condition is met：
If S0For the original state variable of the channel between node and aggregation node, then after experiencing the n moment, the channel is preferably
The Probability p (n) of state is：
(2.3) test human body be expert at, it is static and act the data transmission scenarios squatted etc. under different gestures；
(2.4) above Markov model is fitted according to these measured datas, so can according to formula (1)-
Formula (3) predicts the channel status between any moment lower node and aggregation node.
(3) on the basis of step (1) and step (2), data transfer flow process such as Fig. 1-Fig. 3 institutes involved in the present invention
Show.
(3.1) as shown in figure 1, when node collects data, the channel model first in step (2) judges current
Channel status between node and aggregation node.If channel status between the two is good, data are transmitted directly to converge
Poly- node；Otherwise, via node is selected to carry out data forwarding using enhancing learning algorithm, detailed process is shown in step (3.3).Separately
Outward, it should be noted that being transmitted in the form of message when data are sent, the field of data message includes sending node
Position, the dump energy of sending node, the quality of the optimal path of sending node to Sink node, the mark of next-hop node
And load data.Wherein next-hop node is used to indicate and is forwarded by which node.
(3.2) as shown in Fig. 2 when node listens to the data message that other nodes are sended over, message is extracted first
In information, will be inserted or updated in local neighbor table about the information of sending node.By this behave, node can be
The up-to-date information of neighbor node is constantly obtained during transmission data.Then, present node judges whether itself is message
In indicated next-hop node, if it is not, then directly abandoning data message；Otherwise, sentenced according to the channel model of step (2)
Break itself channel status and aggregation node between, if in good condition, data is transmitted directly to into aggregation node；Otherwise,
Via node is selected to carry out data forwarding using learning algorithm is strengthened, detailed process is shown in step (3.3).
(3.3) as shown in figure 3, when needing via node to carry out data forwarding, using learning algorithm is strengthened route is selected
Path.Using the path selection process of present node to aggregation node as an enhancing learning tasks, and it is by the task description
One Markovian decision process MDP (S；A；P；R), wherein S is all possible state set, and A is possible set of actions,
What P was represented is the probability of state transfer, and R is the value of feedback obtained from environment.Agency is in state siTake action ai, then can receive
To one of environment value of feedback r immediatelyi.One Markovian decision process is then by this series of actions si, state aiImmediately
Value of feedback riThe set for being constituted.
(3.3.1) present node calculates the corresponding Q-value of each neighbor node according to formula (4), the section for selecting Q-value maximum
Point is used as next-hop node.
Qnew(i, j)=(1- α) Qold(i, j)+α (L (i, j)+Rj) (4)
In formula (4), Qnew(i, j) represents working as the path quality that present node i passes through both candidate nodes j arrival aggregation node
Front value, Qold(i, j) is then history value；α is a constant, represents learning rate, and this Graph One factor determines historical information and current information
Two-part weight.If human posture's change is more frequently, current information accounts for higher weight., whereas if monitoring
It is bed patient, then historical information accounts for higher weight.The present invention considers both sides factor, and α values are 0.5；RjGeneration
Quality of the node j that table is locally stored to the optimal path of aggregation node；L (i, j) is represented from node i to the direct of node j
Link-quality, can be calculated by formula (5).
In formula (5), EjRepresent the dump energy of node j, Aj、BjAnd CjRepresent respectively according to node deployment position, node weight
The grade (referring to described in step 1) of the node j that the property wanted and node workload are divided, DijThen represent node i to node j away from
From.
If (3.3.2) present node i selects node j as next-hop forward node, according to formula (6) itself is updated
RiValue.
Ri=Qnew(i, j) (6)
In formula (6), RiThe quality of the optimal path of the node i to aggregation node that expression is currently found.Formula (6) table
Bright, this moment, the optimal path of present node i to aggregation node is then to reach aggregation node by node .j.
