CN101262428A - Potential field routing method in sparse ad-hoc network based on multi-target optimization - Google Patents

Potential field routing method in sparse ad-hoc network based on multi-target optimization Download PDF

Info

Publication number
CN101262428A
CN101262428A CNA2008100442980A CN200810044298A CN101262428A CN 101262428 A CN101262428 A CN 101262428A CN A2008100442980 A CNA2008100442980 A CN A2008100442980A CN 200810044298 A CN200810044298 A CN 200810044298A CN 101262428 A CN101262428 A CN 101262428A
Authority
CN
China
Prior art keywords
node
route
link
network
potential field
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
Application number
CNA2008100442980A
Other languages
Chinese (zh)
Other versions
CN101262428B (en
Inventor
江虹
李磊民
赵海龙
黄玉清
刘雷
韩宾
张晓琴
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southwest University of Science and Technology
Original Assignee
Southwest University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Southwest University of Science and Technology filed Critical Southwest University of Science and Technology
Priority to CN2008100442980A priority Critical patent/CN101262428B/en
Publication of CN101262428A publication Critical patent/CN101262428A/en
Application granted granted Critical
Publication of CN101262428B publication Critical patent/CN101262428B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Mobile Radio Communication Systems (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention relates to the technical field of ad-hoc self-organizing network routing. In a sparse ad-hoc, time for building an effective routing between source/destination nodes is very long or even the effective routing cannot be built due to reasons that node interval distance is larger than effective communication distance thereof or an obstacle and a certain node consumes energy and withdraws from the network, etc. The special network condition often takes place in a war field, tunnel detection, emergent risk-removing and other occasions. Therefore, how to ensure effective information forwarding in the sparse ad-hoc network is an issue to be urgently solved. The invention provides a potential field routing algorithm based on multi-objective optimization in the sparse ad-hoc network to realize mutual communication between source/destination nodes in the manner of optimum 'ferrying node' to transmit data. The invention provides a novel 'link + node' routing algorithm based on link delay, bandwidth, packet loss ratio, node residual energy, moving direction, destination node joint probability and other objectives. Meanwhile, the introduction of method of potential field can effectively control propagation direction of routing searching and grouping, reduce information load in the network and improve arrival rate of data.

Description

In the sparse ad-hoc network based on the potential field routing algorithm of multiple-objection optimization
Technical field
The present invention is a kind of novel routing algorithm that is applied to sparse ad-hoc network.Belong to radio communication and information dissemination technology field.
Background technology
The Ad-hoc network is a kind of new network that occurs along with the wireless communication technology fast development, is a kind of mobile network who does not have wired infrastructure to support.Node in the network is made of mobile host, has the network topology structure of dynamic change, the limited main frame energy, characteristics such as limited wireless communication bandwidth.Because the network coverage of each node is limited,, then need to assist to set up route by the via node between them if the node of two apart from each others will communicate.Route in the Ad-hoc network mainly comprises priori formula and on-demand two big classes, and the work of this two classes route all has a prerequisite: source end and destination must have complete end to end route in information interaction in the time period.If interstitial content is limited in the whole network, the very difficult assurance of the feasible sparse ad-hoc network that constitutes is end-to-end to have complete route in information interaction in the time period.At this moment, common ad-hoc Routing Protocol can't be in sparse operational environment operate as normal, a kind of Routing Protocol that is fit to sparse operational environment must be arranged.
For the ad-hoc network that connects fully, also fewer for the research of sparse ad-hoc network route technology, present research mainly contains two big classes: quasi-protocol and forwarding quasi-protocol flood.The quasi-protocol that floods mainly is Epidemic and mutation thereof.At document Amin Vahdat and David Becker.Epidemic routing for partially connected ad hoc networks, Technical ReportCS-2000-06, Duke University, among the July 2000, a plurality of copies of message spread between the node of random motion, and finally arrive destination, its advantage is not need any knowledge of network, protocol capabilities is that network size and message number that agreement can be fit to are subjected to very big restriction with the enough condition greatly of Internet resources such as buffer memory or bandwidth.Document Mirco Musolesi and Cecilia Mascolo, Controlledepidemic style dissemination middleware for mobile ad hoc networks, Technicalreport, CS-UCL Reasearch Note, Mar.2006 improves Epidemic, try hard to improve the forwards success rate, and reduce taking Internet resources.
The thinking of transmitting quasi-protocol is to use the controlled movement node to realize forwarding to information.At document Wenrui Zhao, Mostafa Ammar, etc., Controlling the mobility of multipledata transport ferries in a delay-tolerant network, In Proceedings of the 24thIEEE Conference on Computer Communications (INFOCOM ' 5), among the Mar.2005, utilize one or more nodes on predetermined path, to give certain node delivery message.Document Anders Lindgren and Kaustubh S.Phanse, Evaluation of queueing policies andforwarding strategies for routing in intermittently connected networks, InProceedings of the First International Conference on Communication SystemSoftware and Middleware (COMSWARE ' 6), Jan.2006 has proposed according to the node characteristics of motion, adopts the method for probability theory to determine the most possible node of transmitting data.Patent EP20040019966Tan Kun, Zhang Qian, Zhu Wen wu, Metaspace:communication middleware forpartially connected mobile ad hoc networks proposes to utilize the mobile node buffer memory to transmit message, and the connection of adopting middleware mechanism to finish between separated network is transmitted.Open day at patent CN1741501A happy, Li Zhongcheng, Liu Min etc., in the radio network group relationship routing system that part connects and the method, proposition use group concerns to come the path label feature, and as the guidance foundation of routing procedure.
Summary of the invention
The effective route that the objective of the invention is to solve in sparse ad-hoc network is set up, and realizes the reliable transmission of data.Because in sparse ad-hoc network, the distance between the node is far away, it is big to set up complete route difficulty, and the present invention comes information-carrying by the ferry-boat node.Ferry-boat transmission comprises three phases: route hunt phase, ferry-boat node determine the stage with link, the implementation phase of the ferry-boat.Route utilizes potential field method effectively to control the direction of propagation that route searches grouping hunt phase, guarantees it all the time towards the transmission of the direction of destination node (data center), finishes and the searching of near as far as possible all the imperfect routes of destination node.The present invention realizes that by the classification mode of selecting the superior uniting of ferry-boat node and communication link is definite.At first, certain link potential energy minimum node remains the energy with node, and moving direction is connected probability and determines candidate's set of node of ferrying with destination node, route initiation node is a foundation with Link Selector Parameter and node selection parameter then, determines optimum ferry-boat node from the candidate ferries set of node.Implement delivery control by selected ferry-boat node implementation phase of ferry-boat, and finish information interaction based on selected imperfect link.Main functional modules as shown in Figure 1.
Among Fig. 1, ferry-boat starts finishes initial work, starts incomplete link road by searching function when detecting when route surpasses certain threshold value break period; In route searches, search all initiate with route that node links to each other but with the near as far as possible incomplete link road of destination node by, in this process, the direction of propagation of routing packets is controlled by potential field.Node selects potential energy minimum node that module refers to receive that route searches grouping to select parameter to finish the ferry selection of set of node of candidate according to node.Node and link optimized choice module show the way and initiate node is finished node and link according to node selection parameter and link parameters optimization optimization selection by searching, after certain node transmission route searches grouping, received the response message of some node in stand-by period in route, then obtained optimum " ferry-boat node+link " combination according to predetermined assessment algorithm.On the contrary, if this node is not received any route replies in finite time, then self effectively transmits as ferry-boat node realization data.Whether node responds its route of receiving and searches grouping be subjected to the ferrying restriction of node optimization target and self potential energy size in the network.Purport of the present invention is in sparse ad-hoc network, between source/order node under the no complete available routing condition, by the ferry-boat node of optimum and the reliable communication between transmission link combination realization source/order node.
1, control and objective optimization algorithm
(1) " potential field " control
In whole network, suppose to exist a data center (generally being to transfer control desk), its position is known for other node in the network, it receives other node and transmits and next miscellaneous service data, carries out the observing and controlling analysis for the user.Other mobile node has certain intelligent, can detect the barrier in the certain distance scope.Introducing " potential field " is to initiate the direction of propagation that route that node sends searches grouping (RSD) in order to control route, makes routing packets have the direction cognitive ability in network.Can utilize the GPS location technology to determine the distance of each nodal distance data center distance, with the parameter of this distance as potential energy p_field.As 1 gesture point, the distance that node arrives data center is far away more with the data center position, node is described more away from 1 gesture point, and then the potential energy of node present position is high more.As shown in Figure 2, position L I+1, L i, L 3, L 2, L 1Pairing potential energy reduces gradually, i.e. U I+1>U i>U 3>U 2>U 1Because the relative potential field position of data center is generally constant during certain subtask, in the netinit stage, each node can obtain the definite position of data center in the network.Therefore, after this node will no longer need with data center exchange position information, can calculate the potential energy size of self present position, to satisfy the requirement that should reduce the unnecessary information exchange in the sparse ad-hoc network as far as possible.Secondly, owing to the nodal distance data center that the potential energy value is more little is near more, its probability of finishing route foundation is just big more.Owing to,, give its corresponding weighted value χ for embodying the influence degree that potential parameter is selected the ferry-boat node according to the optimum ferry-boat of multiple-objection optimization policy selection node Pe
The potential field control chart arrives at RSD before the node of certain bar link potential energy value minimum referring to shown in Figure 2, if certain node is received RSD, judges whether and receive according to the potential energy information in the packet.If this RSD from than the little nodes of locations of self potential energy, then directly abandons this grouping.Promptly in this process route to search mobile that grouping (RSD) embodies be unidirectional, as the N that describes among Fig. 2 I+1→ N i→ N 3→ N 2→ N 1If there is communication barrier thing in the potential energy minimum node detection on this link to the place ahead, node is provided with circuitous mark P _ RSD for this RSD, then in the circuitous jumping figure max_pass scope of maximum, allow potential energy to receive this P_RSD than own big node, make route search the grouping barrier that can make a circulation, as the N that describes among Fig. 2 1→ N 2Utilize formula (1) to calculate each node potential energy value, the compensation coefficient when coefficient θ is converted into plane rectangular coordinates for GPS information.Potential field searches the control flow of grouping referring to shown in Figure 3 to route.
U i ( N i ) = θ × ξ 0 ( ( x i - x 0 ) 2 - ( y i - y 0 ) 2 ) L ( N i = ( x i , y i ) ) - - - ( 1 )
(2) routing optimality parametric description
Unite selection in order to be implemented in the sparse ad-hoc network " node+link ", introduced the multiple-objection optimization strategy, " link-quality " and " node local information " as a plurality of parameters optimization, and selected the difference of influence degree according to each parameter to route, determine the size of respective weights value.
1. the node of ferrying is selected parameter
When selecting " optimum ferry-boat node ", as parameters optimization, be respectively " the intermediate node direction of motion " with 3 local informations of node, " the intermediate node residue energy " and " intermediate node can connect probability with data center ".When selecting " optimum ferry-boat node ", need to consider the factor of node motion direction, if certain movement direction of nodes is opposite with the data center position, then this point should be got rid of outside the range of choice of " the optimum node of ferrying ".As Fig. 4, the intermediate node moving direction
Figure A20081004429800073
For (180 °, 180 °].Secondly because data center as 1 gesture point, is positioned at the tail end of network configuration, so intermediate node and the angle ω of data center's line in the standard straight angle coordinate system ° be (90 °, 90 °), the differential seat angle that defines the two is:
Figure A20081004429800074
In order to make the direction of motion of selecting node, require to satisfy β ° of minimum, promptly as far as possible near data center's place direction
Figure A20081004429800075
Minimum value.
Figure A20081004429800081
For " intermediate node can connect probability with data center " parameter, add up to obtain by intermediate node.Behind network startup, node detects whether there is self route to data center with certain frequency broadcast probe bag, thereby obtains interior node of certain timing statistics and the successful connection probability c_rate between the data center.Probability is big more, and it is big more to illustrate that node and data center set up the chance of communication link.But when there are data interaction in certain node and data center, then suspend the transmission of detection packet, node is used with the data interaction of data center and is replaced detection packet to be connected probability statistics.At node residue energy parameter, can suppose that the node residue energy is the monotonic decreasing function of operating time, make node can confirm the residue energy value left_energy of self according to the function of time.If the node residue energy is lower than certain threshold value, node might be finished ferry-boat data task, thereby node can not be chosen as optimum ferry-boat node.Residue energy of node is sufficient more, and the probability that node is chosen as the ferry-boat node is just big more.In sum, with the selection foundation of tlv triple (min (β °), max (c_rate), max (left_energy)) as " optimum ferry-boat node ".Secondly, under the different application environment, " the intermediate node direction of motion ", " the intermediate node residue energy " and " intermediate node can connect probability with destination node ", three parameters had different weighing factors to the selection of " optimum ferry-boat node ", therefore, can utilize corresponding weighted value χ d, χ e, χ cDecide each parameter to determining the influence degree of ferry-boat node.
2. communication link is selected parameter
Route is initiated link selection between node and " the optimum node of ferrying " with chain-circuit time delay, packet loss, and bandwidth is as parameters optimization.The desirable principle that " time delay is more little, and packet loss is more little, and bandwidth is big more " selected as communication link.Set (Delay i j, Drop i j, Bandwidth i j) as the parameters optimization group of node i to certain bar link between the node j.t iThe expression node i is to the time delay of node i+1; d iThe expression node i is to the packet loss of node i+1; b iThe expression node i is to the bandwidth of node i+1, as formula (3,4,5).
Delay i j = Σ l = i j - 1 tl - - - ( 3 )
Drop i j = 1 - Π κ = i ( 1 - d κ ) - - - ( 4 )
Bandwidth i j = min ( b i , b i + 1 , . . . , b j - 1 ) - - - ( 5 )
Because sparse ad-hoc network particularity, in transfer of data, the influence degree of chain-circuit time delay decreases.Most important characteristic is embodied in the integrality and the reliability of transmission data, i.e. packet loss and bandwidth.In like manner, use weighted value χ De, χ Dr, χ bDetermine the time delay in sparse ad-hoc network, packet loss, bandwidth is to the influence degree of selected optimum link.And, can calculate the numerical value that embodies link-quality according to above-mentioned condition and formula (6), thereby obtain effective estimation, for the selection of " link " in this routing algorithm provides important reference single-hop or multi-hop link performance.G lBe link quality estimation function, Link lThe quality of expression link l.Under actual conditions, should there be an ideal value E (Link in the communication link quality that is determined by this three parameter l), the link-quality optimum value that the expression parametric joint is determined.
Link l ( Delay , Drop , Bandwidth ) = G l ( χ de * Delay , χ dr * Drop , χ b * Bandwidth ) ( χ de + χ dr + χ b ) + χ pe + χ d + χ e + χ r = 1 - - - ( 6 )
2, route implementing process design
(1) node+link searches flow process
Whether setup parameter " exists communication barrier thing ", and " potential energy ", " the intermediate node direction of motion ", " the intermediate node residue energy " is crucial factor of influence, the propagation of RSD in network influenced by these four factors mainly.After node was received RSD, according to its potential energy value size, whether RSD existed circuitous sign, and whether node is that the information such as potential energy minimum node on this link judge that how handling this route searches grouping.When the potential energy minimum node on certain link is received RSD, there is communication barrier thing if detect, then for RSD is provided with circuitous sign, allows potential energy to receive P_RSD, thereby reach the purpose of forcing out barrier greater than the node of oneself.When P_RSD arrives certain during the node of cut-through thing, this node carries out reset operation to P_RSD, and continues to transmit RSD.Secondly, the moving direction of node should place, as close as possible data center position direction.If the moving direction of node and data center's place deviation of directivity have surpassed 90 °, then node is not initiated node toward route and is replied route replies grouping (RRD).In sum, the restrictive condition of crucial factor of influence " potential energy ", " whether having communication barrier thing " and " the intermediate node direction of motion " is (8,9).
Figure A20081004429800092
Figure A20081004429800093
The 3rd factor of influence, " the intermediate node residue energy " also plays crucial effects.The probability that intermediate node is connected with data center has material impact to its ferry-boat energy resource consumption, connects the least residue energy value E that probability is determining that this node is allowed as " optimum ferry-boat node " Min ζ iNode and data center to be connected probability big more, the least residue energy value E of permission Min ζ iJust more little, the probability that node becomes " optimum ferry-boat node " alternative objects is just big more.Formula (10) has been represented node as the needed residue energy of ferry-boat node condition, and wherein the ξ function is node is connected probability c_rate with data center a subtraction function.
E ( left _ energy ) > ( E min ζi = σ × ξ ( c _ rate ) ) - - - ( 10 )
The definition of each data field is as shown in table 1 among the RSD:
The formal definition of table 1RSD
Figure A20081004429800103
Wherein, bandwidth, packet loss, potential energy value and route record are brought in constant renewal in communication process.Delay is initialized as route and initiates the time that node sends RSD, and MAX-hops represents the maximum hop count restriction that routing packets can be transmitted, and MAX-time represents the maximum life span of this routing packets in network, and node is deleted this grouping after this time.In wireless network, the link generation interrupt probability of being set up through multi-hop number more is just big more.Rationally control the propagation jumping figure of RSD, can effectively improve the stability of the route of setting up.After certain node is received RSD, whether the satisfy condition restriction of (8) of decision node itself, satisfy condition and then upgrade bandwidth in the routing packets, packet loss, parameter values such as potential energy value, and will oneself add among the route record Route-log, transmit this RSD to other node then, upwards a hop node sends a confirmation simultaneously, and expression also exists than it more near the communication node of data center.After a last hop node is transmitted RSD, in the time-out time of setting, do not receive the affirmation information that other node is replied, node promptly is the potential energy minimum node on this link.The potential energy minimum node is then taked different processing modes according to whether detecting barrier.If do not detect barrier, then change Rule of judgment (9,10) over to.The restriction of (9,10) if movement direction of nodes satisfies condition then generates a route replies RRD, otherwise, abandoning the RSD that receives, potential energy minimum node processing flow chart is as shown in Figure 5.The formal definition of RRD is as shown in table 2.The RRD that produces arrives route along the Route-log record among the corresponding RSD and initiates node.On the contrary, if the potential energy minimum node detection to communication barrier thing, then for this RSD is provided with circuitous mark P _ RSD, allows to receive P_RSD at the circuitous jumping figure scope max_pass internal potential energy of maximum than own big node.In back haul, when certain node does not detect barrier, and satisfy condition simultaneously (9,10), then finished the transmission of routing packets, otherwise, think that the route on this link is set up failure.The process chart of P_RSD is seen shown in Figure 6.
The formal definition of table 2RRD
Figure A20081004429800111
Wherein, Link-quality represents that route initiates node and reply link-quality between the node, by link-quality tlv triple (Delay, Bandwidth, Drop-rate) expression.Wherein, the occurrence of Delay obtains from the pairing domain of definition of RSD, and promptly route is initiated the time that node sends RSD.Route is initiated node and in the stand-by period, is received the RRD that other node is replied in the route replies of stipulating, then according to the multiple-objection optimization principle, selects optimum " path+node " combination and finishes searching of this route.Secondly, when not receiving any RRD as if route initiation node, route is initiated node oneself becomes " ferry-boat node ", move to target node position, when it enters the communication range of destination node, the data of self storing are sent to destination node, finish the ferry-boat of data message from then on.
Description of drawings
Fig. 1 primary circuit routing high-level schematic functional block diagram
Fig. 2 potential field searches packet propagation direction control schematic diagram to route
Fig. 3 potential field control flow chart
Fig. 4 movement direction of nodes deviation schematic diagram
The workflow diagram of Fig. 5 potential energy minimum node
Fig. 6 node is received the decision flow chart of P_RSD
Fig. 7 route is initiated the workflow diagram of node
Route is set up the process schematic diagram in the sparse ad-hoc network of Fig. 8
Fig. 9 route is initiated node self the professional schematic diagram data of ferrying
Embodiment
Route is initiated node behind broadcasting RSD, waits for the route replies of other node.Route is initiated node according to whether receiving RRD, takes two kinds of scheduling processing schemes shown in Figure 7.
1. route is initiated the processing scheme that node is received RRD:
For this situation, route is initiated node according to the Delay value among current time and the RRD, calculates the chain-circuit time delay of replying between node and the route initiation node.At this moment, after route initiation node is received many groups " node and link corresponding " combination,, calculate the quality estimated value of whole piece link, and then consider node selection problem according to formula (6).In actual environment, because the selection parameter " the network intermediate node direction of motion " of node, there are an ideal value in " the intermediate node residue energy ", " intermediate node can connect probability with destination node " and potential energy value p_field, be defined as E (β °) respectively, E (left_energy), E (c_rate) and E (p_field).Such as, when not considering the actual environment factor affecting, movement direction of nodes and data center's place orientation angles deviation ideal value are 0, node residue energy ideal value is the maximum energy that node can hold, it is 1 that node can connect the probability ideal value with destination node, and the ideal value of potential energy value is 0.Try to achieve the numerical value that embodies node quality and link quality according to the minimum deviation quadratic sum algorithm of belt restraining condition, promptly the joint estimate of each parameter departs from the size of ideal value.Select the optimum combination of " ferry-boat node+link " according to numerical values recited.At first, according to the time delay value that the Delay among the RRD obtains, G in (6) formula lRepresent with formula (11).Then, according to the minimum deviation quadratic sum algorithm structure link of belt restraining condition and the net assessment function LNE of node, as the formula (12).
Link l ( Delay , Drop , Bandwidth ) = χ b * Bandwidth α - χ de * Delay β - χ dr * Drop γ ( χ de + χ dr + χ b ) + χ pe + χ d + χ e + χ r = 1 - - - ( 11 )
LNE ( l ) = [ Link l - E ( Link l ) ] 2 + Σ i = pe , d , e , r χ i × [ ψ ( i ) - E ( i ) ] 2 - - - ( 12 )
In the formula (12), variable i is illustrated in link l and goes up the information that produces the RRD node, comprises the potential energy value, the direction of motion, and the residue energy, with the probability that is connected of data center, the ideal value of E (i) expression above-mentioned parameter.Try to achieve the value of the pairing net assessment function LNE of each " node and link " combination.Therefrom select minimum value min (LNE) then, its corresponding variable is exactly " ferry-boat node+path " optimum combination that this routing algorithm searches.After route and node were determined, route was initiated node and is sent data message and then directly do not arrive data center, but arrives the ferry-boat node of selecting by the link of selecting, and node receives storing data information.In the motion process of ferry-boat node, if successfully connect with data center, then node is given data center with the data forwarding of storage, assists to finish the task of whole transfer of data.Data message then is similar to " ferry-boat " mode, arrives data center through middle " ferry-boat node ".
Below in conjunction with Fig. 8 this situation is further described.In the sparse ad-hoc network that Fig. 8 describes, route is initiated node has data need send to data center, data center is too far away apart from route initiation nodal point separation, and does not have enough intermediate nodes to help to set up complete available route between them but as shown in the figure.For this situation, route is initiated node S broadcasting RSD, and it flows to as shown in the figure.Node N2 is after N3 receives RSD, through judging that satisfy condition (8) then upgrade this RSD of relevant information continuation forwarding among the RSD.Through the propagation of RSD, four potential energy minimum node N4 have been found in the network, N6, N7, N8.But because node N7 does not satisfy the requirement of the direction of motion, this node abandons the RSD that receives, node N4 does not satisfy the restrictive condition of the least residue energy, abandons the RSD that receives equally.So, through the screening of crucial factor of influence, having only node N6, N8 satisfies condition, and initiates node to route and replys RRD.Wherein, the RRD that N6 produces has two to arrive the S nodes, but the path of process inequality.
{(S-N 2-N 6)|(N 6)};
{(S-N 3-N 6)|(N 6)};
{(S-N 5-N 8)|(N 8)}
After RRD arrived node S, according to the net assessment function LNE of path and node, the numerical value of the RRD domain of dependence and the weight of different parameters, route were initiated node from 3 top combinations, select optimum combination.Such as, because { (S-N 5-N 8) | (N 8) link-quality is relatively poor, { (S-N then 3-N 6) | (N 6) be the optimum combination of selecting, route is initiated node with the path { (S-N of data by selecting 3-N 6) arriving node N6, node N6 receives storing data information.In the motion process of ferry-boat node N6, be connected as if successfully setting up with data center, then node is given data center with the data forwarding of storage, assists to finish the task of whole transfer of data.
2. route is initiated the processing scheme that node is not received RRD:
The reason that causes this situation has: (1) but route initiates not have in the communication range of node other node.(2) all nodes do not satisfy the restrictive condition of crucial factor of influence after receiving RSD, do not initiate node to route and reply RRD.(3) because link-quality is too poor, the RRD that node is replied is lost halfway, does not successfully arrive route and initiates node.
At above-mentioned situation, route is initiated node oneself becomes " ferry-boat node ", and as shown in Figure 9, route is initiated node and changed the direction of motion, the direction motion to data center.In this process, constantly detect the effective route that arrives data center, when route appearance and stable, route is initiated node business datum is transmitted to data center.Route is initiated node and is finished the foundation of route and the transmission of business datum by sacrificing self limited energy.

Claims (7)

1, a kind of novel routing algorithm that is applied in the sparse ad-hoc network is characterized in that: the route that potential field method and multi-objective optimization algorithm are applied in the ad-hoc network is set up.Whole process is as follows: route searches and is grouped in potential field method control down, propagates to destination node (data center) direction; Local information with node is selected parameters optimization as node, satisfies the node of this parameter limit condition and initiates node answer route replies to route; Route is initiated node and is taked different scheduling strategies to realize and the communicating by letter of data center according to the route replies of whether receiving other node.
2, according to the described potential field routing algorithm based on multiple-objection optimization of claim 1, it is characterized in that: comprise a plurality of operational modules, route searches the module duration of work and controlled by potential field method, guarantees that route searches the correct transmission of grouping; Node selects module to be subjected to node to select parameter to control, and selects candidate's set of node of ferrying; It is foundation that node and link optimized choice module are selected parameter and Link Selector Parameter with node, guarantees to select the superiority of " node+link " combination; The ferry-boat node of implementing to refer to ferry receives after route initiates the business information of node, carries out the ferry-boat transformation task to the business information destination.
3, according to the described potential field routing algorithm based on multiple-objection optimization of claim 1, it is characterized in that: the propagation that potential field method control makes route search and is grouped in the network has the direction cognitive ability.With the data center position is 1 gesture point, and the potential energy value that position far away more, range data center has is big more.Route searches grouping and propagates along the direction that the potential energy value reduces, and can effectively control the range of scatter that route searches grouping, reduces the load of network.
4, according to the described potential field routing algorithm of claim 1 based on multiple-objection optimization, it is characterized in that: at first, local information with the network intermediate node is selected parameter as node, i.e. " the intermediate node direction of motion ", " the intermediate node residue energy " and " intermediate node can connect probability with data center " three parameters are determined candidate's set of node of ferrying jointly.Next, route initiates node and optimum is ferried internodal link selection with chain-circuit time delay, packet loss, bandwidth is as parameters optimization, and " time delay is more little, and packet loss is more little, and bandwidth is big more " is as the desirable principle of communication link selection.
5, according to the described potential field routing algorithm of claim 1, it is characterized in that: have only the node that satisfies following two kinds of situations receiving that just initiating node to route after route searches grouping replys route replies based on multiple-objection optimization.First kind of situation: the potential energy minimum node on certain bar link is just initiated node to route and is replied route replies under the restrictive condition that satisfies node selection parameter; Second kind of situation: when there is communication barrier thing in the potential energy minimum movement direction of nodes on certain bar link, satisfy other certain node that node is selected the parameter limit condition if exist in the circuitous jumping figure scope of maximum, then this node is initiated node to route and is replied route replies.
6, according to the described potential field routing algorithm based on multiple-objection optimization of claim 1, it is characterized in that: route is initiated node according to the route replies of whether receiving that at the appointed time other node is replied in the network, takes different processing modes.Route is initiated node if receive the route replies of other node, then the minimum deviation quadratic sum algorithm according to the belt restraining condition makes up link and node net assessment function, select the optimum combination of " node+link ", the route of utilizing this combination to finish between route initiation node and the data center is set up and data interaction.On the contrary, route is initiated node if do not receive any route replies, then self starts ferry-boat mechanism, moves and then finish the transmission of data to the data center position.
7, according to the described potential field routing algorithm of claim 1 based on multiple-objection optimization, it is characterized in that: when certain sparse ad-hoc network is divided into a plurality of zones far away that are separated by, in the time of can't setting up communication link each other, in each zone, successively utilize the described routing algorithm of claim 1 to select " node+link " optimum combination, adopt the control mode that is fit to zones of different to finish the intercommunication mutually of interregional node.
CN2008100442980A 2008-04-24 2008-04-24 Potential field routing method in sparse ad-hoc network based on multi-target optimization Expired - Fee Related CN101262428B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2008100442980A CN101262428B (en) 2008-04-24 2008-04-24 Potential field routing method in sparse ad-hoc network based on multi-target optimization

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2008100442980A CN101262428B (en) 2008-04-24 2008-04-24 Potential field routing method in sparse ad-hoc network based on multi-target optimization

Publications (2)

Publication Number Publication Date
CN101262428A true CN101262428A (en) 2008-09-10
CN101262428B CN101262428B (en) 2011-06-01

Family

ID=39962641

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2008100442980A Expired - Fee Related CN101262428B (en) 2008-04-24 2008-04-24 Potential field routing method in sparse ad-hoc network based on multi-target optimization

Country Status (1)

Country Link
CN (1) CN101262428B (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101710895B (en) * 2009-11-30 2012-11-14 中国人民解放军信息工程大学 Method for obtaining multiple next hop routing, device and router
CN102984200A (en) * 2012-09-13 2013-03-20 大连理工大学 Method applicable for scene with multiple sparse and dense vehicular ad hoc networks (VANETs)
CN103501524A (en) * 2013-09-18 2014-01-08 南京信息工程大学 Sparse self-organized monitoring network
CN104702500A (en) * 2015-03-20 2015-06-10 中国矿业大学 Distributed robust routing method of fully mechanized working face wireless network
CN105451293A (en) * 2014-07-17 2016-03-30 株式会社理光 Forwarding method in wireless network and method and device for determining forwarding strategy
CN106465053A (en) * 2014-06-25 2017-02-22 微软技术许可有限责任公司 Courier network
CN107277884A (en) * 2017-05-18 2017-10-20 中国科学院计算技术研究所 A kind of static link dispatching method for 6TiSCH multi-hop wireless networks
CN108184246A (en) * 2018-03-06 2018-06-19 中国船舶重工集团公司第七二四研究所 Mobile ad-hoc network routing optimization method based on position and service quality
CN112326925A (en) * 2020-10-18 2021-02-05 西南科技大学 Method for evaluating stability of tunnel surrounding rock based on matter element analysis

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7027426B2 (en) * 2002-08-05 2006-04-11 Harris Corporation Multi-channel mobile ad hoc network
CN1849597A (en) * 2003-06-06 2006-10-18 网状网络公司 Method to provide a measure of link reliability to a routing protocol in an ad hoc wireless network
CN1599487A (en) * 2003-09-19 2005-03-23 皇家飞利浦电子股份有限公司 Routing selecting method for radio communication system and mobile terminal executing the method
CN100566292C (en) * 2005-09-22 2009-12-02 中国科学院计算技术研究所 Radio network group relationship routing system and method that part connects

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101710895B (en) * 2009-11-30 2012-11-14 中国人民解放军信息工程大学 Method for obtaining multiple next hop routing, device and router
CN102984200A (en) * 2012-09-13 2013-03-20 大连理工大学 Method applicable for scene with multiple sparse and dense vehicular ad hoc networks (VANETs)
CN102984200B (en) * 2012-09-13 2015-07-15 大连理工大学 Method applicable for scene with multiple sparse and dense vehicular ad hoc networks (VANETs)
CN103501524A (en) * 2013-09-18 2014-01-08 南京信息工程大学 Sparse self-organized monitoring network
CN103501524B (en) * 2013-09-18 2016-03-23 南京信息工程大学 A kind of sparse self-organizing monitor network
CN106465053A (en) * 2014-06-25 2017-02-22 微软技术许可有限责任公司 Courier network
CN105451293B (en) * 2014-07-17 2018-11-13 株式会社理光 Retransmission method in wireless network, the method and apparatus for determining forwarding strategy
CN105451293A (en) * 2014-07-17 2016-03-30 株式会社理光 Forwarding method in wireless network and method and device for determining forwarding strategy
CN104702500A (en) * 2015-03-20 2015-06-10 中国矿业大学 Distributed robust routing method of fully mechanized working face wireless network
CN104702500B (en) * 2015-03-20 2017-11-14 中国矿业大学 The Robust distributed method for routing of fully-mechanized mining working wireless network
CN107277884A (en) * 2017-05-18 2017-10-20 中国科学院计算技术研究所 A kind of static link dispatching method for 6TiSCH multi-hop wireless networks
CN107277884B (en) * 2017-05-18 2019-11-29 中国科学院计算技术研究所 A kind of static link dispatching method for 6TiSCH multi-hop wireless network
CN108184246A (en) * 2018-03-06 2018-06-19 中国船舶重工集团公司第七二四研究所 Mobile ad-hoc network routing optimization method based on position and service quality
CN108184246B (en) * 2018-03-06 2021-05-28 中国船舶重工集团公司第七二四研究所 Mobile self-organizing network route optimization method based on position and service quality
CN112326925A (en) * 2020-10-18 2021-02-05 西南科技大学 Method for evaluating stability of tunnel surrounding rock based on matter element analysis

Also Published As

Publication number Publication date
CN101262428B (en) 2011-06-01

Similar Documents

Publication Publication Date Title
CN101262428B (en) Potential field routing method in sparse ad-hoc network based on multi-target optimization
CN1906898B (en) Method and system for efficient routing in ad hoc networks
Nayyar et al. Comprehensive analysis of routing protocols surrounding underwater sensor networks (UWSNs)
Malhotra et al. A comprehensive review on recent advancements in routing protocols for flying ad hoc networks
KR20190017906A (en) Routing data packets in wireless networks with directional transmissions
Sutagundar et al. Location aware event driven multipath routing in Wireless Sensor Networks: Agent based approach
Kumar Performance evaluation of energy consumption in MANET
CN104602313A (en) Environment-adapted routing method of wireless sensor network for maritime search and rescue
Sirmollo et al. Mobility‐Aware Routing Algorithm for Mobile Ad Hoc Networks
CN104869604A (en) RPL-based layer type route establishment method of wireless sensor network
Kumar et al. Fuzzy logic based efficient multipath routing for mobile adhoc networks
GhasemAghaei et al. Ant colony-based many-to-one sensory data routing in wireless sensor networks
Al Rubeaai et al. 3D real-time routing protocol with tunable parameters for wireless sensor networks
Prabu et al. Energy efficient routing in MANET through edge node selection using ESPR algorithm
Al-Salti et al. An efficient and reliable grid-based routing protocol for UWSNs by exploiting minimum hop count
Abdou et al. Priority levels based multi-hop broadcasting method for vehicular ad hoc networks
Meghanathan A location prediction-based reactive routing protocol to minimize the number of route discoveries and hop count per path in mobile ad hoc networks
Mamoun A new reliable routing algorithm for MANET
Saidi et al. Opportunistic routing in wireless sensors networks
CN102761950A (en) Self-adaption self-optimizing QoS (quality of service) transmission method for heterogeneous network
Tan et al. A distributed and dynamic data gathering protocol for sensor networks
Chelbi et al. Event-driven wireless sensor networks based on consensus
Junnarkar et al. Novel quality of service (qos) improvement routing protocol for manet using ant colony optimization
Dsouza et al. Energy and congestion aware simple ant routing algorithm for MANET
Nehra et al. Routing with load balancing in ad hoc network: A mobile agent approach

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20110601

Termination date: 20140424