CN103428821A - Asynchronous neighbor discovery method - Google Patents
Asynchronous neighbor discovery method Download PDFInfo
- Publication number
- CN103428821A CN103428821A CN2013103147982A CN201310314798A CN103428821A CN 103428821 A CN103428821 A CN 103428821A CN 2013103147982 A CN2013103147982 A CN 2013103147982A CN 201310314798 A CN201310314798 A CN 201310314798A CN 103428821 A CN103428821 A CN 103428821A
- Authority
- CN
- China
- Prior art keywords
- node
- message
- neighbours
- time slot
- nodes
- 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.)
- Pending
Links
Abstract
The invention discloses an asynchronous neighbor discovery method. The asynchronous neighbor discovery method includes steps that 1), a node x and a node y are defined; 2), a message m used for identifying a sender and sending to a receiving node is defined; 3), when the node x discovers the node y in the node x, if the node x and the node y are neighbors, then the node x receives at least one message m from the node y; 4), a set of limit nodes is given, and each node has a limit set of a neighboring node. The asynchronous neighbor discovery method has the advantages that the distributed neighbor discovery method is adopted, so that global synchronization is not needed and resources can be effectively saved under the condition of large-scaled nodes.
Description
Technical field
The present invention relates to the communications field, be specifically related to a kind of asynchronous neighbor discover method.
Background technology
In wireless network, wireless sensor network particularly, it is a difficult task that neighbours find.When node does not move, node can be paid the neighbours that disposable cost study exists, to optimize media interviews and route support.
Common neighbor node finds that agreement adopts master controller to coordinate.In these class methods, all node reports are to master controller, and master controller determines that nodal information comprises calculating its position, neighbor node and out of Memory, then circulates a notice of one by one to each node.The master controller mode needs lot of energy, and when node is on a grand scale, the present invention is distributed algorithm especially, adopts node to cooperate with each other, and does not have master controller to coordinate.
In the distributed algorithm of Baker and Ephremides, all nodes participate in 2 and take turns robin scheduling (schedule round robin).The hypothesis node can without error code from the condition of neighbours' receipt message.At each, take turns, each node distributes the information of a time slot for the neighbor node issuing its identity and found, and other nodes are in time slot monitoring in addition, and takes turns the interior information of neighbor nodes that can determine its neighbours and neighbours two.Yet this algorithm needs the global network time synchronized.Neighbours' discovery algorithm normally moves the earliest, and network behavior subsequently just can suppose that some mechanism is synchronous to strengthen, but, for operation at first, can not possess the basis of this hypothesis.Secondly, along with the network node scale increases, by safeguarding global synchronization, will become heavy resource consumption burden, same, adopt hardware means to safeguard global synchronization as GPS, also will bring extra huge cost.
Summary of the invention
The object of the present invention is to provide a kind of asynchronous distributed neighbor discovering method, comprise the following steps:
1) define a node x and a node y, wherein node x exceeds the signal to noise ratio requirement of node y, and wherein, described node x, node y all have the estimated value of neighbours' number, and unique identity identification information is arranged;
2) define a message m, this message m is for identifying the sender, and this message m is used for sending to receiving node, in distance d, if do not have other nodes to launch simultaneously, and receiving node success receipt message m;
3) when described node x finds node y, if node x and y are neighbours, node x receives message m at least one times from node y;
4) set of a given limited node, each node has a neighbor node finite aggregate.
In a preferred embodiment of the invention, node x, node y all have a kind of in network interface ID, MAC Address or CPU ID, and described network interface ID, MAC Address or CPU ID can be queried.
In a preferred embodiment of the invention, message m should comprise a long preambles.
In a preferred embodiment of the invention, in step 4), the set of a given limited node, each node has a neighbor node finite aggregate; If node x, y are neighbours, the finite aggregate of People Near Me relation comprises (x, y) and (y, x).
Especially, it should be noted that, finite aggregate is to possible node location and the statistical expection of clock skew
Beneficial effect of the present invention is:
Adopt distributed neighbor discovering method, do not need global synchronization, in the situation that the larger effectively saving resource of node.
Embodiment
For technological means, creation characteristic that the present invention is realized, reach purpose and effect is easy to understand, below in conjunction with embodiment, further set forth the present invention.
A kind of asynchronous neighbor discover method of the present invention comprises the following steps:
1) define a node x and a node y, wherein node x exceeds the signal to noise ratio requirement of node y; Wherein, node x, node y all have the estimated value of neighbours' number, and each node can broadcast or received at any time a broadcast, but node sending and receiving simultaneously.Under the hypothesis of only having neighbours to send, receive the broadcast zero defect that neighbours send.
In addition, each node has the unique identity identification information of himself, a node has its unique identification can have various ways to realize, a kind of as in network interface ID, MAC Address or CPU ID, and described network interface ID, MAC Address or CPU ID can be queried, and its uniqueness is guaranteed by equipment manufacturers.
Estimate that neighbours' number usually is required to meet in multihop network.Known region is disposed the sensor node number can be used for estimating neighbours' number, estimates also will consider the wireless coverage at the communication environments lower node of expectation.
2) define a message m, this message m is for identifying the sender, and this message m is used for sending to receiving node, in distance d, if do not have other nodes to launch simultaneously, and receiving node success receipt message m.Suppose that m can send to receiving node in time T m, if do not have other nodes to launch simultaneously, also at receiving terminal, do not transmit when collision in distance d, receiving node success receipt message m.
Consider and there is no global synchronization, message m should comprise a long preambles (Preamble), and therefore, this makes when starting and can receive m rather than middlely to insert that to receive be very important receiving node.
3) when described node x finds node y, if node x and y are neighbours, node x receives message m at least one times from node y.
4) set of a given limited node, each node has a neighbor node finite aggregate.
If x and y are neighbours, the finite aggregate of People Near Me relation comprises (x, y) and (y, x).
If
Wherein the right is to possible node location and the statistical expection of clock skew.If set the t long enough to complete first round scheduling, F (t)=1 arranged.
Suppose that each node has its time slot time, time slot is isometric but internodal time slot has a random offset.At each time slot, node state T, R} selects, { its probability is pT and pR, and meets: pT+pR=1 for T, the corresponding sending and receiving of R}.Between node, work alone, and its state is independently selected at each time slot.
Slot length is definite as follows, if determine to send at each time slot node, its sends W message m copy, the fixed value that wherein W is positive integer in network.Under time slot work (meaning system global synchronization), W > 1 optional.But, for non-time slot work, optimum W is undetermined.
Relevant with the transceiver technology adopted is peak transfer rate and transmitting-receiving, state switching time, suppose that the sending and receiving state can ignore switching time.If sending a required time of message m is Tm, WTm is for sending W the time that message copying is required.Therefore, a slot length is T=WTm.
At a receiving slot, node is opened the receiver decoding input signal, and the node processing input message is received with the message that determines whether zero defect.If have, can from message content, obtain the identification of sending node so, and if for this receiving node, before the information of sending node, not know, receiving node adds neighbor list by it.Any additional information in message can be stored and upgrade in receiving node.
Due to assumes synchronization, repeat to send message and be no advantage.Therefore, make W=1.If a fixed time period t is neighbours' discovery time.Comprise S=t/Tm time slot this period.
If h is the nodes that produces a successful receipt message m in a time slot, and parameter p T, W and N fix, and h is a stochastic variable so, and because the state of N-1 neighbours in this time slot of node is random, we will calculate E (h).
If nodes X has N-1 neighbours, due to N node disjoint work, the time slot of each node is independent of other nodes to be selected, and it can think a N * S form, and in form, each unit comprises an independent same distribution Bei Nuli (Bernoulli) stochastic variable.Look down (i.e. a time slot) from each row of form, receiving node and sending node are arranged.At each row, make receiving node corresponding in the neighbours that R and T are X and the quantity of sending node.H is the nodes time slot is heard by X arbitrarily, and obviously, h can not surpass 1, because surpass neighbours in a time slot, sends, and X can only hear collision.Therefore, h is a 0-1 stochastic variable, indicates whether that the neighbours of X send.Quantity can be expressed as the probability of hearing Y at a time slot X.X hears that the mathematic expectaion of message is:
Because all nodes are identical, expect to have symmetry: free in institute, node x listens and obtains other another sending node message, and this node is uniformly distributed in N-1 neighbours.Therefore, node x hears that the expectation of the number of times of node y is E (h)/(N-1), and wherein node y is the specific neighbor node of node x.
Upper table has been described the time slot analysis of neighbours' discovery algorithms.During N is capable each line description the status switch of a node.The S row are arranged, interval during corresponding algorithm.The output of each symbology independence identical test in table, because all nodes adopt identical p
TAnd p is arranged
R=1-p
T.. each row has been described the system mode of a time slot.The content of arbitrary row is obeyed a binomial distribution.If X is at state R, and in neighbours one is at state T, and from its N-1 neighbour one of nodes X successfully receives m.This is at row 5,6,8 and S (as long as no 2 nodes 4,5 ..., N-2 causes collision at state T).Notice that X hears neighbours N-1 at row S, but therefore this do not bring benefit in generation before.
By investigating X, the performance of individual node X can determine, each time slot is the test of once hearing neighbours.When the less and S of E (h)/(N-1) is very large, can thinks S time slot the lot of experiments of the little probability of success, and can hear that neighbours' number of times is approximately the Poisson stochastic variable of average E (h)/(N-1) to X.In this case, any node is found the probability that the probability of other node of its neighbours is the Poisson stochastic variable:
This equation is very crucial for analysis, and design can be controlled S by W, by pT, affects E (h).Select pT to maximize E (h), definition optimal value pT is:
This result is intuitively, can be expressed as
Node send in turn probability simulation.A sensor network generally is in multi-hop environment and moves, and considers the center of each Nodes at its disk (Disk), and this disk comprises random nodes and there is no need to equal
When all nodes adopt
Performance may be suboptimum, because with some node experience
Employing is greater than the 1/N probability and sends (thereby causing more multipacting to be hit), another node experience
Employing is less than the 1/N probability and sends (staying too many free timeslot), and wherein N is actual node quantity in a disk, and
Be average neighbours' number of network range, yet neighbours still can be found.
In view of h in time slot analysis is stochastic variable, because N-1 neighbours' state is random, under asynchronous condition, the skew of neighbours' time slot is the factor that another kind brings randomness.
Consider nodes X a time slot (0, T) monitor, wherein T=WTm.If X has N-1 neighbours.These neighbours also have the time slot of equal length T, but with the time slot of X, random offset are arranged.Making x is a N-1 offset vector, and (0, T), xi means the skew of the neighbor node i of corresponding X to be evenly distributed on the time interval.Analyze must consider neighbor node i two kinds of possible states " left side " state at (xi-T, xi) and " right side " state at (xi, xi+T).
Because slot length is T, can not receive 3 or more message at a single time slot.The example of hearing 2 message generations is as follows:
If W=4, and (0, T) interior is state R in the time interval to make X.Node Y sends at (2T/3, T/3), and X receives at the last mini time slot of Y.Z sends at (T/2,3T/2), and X receives at first mini time slot of Z.Other neighbours of all X (0, T) in state RR.
Unless W=1, the most of monitoring is normal monitoring.Minority is near monitoring., ignore monitor more here, monitor few the generation more because, the performance of A is likely underestimated in analysis like this.The labor of normal monitoring and near monitoring comprises several situations.
By observing the residing state TT of neighbours of X, TR, RT, or RR, claim these neighbours to send for two accordingly, left transmission, right transmission and without sending node, use D, L, R with (N-1)-(D+R+L) mean their quantity.
If D >=2, X will can not receive any information.If D=0, X can (but not be from left sending node or right sending node simultaneously; Monitoring more than will being like that) successfully receive m.During D=, two kinds of situations are arranged:
1) a kind of situation of left sending node be (0, T) peak excursion is arranged, with a, mean; The another kind of situation of left sending node is that second largest skew is arranged;
2) if a-a ' is greater than the length of a mini time slot, and do not have right sending node to disturb, so left sending node has skew a, and can successfully send message to X.In this case, right sending node is similar.
Even left sending node and right sending node are overlapping, one in them can successfully send to X.If the time from a ' to b comprises the whole message from left sending node, X can receive.If the width of b-a ' is a mini time slot or less, m can not be received.If b-a ' is two or more slot lengths, be assumed to be simply transmission success.
When D=1, X can only listen two sending nodes.If left transmission and right transmission reserve enough intervals, b-a is enough large, and X also hears two the transmission.In addition, the interval of a mini time slot is not enough, leading because X need to hear; A 2mini slot time will guarantee successfully to receive.On average, due to the skew of two sending nodes be evenly distributed on (0, T), 1.5, the interval that we think mini time slot or the reception that more pulls off coup, the short failure than this.
Part has before provided W and pT optimum under given N condition.For N, suggestion adopts whole network average
Yet, for the most of nodes in network, the actual neighbors number may be not equal to average neighbours' number.Can consider whether pT can use the value different from the value of expectation maximality energy under average case.
By emulation, point out, suitable parameter setting, the present invention can almost find all neighbours.
Above demonstration and described basic principle of the present invention and principal character and advantage of the present invention.The technical staff of the industry should understand; the present invention is not restricted to the described embodiments; that in above-described embodiment and specification, describes just illustrates principle of the present invention; without departing from the spirit and scope of the present invention; the present invention also has various changes and modifications, and these changes and improvements all fall in the claimed scope of the invention.The claimed scope of the present invention is defined by appending claims and equivalent thereof.
Claims (5)
1. an asynchronous neighbor discover method, is characterized in that, comprises the following steps:
1) define a node x and a node y, wherein node x exceeds the signal to noise ratio requirement of node y, and wherein, described node x, node y all have the estimated value of neighbours' number, and unique identity identification information is arranged;
2) define a message m, this message m is for identifying the sender, and this message m is used for sending to receiving node, in distance d, if do not have other nodes to launch simultaneously, and receiving node success receipt message m;
3) when described node x finds node y, if node x and y are neighbours, node x receives message m at least one times from node y;
4) set of a given limited node, each node has a neighbor node finite aggregate.
2. a kind of asynchronous neighbor discover method as claimed in claim 1, is characterized in that, described node x, node y all have a kind of in network interface ID, MAC Address or CPU ID, and described network interface ID, MAC Address or CPU ID can be queried.
3. a kind of asynchronous neighbor discover method as claimed in claim 1, is characterized in that, described message m should comprise a long preambles.
4. a kind of asynchronous neighbor discover method as claimed in claim 1, is characterized in that, in described step 4), and the set of a given limited node, each node has a neighbor node finite aggregate; If node x, y are neighbours, the finite aggregate of People Near Me relation comprises (x, y) and (y, x).
5. a kind of asynchronous neighbor discover method as claimed in claim 4, is characterized in that, described finite aggregate is to possible node location and the statistical expection of clock skew.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013103147982A CN103428821A (en) | 2013-07-24 | 2013-07-24 | Asynchronous neighbor discovery method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013103147982A CN103428821A (en) | 2013-07-24 | 2013-07-24 | Asynchronous neighbor discovery method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103428821A true CN103428821A (en) | 2013-12-04 |
Family
ID=49652805
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013103147982A Pending CN103428821A (en) | 2013-07-24 | 2013-07-24 | Asynchronous neighbor discovery method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103428821A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104185243A (en) * | 2014-09-16 | 2014-12-03 | 北京航空航天大学 | Method for rapidly finding neighbors |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110029204A (en) * | 2009-09-15 | 2011-03-23 | 에스케이 텔레콤주식회사 | Method for neighbor discovery for directional mac, and node therefor |
| CN103002525A (en) * | 2012-12-26 | 2013-03-27 | 上海交通大学 | Ad Hoc network asynchronous neighbor node scan method based on directional antennae |
-
2013
- 2013-07-24 CN CN2013103147982A patent/CN103428821A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110029204A (en) * | 2009-09-15 | 2011-03-23 | 에스케이 텔레콤주식회사 | Method for neighbor discovery for directional mac, and node therefor |
| CN103002525A (en) * | 2012-12-26 | 2013-03-27 | 上海交通大学 | Ad Hoc network asynchronous neighbor node scan method based on directional antennae |
Non-Patent Citations (1)
| Title |
|---|
| 费晓飞等: "无线传感器网邻居发现算法研究", 《微计算机信息》, 30 April 2009 (2009-04-30) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104185243A (en) * | 2014-09-16 | 2014-12-03 | 北京航空航天大学 | Method for rapidly finding neighbors |
| CN104185243B (en) * | 2014-09-16 | 2018-05-01 | 北京航空航天大学 | A kind of fast neighbor finds method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gallais et al. | Localized sensor area coverage with low communication overhead | |
| Gandham et al. | Distributed minimal time convergecast scheduling in wireless sensor networks | |
| Buratti et al. | Performance analysis of IEEE 802.15. 4 non beacon-enabled mode | |
| Gong et al. | Energy-efficient clustering in lossy wireless sensor networks | |
| US9723538B2 (en) | Systematic neighbor detection in shared media computer networks | |
| KR20180068848A (en) | A multi-hop networking protocol for wide-area energy harvesting sensor network deployments | |
| JP2010035068A (en) | Wireless network system | |
| JP2010522487A5 (en) | ||
| EP2925047B1 (en) | Method for transmitting and receiving data | |
| CN109121096B (en) | Method and apparatus for determining broadcast nodes in a mesh network | |
| Bhatia et al. | A distributed TDMA slot scheduling algorithm for spatially correlated contention in WSNs | |
| US20180124206A1 (en) | Communications network and method for transmitting data in a communications network | |
| Van Luu et al. | An efficient algorithm for scheduling sensor data collection through multi-path routing structures | |
| Zhao et al. | Scheduling data collection with dynamic traffic patterns in wireless sensor networks | |
| Huang et al. | SINR based Maximum Link Scheduling with Uniform Power in Wireless Sensor Networks. | |
| Kim et al. | A fast joining scheme based on channel quality for IEEE802. 15.4 e TSCH in severe interference environment | |
| CN108184251B (en) | Energy-saving scheduling method based on optimal DAG (demand oriented architecture) of life cycle | |
| Kakalou et al. | A reinforcement learning-based cognitive MAC protocol | |
| CN101755425B (en) | Method of managing transmission within a wireless communications network | |
| Radi et al. | Network initialization in low-power wireless networks: a comprehensive study | |
| Elhabyan et al. | Weighted tree based routing and clustering protocol for WSN | |
| CN104619005A (en) | Underwater wireless sensor network media medium access control method | |
| Xu et al. | Splitting tree algorithm for decentralized detection in sensor networks | |
| Kai et al. | Throughput analysis of CSMA wireless networks with finite offered-load | |
| CN103428821A (en) | Asynchronous neighbor discovery 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 | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20131204 |