(3.3.3) present node i updated the data message before message is sent, and by the positional information of itself, remained
Quality R of complementary energy, the optimal path to aggregation node for currently being foundiNumber is substituted into selected next-hop node j
According to the head of message.Then, node i sends the data message after renewal.In this way, node is sending data
Incidentally value of feedback R during message, which saves the energy expense for sending control message.
(3.3.4) when next-hop node j listens to data message, carry out according to step (3.2).Until data message is supported
Up to aggregation node, one time data transmission procedure terminates.The present invention is by such a learning process come the choosing of constantly path optimizing
Select, and value of feedback R is carried in data message transmission, the energy that transmission control message is saved in this way is opened
Pin.
Claims (6)
1. it is a kind of based on the wireless body area network data transmission method for strengthening learning algorithm, it is characterised in that to comprise the following steps：
It is step 1, special for the energy level of sensing node in wireless body area network, data service this isomerism different with workload
Point, is different grades by node division.According to deployed position divided rank A of sensing node, body surface node it is higher ranked,
Body surface node for data forwarding is used as far as possible；According to importance divided rank B of node, the grade of important node is relatively low, makes important
Node undertakes less forwarding task；According to workload divided rank C of node, very short time can complete the node of acquisition tasks
It is higher ranked, making to undertake forwarding task this kind of node more；
Step 2, current sensor node are collected after data, are judged and the channel status between aggregation node, if channel status
Well, then current sensor node directly sends the data to aggregation node, this DTD；Otherwise, into step
3；
Step 3, using current sensor node as source node；
Step 4, the via node for selecting to lead to aggregation node for source node using enhancing learning algorithm；
Step 5, data are reached after the selected via node of step 4, judge the letter between current hop node and aggregation node
Road state, if channel status is good, current hop node directly sends the data to aggregation node, this data transfer knot
Beam；Otherwise, using current hop node as source node, return to step 4.
2. as claimed in claim 1 a kind of based on the wireless body area network data transmission method for strengthening learning algorithm, its feature exists
In in the step 2 or the step 5, using Markov channels model current sensor node or current hop being judged
Channel status between node and aggregation node.
3. as claimed in claim 1 a kind of based on the wireless body area network data transmission method for strengthening learning algorithm, its feature exists
In, it is transmitted in the form of data message when the current sensor node or current hop node transmission data,
Be sending node by the current sensor node or the current hop node definition that send data, then the word of data message
Section include the position of sending node, the dump energy of sending node, the quality of the optimal path of sending node to aggregation node, under
The mark and load data of one hop node.
4. as claimed in claim 3 a kind of based on the wireless body area network data transmission method for strengthening learning algorithm, its feature exists
In in the step 5, when current hop node listens to the data message for sending over, first in extraction data message
Information, will be inserted or updated in local neighbor table about the information of sending node, and then, current hop node judges that itself is
No is next-hop node indicated in data message, if it is not, then directly abandoning data message；Otherwise, in judging currently
After the channel status between node and aggregation node, if channel status is good, current hop node directly sends the data to
Aggregation node, otherwise, using current hop node as source node, return to step 4.
5. as claimed in claim 1 a kind of based on the wireless body area network data transmission method for strengthening learning algorithm, its feature exists
In using learning algorithm is strengthened being that source node is selected to lead to the via node of aggregation node and comprised the following steps in the step 4：
Step 4.1, source node calculate the quality in the path that aggregation node is reached by each neighbor node, select path quality most
Big neighbor node as next-hop node, wherein, source node i by adjacent node j reach aggregation node current path matter
Measure as Qnew(i, j), then have：
Qnew(i, j)=(1- α) Qold(i, j)+α (L (i, j)+Rj), in formula, α is learning rate, Qold(i, j) passes through for source node i
Adjacent node j reaches the path quality history value of aggregation node, and L (i, j) is the direct link from source node i to adjacent node j
Quality, RjFor the quality of the adjacent node j that locally stored to the optimal path of aggregation node；
If step 4.2, source node i select adjacent node j as next-hop forward node, R is updatedi, RiRepresent source node i extremely
The quality of the optimal path of aggregation node：
Ri=Qnew(i, j)；
Step 4.3, source node i updated the data message before data message is sent, and by the positional information of itself, remained
Quality R of complementary energy, the optimal path to aggregation node for currently being foundiWith selected next-hop node, i.e. adjacent segments
Point j is substituted into the head of data message, and then, source node i sends the data message after renewal.
6. as claimed in claim 5 a kind of based on the wireless body area network data transmission method for strengthening learning algorithm, its feature exists
In the L (i, j) is calculated by below equation：
In formula, EjRepresent the dump energy of node j, Aj、BjAnd CjRepresent respectively according to section
The grade of the adjacent node j that point deployed position, node importance and node workload are divided, DijSource node i is then represented to phase
The distance of neighbors j.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710011775.2A CN106686533B (en) | 2017-01-06 | 2017-01-06 | Wireless body area network data transmission method based on reinforcement learning algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710011775.2A CN106686533B (en) | 2017-01-06 | 2017-01-06 | Wireless body area network data transmission method based on reinforcement learning algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106686533A true CN106686533A (en) | 2017-05-17 |
| CN106686533B CN106686533B (en) | 2020-05-26 |
Family
ID=58849413
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710011775.2A Expired - Fee Related CN106686533B (en) | 2017-01-06 | 2017-01-06 | Wireless body area network data transmission method based on reinforcement learning algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106686533B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108064063A (en) * | 2017-12-26 | 2018-05-22 | 江苏金陵机械制造总厂 | It is a kind of based on energy distance than routing of sensor networks implementation method |
| CN108881467A (en) * | 2018-07-04 | 2018-11-23 | 深圳凯达通光电科技有限公司 | Industrial production machines people's real-time monitoring system |
| CN109187874A (en) * | 2018-09-26 | 2019-01-11 | 东莞幻鸟新材料有限公司 | Regional air quality intelligent real-time publishing system |
| CN112512092A (en) * | 2020-11-03 | 2021-03-16 | 中国科学院深圳先进技术研究院 | Multi-node human body communication networking method and device |
| CN113052220A (en) * | 2021-03-16 | 2021-06-29 | 洛阳城市建设勘察设计院有限公司郑州工程分公司 | Sealing performance strength detection system, terminal and medium for direct-buried heat supply pipeline research |
| CN113093654A (en) * | 2021-03-29 | 2021-07-09 | 武汉丰乐精密制造有限公司 | Automobile precision device processing control system, control method, terminal and medium |
| CN113190403A (en) * | 2021-04-22 | 2021-07-30 | 杭州雅观科技有限公司 | Monitoring system, computer equipment, terminal and medium for operation state of big data platform |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009114626A2 (en) * | 2008-03-11 | 2009-09-17 | The Regents Of The University Of California | Wireless sensors and applications |
| CN103974366A (en) * | 2014-04-28 | 2014-08-06 | 南京邮电大学 | Wireless body area network routing method based on reinforcement learning |
| CN105228159A (en) * | 2014-06-18 | 2016-01-06 | 北京邮电大学 | Based on the wireless sense network coverage enhancement algorithm of gridding and improve PSO algorithm |
-
2017
- 2017-01-06 CN CN201710011775.2A patent/CN106686533B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009114626A2 (en) * | 2008-03-11 | 2009-09-17 | The Regents Of The University Of California | Wireless sensors and applications |
| CN103974366A (en) * | 2014-04-28 | 2014-08-06 | 南京邮电大学 | Wireless body area network routing method based on reinforcement learning |
| CN105228159A (en) * | 2014-06-18 | 2016-01-06 | 北京邮电大学 | Based on the wireless sense network coverage enhancement algorithm of gridding and improve PSO algorithm |
Non-Patent Citations (2)
| Title |
|---|
| RAMTIN KAZEMI 等: "Dynamic power control in Wireless Body Area Networks using reinforcement learning with approximation", 《2011 IEEE 22ND INTERNATIONAL SYMPOSIUM ON PERSONAL, INDOOR AND MOBILE RADIO COMMUNICATIONS》 * |
| 宝磊: "基于增强学习的无线体域网路由算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108064063A (en) * | 2017-12-26 | 2018-05-22 | 江苏金陵机械制造总厂 | It is a kind of based on energy distance than routing of sensor networks implementation method |
| CN108881467A (en) * | 2018-07-04 | 2018-11-23 | 深圳凯达通光电科技有限公司 | Industrial production machines people's real-time monitoring system |
| CN109187874A (en) * | 2018-09-26 | 2019-01-11 | 东莞幻鸟新材料有限公司 | Regional air quality intelligent real-time publishing system |
| CN112512092A (en) * | 2020-11-03 | 2021-03-16 | 中国科学院深圳先进技术研究院 | Multi-node human body communication networking method and device |
| CN112512092B (en) * | 2020-11-03 | 2023-03-24 | 中国科学院深圳先进技术研究院 | Multi-node human body communication networking method and device |
| CN113052220A (en) * | 2021-03-16 | 2021-06-29 | 洛阳城市建设勘察设计院有限公司郑州工程分公司 | Sealing performance strength detection system, terminal and medium for direct-buried heat supply pipeline research |
| CN113093654A (en) * | 2021-03-29 | 2021-07-09 | 武汉丰乐精密制造有限公司 | Automobile precision device processing control system, control method, terminal and medium |
| CN113190403A (en) * | 2021-04-22 | 2021-07-30 | 杭州雅观科技有限公司 | Monitoring system, computer equipment, terminal and medium for operation state of big data platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106686533B (en) | 2020-05-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106686533A (en) | Wireless body area network data transmission method based on reinforced learning algorithm | |
| US10433726B2 (en) | Systems, methods, and devices for remote health monitoring and management using internet of things sensors | |
| Lara et al. | A mobile platform for real-time human activity recognition | |
| CN107456743A (en) | Exercise guidance method and system | |
| CN106357802A (en) | System and method for evaluating health conditions based on internet of things | |
| CN106940757A (en) | Health supervision service method and collaboration host computer system based on Internet of Things | |
| CN103598873B (en) | A kind of physiological signal intelligent monitor system based on adaptive wireless sensor network | |
| CN104035396B (en) | Distributed Activity recognition method based on wireless sensor network | |
| CN107103182A (en) | A kind of heart disease Warning System and method based on deep learning algorithm | |
| CN110505597A (en) | A kind of data transmission method of wireless sensor network | |
| CN105891775A (en) | Smart monitoring method and apparatus based on region positioning | |
| CN109525956A (en) | The energy-efficient method of data capture of sub-clustering in wireless sense network based on data-driven | |
| De Paola et al. | An ambient intelligence system for assisted living | |
| CN103391594A (en) | Wireless body area network routing optimization method and device | |
| CN105796088A (en) | Craniocerebral pressure non-invasive monitoring and analyzing system | |
| Parameshachari et al. | Health-care monitoring of patient using cnn based model in internet of things | |
| Bisio et al. | Towards IoT-based eHealth services: A smart prototype system for home rehabilitation | |
| CN106250712A (en) | A kind of ureteral calculus Forecasting Methodology based on increment type neural network model and prognoses system | |
| Karimunnisa et al. | Advanced IOT based system for cricketers health supervision | |
| CN106384013A (en) | Incremental neural network model-based type-II diabetes prediction method and prediction system | |
| CN106295238A (en) | A kind of hypertensive nephropathy Forecasting Methodology based on increment type neural network model and prognoses system | |
| Bahanfar et al. | Reliable communication in wireless body area sensor network for health monitoring | |
| Ullas et al. | Sensor overlay for remote patients monitoring | |
| CN112954597B (en) | Indoor positioning method for nursing care | |
| CN105069728A (en) | Emotion inference method based on wireless sensor network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200526 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |