CN110461018B - Opportunistic network routing forwarding method based on computable AP - Google Patents
Opportunistic network routing forwarding method based on computable AP Download PDFInfo
- Publication number
- CN110461018B CN110461018B CN201910756850.7A CN201910756850A CN110461018B CN 110461018 B CN110461018 B CN 110461018B CN 201910756850 A CN201910756850 A CN 201910756850A CN 110461018 B CN110461018 B CN 110461018B
- Authority
- CN
- China
- Prior art keywords
- node
- probability
- message
- nodes
- forwarding
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses an opportunistic network routing forwarding method based on a calculable AP (access point). AP nodes with calculable capability are introduced into an opportunistic network, the core function of the AP nodes is responsible for the routing decision of the whole network, the whole network is constructed into an uncertain graph inside the node according to the meeting probability among all the nodes, the network structure and the topology evolution condition are analyzed from the global perspective, and the forwarding probability of an end-to-end path is calculated, so that a routing decision table is formed for each node. Meanwhile, the AP node is also an intermediate forwarding node and provides message forwarding tasks for other mobile nodes. Each mobile node maintains a vector table for recording the probability of encountering it with all other nodes, uploads the probability vector table to the AP node when in the AP signal coverage area, and receives the routing decision table of the node from the AP; in the process of message forwarding, the node only judges whether to forward the message to the next hop according to the routing decision table, thereby greatly improving the efficiency of message forwarding.
Description
Technical Field
The invention relates to the field of short-distance wireless communication, in particular to an opportunistic network routing forwarding method based on a calculable AP.
Background
In recent years, with the development of a large number of portable intelligent devices with low cost and short-distance wireless communication capability, a novel mobile ad hoc network mode, namely an opportunity network (Opportunistic network), appears to provide Opportunistic communication and information sharing services for mobile devices, and the method is widely applied to specific fields such as space Networks, underwater Networks, handheld device networking and unmanned aerial vehicle networking, and has a good effect on solving the problems of access network traffic pressure overload, spectrum resource shortage and the like. The nodes in the opportunistic network have the characteristics of typical mobility, connection discontinuity, openness and the like, the short-distance wireless communication technology is generally adopted to realize communication, and when the two nodes are in the radio frequency signal coverage range, a wireless connection link is automatically established for data exchange. Under the condition, a network cannot ensure a stable topological structure, a complete end-to-end communication link is lacked at most of time, and the whole network communication is realized by adopting a storage-carrying-forwarding mechanism and depending on the meeting opportunity brought by node movement to cooperatively transmit messages. In the dynamic network environment with frequently changing topology structure, routing decision becomes a key element influencing network performance, so how to quickly decide to select the most appropriate next hop forwarding node in the message delivery process and how to reduce the overhead of each node for maintaining routing information becomes a key problem to be solved urgently in the current opportunistic network routing research.
The existing opportunistic routing forwarding method assumes that nodes in a network are completely distributed and in a peer-to-peer relationship, generally adopts a multi-copy routing forwarding strategy to guarantee the success rate and the minimum time delay of message forwarding, typically represents Epidemic infectious forwarding routing, the routing is similar to a propagation strategy of a flooding mechanism, and each node copies carried message copies to all encountered nodes. In order to solve the problem of low forwarding efficiency caused by excessive redundant message copies, a limited copy routing forwarding strategy based on node encounter probability or forwarding utility is provided, such as PROPHET and MaxProp. Each node in the network maintains a vector table of the probability of meeting with other nodes, when two nodes meet, the probability vector table is exchanged to obtain the probability from the meeting node to the target node, so as to judge whether the node is used as the next hop forwarding node or not, the method has the advantages of reducing the redundancy of network messages and improving the success rate of forwarding, and has the defects that the nodes need to exchange the probability vector table first to realize the forwarding of the messages when meeting each time, and a plurality of message header processing and vector table query operations need to be carried out, which consumes a large amount of time and causes the reduction of the forwarding efficiency, and in addition, the method can not obtain the communication path between end to end from the perspective of the global topology of the network. The forwarding method based on social relations or social networks, such as SMART, SSAR, Bubble Rap, SGBR, SimBet and SEBAR, estimates nodes by using social characteristics (such as centrality or community) among the nodes, divides the network into a plurality of communities, and then calculates the centrality of each node as the basis for forwarding messages in and among the communities, and has the defect that the centrality of the node is calculated only by local information.
In summary, each node in the existing message routing policy network needs to maintain a local routing table or a probability vector table, and storage of the routing table, calculation of the meeting probability between nodes, multiple query and matching of the routing table during message forwarding, and the like will cause each node to consume additional computing resources, storage resources, and network bandwidth resources.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides an opportunistic network routing forwarding method based on a calculable AP, which introduces an Access Point (AP) node with calculable capability into a network from the perspective of a network global topology structure, is relatively suitable for opportunistic networks with limited node resources, can effectively reduce resource overhead paid by each node for maintaining routing information, and improves the routing forwarding efficiency of the whole network.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows:
an opportunistic network routing forwarding method based on a computable AP, wherein n mobile nodes v in the opportunistic networkiAnd a fixed AP node; each mobile node maintains a record of its own identity andvector table of the meeting probability of all other nodes; the method comprises the following steps:
s1, when the mobile node enters the signal coverage of the AP node in the moving process, the mobile node uploads the encounter probability vector table to the AP node;
s2, the AP node constructs an uncertain graph representing the connection relation between the nodes of the whole network according to the historical encounter situation among all the nodes in the network according to the encounter probability vector table uploaded by all the mobile nodes;
s3, the AP node calculates and generates a global routing decision table by using the uncertain graph, further generates routing decision tables of all nodes, and issues the routing decision table to the mobile node while uploading the encounter probability vector table each time by the mobile node;
s4, after receiving the route decision list, the mobile node detects the ID of the target node of the message in sequence, finds the forwarding node set corresponding to the target node, and first judges whether the AP node is in the forwarding node set of the message:
if the AP node is in the set, the mobile node uploads the message to the AP node, and the AP node replaces the forwarding; if the AP node is not in the set, the mobile node carries the message to be forwarded to a proper opportunity;
s5, in the moving process, when the mobile node meets other nodes, judging whether the node belongs to the forwarding node set of the message, if so, forwarding the message to the meeting node, and if not, carrying the message continuously.
Furthermore, each mobile node maintains a vector table for recording the encounter probability of the mobile node with all other nodes in the network, and the calculation and maintenance process of the encounter probability specifically comprises the following steps:
(1) any pair of nodes (v) in a computing networki，vj) Average encounter interval therebetween:
wherein the content of the first and second substances,
denotes vi，vjIn the time interval of meeting at m and m +1 times,
representing a node vi，vjThe moment of the m-th encounter, where m is 0,
then
Denoted as node vi，vjN consecutive encounter interval time series;
(2) calculating two nodes vi，vjThe probability of meeting in time T:
(3) with the operation of the network, an n + 1-dimensional vector table q recording the probability of its encounter with all nodes in time T is generated in each mobile node by using the two calculation formulasi＝(qi,0,qi,1,qi,2,…,qi,n-1,qi,AP) Wherein q isi,i1 denotes the probability of the node itself, if node vi，vjWithout having historically met qi,j0 denotes a node vi，vjThe encounter probability of 0.
Further, in the steps S4 and S5, the mobile node detects the lifetime of the carried message in each timeslot, discards the message whose lifetime has expired from the buffer, and when receiving a new message forwarding request, discards the earliest received message in the buffer if the remaining space is insufficient, so as to make enough free space for storing the new message.
Further, in step S3, the specific step of the AP node generating the routing decision table for each node is:
s3-1, after the AP node receives the encounter probability vector tables uploaded by each mobile node, first summarizing the vector tables of all the mobile nodes to form an encounter probability matrix Q:
wherein q isi,i＝1；
S3-2, based on the meeting probability matrix Q, the AP node dynamically constructs an uncertainty graph G (V, E, p) describing the network global topology;
wherein V ═ { V ═ V0,v1,v2,…,vn-1AP is the set of all nodes, E ═ Ei,jIs the set of all edges, if two nodes vi，vjHistorically there is one edge e if a connection is establishedi,j，p:E→(0,1]Is a function of assigning a weight to each edge, the weight having a magnitude of the probability of two nodes meeting, i.e., p (e)i,j)＝qi,j；
S3-3, calculating the maximum path transmission probability between each pair of nodes in the network based on the uncertainty map G ═ V, E, p, to obtain an inter-node maximum path transmission probability matrix representing the global probability routing table:
s3-3-1, for slave node viTo vjN +1 trials are performed, first the path v is determinedi，v0，vjIf it is present, v is comparedi，vjAnd vi，v0，vjThe path probability of (1) is taken as the slave v with a higher probabilityiTo vjThe maximum probability path with the sequence number of the middle vertex not more than 0; adding a node v on the path1And so on, after n +1 times of comparison, the slave node v is finally obtainediTo vjThe maximum forwarding probability path of (a);
s3-3-2, node vi，vjThe intermediate nodes on the path are all taken from the set v0,v1,v2,…,vkProbability value of maximum forwarding probability path
The iterative formula of (a) is:
s3-3-3, calculating in increasing order using the modified Flouard algorithm according to the recursive formula above
The input to the algorithm is an encounter probability matrix Q whose elements represent the weights of the (V, E, p) edges of an undirected graph G, where:
then initializing: let matrix P(-1)Is an encounter probability matrix Q; then the triple nested loop generates the matrix P of the maximum forwarding probabilityn: the variable k of the first layer loop increases from 0 to n, wherein the variable k indicates that the maximum value of the number of path intermediate nodes does not exceed k, and a new matrix of (n +1) × (n +1) is created each time the first layer loop goes through
The variable i of the second layer loop is increased from 0 to n, wherein the variable i represents the number of the source node, the variable j of the third layer loop is also increased from 0 to n, wherein the variable j represents the number of the destination node, and the comparison is carried out every time the third layer loop goes through
And
the magnitude of the value, larger value being assigned
The return result of the final algorithm is the maximum forwarding probability matrix Pn：
S3-4, AP node is all nodes viGenerating a routing decision table:
firstly, a maximum transmission probability matrix P is usednMiddle extraction node viThen each node v in the row vectorjAs target nodes, respectively searching for the satisfied conditions in the corresponding column vectors
Node v ofkFrom node viMeet probability vector table qi＝(qi,0,qi,1,qi,2,…,qi,n-1,qi,AP) Middle search and node vkProbability of encounter qi,kIf the calculated integrated value is not equal to the predetermined value
Delta epsilon (0, 1) is the threshold value, then the node v iskAs a destination node vjNext hop forwarding node of (1); thus being node viEach target node in the row vector generates a candidate next hop forwarding node set, thereby forming a very light weight suitable for the node viThe routing decision table of (1).
Further, in step S3, after the AP node generates the routing decision table, it also generates a new routing message packet for each node in the network, where the message content only includes the routing decision table of the node, and then sends the routing message to each node; and when the node receives the routing message sent by the AP node, the routing decision table is stored or updated.
Further, in the step S5, when the node v is at the node viIn moving process with node vjAfter a connection link is established, a routing forwarding strategy of the message is started, and the specific steps are as follows:
s5-1, firstly, searching whether a message to be forwarded is carried in a cache or not, and if a message set M to be forwarded exists, sequencing the messages in the set M from small to large according to a receiving timestamp;
s5-2, circularly detecting each message M belongs to M, if the message detection is completed, finishing the algorithm, otherwise, obtaining the destination node v of the message M from the head field of the message MdGo to step S5-3;
s5-3, according to the target node vdLookup routing decision table TiAcquiring the next hop sending node set of the message mdJudging node vjWhether or not in setdIn, if vj∈setdIndicating node vjIs the appropriate next hop forwarding node for message m, the message m is forwarded to node vjOtherwise, it is not forwarded, and goes to step S5-2.
Adopt the produced beneficial effect of above-mentioned technical scheme to lie in:
(1) the invention provides an opportunistic network routing method based on a calculable AP (Access Point), which introduces an AP (Access Point) node with calculable capability into a network, wherein the AP node has stronger computing capability and storage capability relative to a mobile node in the network and has continuous power supply equipment. In the invention, the core function of the AP node is responsible for the routing decision of the whole network, the whole network is constructed into an uncertain graph in the AP node according to the meeting probability among all nodes, the network structure and the topology evolution condition are analyzed from the global perspective, and the forwarding probability of the path between end to end is calculated, thereby forming a routing decision table for each node. Meanwhile, the AP node is also an intermediate forwarding node and provides message forwarding tasks for other mobile nodes.
(2) The invention realizes the separation of route calculation and message forwarding. In the invention, each node only maintains a vector table to record the encountering probability with other nodes, uploads the probability vector table to the AP node when in the AP signal coverage range, and receives the routing decision table of the node from the AP. In the process of message forwarding, the node only judges whether to forward the message to the next hop according to the routing decision table, thereby greatly improving the efficiency of message forwarding. In the method, because each node unloads the task of route calculation to the AP with stronger calculation capability, the separation of route calculation and message forwarding is realized, the node resource expense is greatly saved, and the life cycle of the network is increased.
Drawings
FIG. 1 is an upload diagram of an encounter probability vector table;
FIG. 2 is a schematic diagram of the forwarding of messages through an AP node;
FIG. 3 is a schematic diagram of the forwarding of an inter-node message (D2D);
FIG. 4 is a schematic diagram of a sequence of encounter intervals of node pairs;
fig. 5 is a schematic diagram of the global topology of the network.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
The invention discloses an opportunistic network routing forwarding method based on a calculable AP.
The opportunistic network model is as follows: there are n mobile nodes and a fixed AP (Access Point) node in the whole opportunistic network. Each mobile node has a globally unique ID identification viThe method comprises the steps that a buffer space B with the same size and limited capacity and a vector table for recording the probability of encountering other nodes are maintained, any node can generate messages with the same size, the target node of the message can be any other node in a network, each message has a unique ID (identity) Mj, and except the content and the ID of the message, fields used for message control in a Bundle comprise the IDs of a sending node and the target node, a timestamp t generated by the message, a message life cycle TTL, the number nc of copies of the message and the like. The AP node has stronger computing capability and storage capability relative to the mobile node in the network, and hasThe continuous power supply equipment and the AP node can also forward the message, can be used as an intermediate node for message forwarding, and provide message forwarding tasks for other mobile nodes.
The method of the invention comprises the following steps:
s1, when the mobile node enters the signal coverage of the AP node in the moving process (i.e. establishes connection with the AP node), the mobile node uploads its encounter probability vector table to the AP node, as shown in fig. 1;
s2, the AP node constructs an uncertain graph representing the connection relation between the nodes of the whole network according to the historical encounter situation among all the nodes in the network according to the encounter probability vector table uploaded by all the mobile nodes;
s3, the AP node analyzes the network structure and topology evolution situation from the global perspective by using the uncertain graph, calculates and generates a global routing decision table, further generates routing decision tables of all nodes, and issues the routing decision table to the mobile node while uploading the meeting probability vector table each time;
s4, after receiving the route decision list, the mobile node detects the ID of the target node of the message in sequence, finds the forwarding node set corresponding to the target node, and first judges whether the AP node is in the forwarding node set of the message:
if the AP node is in the set, the mobile node uploads the message to the AP node, and the AP node replaces the forwarding, as shown in FIG. 2; if the AP node is not in the set, the mobile node carries the message to be forwarded to a proper opportunity;
s5, in the moving process, when the mobile node meets other nodes (including the mobile node and AP nodes, the same below), it is determined whether the node belongs to the forwarding node set of the message, if it belongs, the message is forwarded to the meeting node, if it does not belong, the message is carried continuously, as shown in fig. 3.
The mobile node detects the survival time of the carried message in each time slot, discards the message with the overtime survival time from the cache, and discards the message received earliest in the cache if the residual space is insufficient when receiving a new message forwarding request so as to vacate enough free space to store the new message.
In the invention, each mobile node maintains a vector table for recording the encountering probability of the mobile node and all other nodes in the network, and the calculating and maintaining process of the encountering probability specifically comprises the following steps:
(1) any pair of nodes (v) in a computing networki，vj) The average encounter interval between them, as shown in fig. 4:
wherein the content of the first and second substances,
denotes vi，vjIn the time interval of meeting at m and m +1 times,
representing a node vi，vjThe moment of the m-th encounter, where m is 0,
then
Denoted as node vi，vjN consecutive encounter interval time series;
the node average encounter interval is an important index for measuring the encounter probability among nodes. A shorter average encounter interval between nodes indicates a higher probability of their meeting over a period of time. Therefore, the internal node measures the meeting probability between nodes by recording the average meeting interval of the node and other nodes.
(2) Calculating two nodes vi，vjThe probability of meeting in time T:
in a mobile opportunistic network, any pairNode vi，vjThe average meeting interval therebetween is subject to the parameter of
The probability density function of the exponential distribution of (1) is:
thus, node vi，vjThe encounter probability in time T can be calculated using the cumulative density distribution function as shown in equation (2):
(3) with the operation of the network, an n + 1-dimensional vector table q recording the probability of its encounter with all nodes in time T is generated inside each mobile node by using formula (2)i＝(qi,0,qi,1,qi,2,…,qi,n-1,qi,AP) Wherein q isi,i1 denotes the probability of the node itself, if node vi，vjWithout having historically met qi,j0 denotes a node vi，vjThe encounter probability of 0.
The calculation process is simple and requires less calculation resources. When the mobile node enters the signal coverage range of the AP node, the maintained probability vector table is automatically transmitted to the AP node after a communication link is established with the AP node.
The AP node has stronger computing power and storage power besides basic message forwarding capability, and can construct an uncertain graph according to the historical encounter condition of each node in the network, and then calculate the maximum forwarding probability of the path between end to end by utilizing a Flouard algorithm, thereby generating a routing decision table for each node. In step S3, the specific steps of the AP node generating the route decision table for each node are:
s3-1, after the AP node receives the encounter probability vector tables uploaded by each mobile node, first summarizing the vector tables of all the mobile nodes to form an encounter probability matrix Q:
wherein q isi,i＝1；
S3-2, based on the encounter probability matrix Q, the AP node dynamically constructs an uncertainty graph G ═ V, E, p describing the network global topology, as shown in fig. 5;
wherein V ═ { V ═ V0,v1,v2,…,vn-1AP is the set of all nodes, E ═ Ei,jIs the set of all edges, if two nodes vi，vjHistorically there is one edge e if a connection is establishedi,j，p:E→(0,1]Is a function of assigning a weight to each edge, the weight having a magnitude of the probability of two nodes meeting, i.e., p (e)i,j)＝qi,j；
The invention utilizes the strong computing capacity and the storage capacity of the AP node to form a routing decision table for each node, when two nodes meet, the main task of the routing strategy is to compare the forwarding probability from the two nodes to a message target node, namely to judge which node has high path forwarding probability to the target node, and then to select the node with high probability as the next hop node of the message. Therefore, the routing problem is converted into the problem of solving the forwarding probability of the path between end to end in the global topological structure chart, and because a plurality of paths possibly exist between end to end, one path with the highest probability is used as the forwarding probability between end to end. Suppose node vi，vjThere is a path betweeni,j:vi,vi+1,vi+2,…vjThe forwarding probability of this path is then the product of the probabilities of encountering each pair of nodes, i.e.
In the formula (3), since the forwarding probability of the path between end to end is a multiplicative relationship, the maximum probability forwarding path between two nodes cannot be directly obtained in the network global topology by using the existing shortest path algorithm. But due to the weight value q of the edgei,j∈(0,1]So if node vi，vjThere are multiple paths between them, then there must be one pathi,j:vi,vi+1,vi+2,…vjSo that P (path)i,j)∈(0,1]The path with the largest value, i.e. the highest forwarding probability, exists. In the invention, the maximum probability path among the nodes of the whole network is solved by improving the Flouard algorithm.
S3-3, calculating the maximum path transmission probability between each pair of nodes in the network based on the uncertainty map G ═ V, E, p, to obtain an inter-node maximum path transmission probability matrix representing the global probability routing table:
s3-3-1, for slave node viTo vjN +1 trials are performed, first the path v is determinedi，v0，vjIf it is present, v is comparedi，vjAnd vi，v0，vjThe path probability of (1) is taken as the slave v with a higher probabilityiTo vjThe maximum probability path with the sequence number of the middle vertex not more than 0; adding a node v on the path1And so on, after n +1 times of comparison, the slave node v is finally obtainediTo vjThe maximum forwarding probability path of (a);
s3-3-2, firstly defining an iterative formula for solving the probability value of the maximum forwarding probability path between the node pairs, and setting
Representing a node vi，vjThe intermediate nodes on the path are all taken from the set v0,v1,v2,…,vkThe probability value of the maximum forwarding probability path. When k is-1, the slave node viTo node vjWithout any intermediate nodes, whichThe sample path has only one, so
Whereby the definition can be recursive
For any path, all intermediate nodes belong to the set v0,v1,v2,…,vn-1AP }, matrix
The matrix of the maximum forwarding probability is given. V to facilitate algorithmic descriptionnRepresenting an AP node. S3-3-3, calculating in increasing order using the modified Flouard algorithm according to the recursive formula above
The input to the algorithm is an encounter probability matrix Q whose elements represent the weights of the (V, E, p) edges of an undirected graph G, where:
then initializing: let matrix P(-1)Is an encounter probability matrix Q; then the triple nested loop generates the matrix P of the maximum forwarding probabilityn: the variable k of the first layer loop increases from 0 to n, wherein the variable k indicates that the maximum value of the number of path intermediate nodes does not exceed k, and a new matrix of (n +1) × (n +1) is created each time the first layer loop goes through
The variable i of the second layer loop is increased from 0 to n, wherein the variable i represents the number of the source node, the variable j of the third layer loop is also increased from 0 to n, wherein the variable j represents the number of the destination node, and the comparison is carried out every time the third layer loop goes through
And
the magnitude of the value, larger value being assigned
The return result of the final algorithm is the maximum forwarding probability matrix Pn：
Proposition 1: matrix PnIs a maximum forwarding probability matrix, i.e.
Is a node vi、vjMaximum forwarding probability of the path in between.
Is provided with
Representing a node vi、vjThe intermediate nodes on the path are all taken from the set v0,v1,v2,…,vkThe probability of the maximum forwarding path of v, node vi、vjHas a maximum forwarding probability of
The correctness of the proposition can be inferred by the following analysis. Because of the slave node viTo node vjAll from the set v0,v1,v2,…,vkThe maximum forwarding probability path has two cases, if v passeskThen, then
If not pass vkThen, then
Solving the most probable path according to the algorithm requirements, then
It is known that
From this can be derived
By analogy, all paths between nodes can be obtained after being considered
So the matrix P returned by the algorithmnIs a maximum forwarding probability matrix, i.e.
Is a node vi、vjMaximum forwarding probability of the path in between.
Proposition 2: node viTo node vjThe forwarding probability of any two nodes on the path of maximum forwarding probability is also the largest.
At node viTo node vjHas two nodes v on the maximum forwarding probability pathk，vdDividing the path into viTo vk(denoted as path1), vkTo vd(denoted as path2), vdTo vj(marked as path3) in three segments, where the probability of the path is p (path) is p (path1) x p (path2) x p (path3), where path1 and path3 are the most probable paths between nodes, and the hypothesis is that v is the most probable path between nodeskTo vdThere are other paths (denoted as path4) with greater probability, satisfying the condition
Then p (path 1). times.p (path 2). times.p (path3) < p (path 1). times.p (path 4). times.p (path3) will be obtained, which contradicts the premise of proposition 2, soAssume an error, proposition 2 holds.
Calculating the maximum transmission probability matrix P between end to endnThen, the AP node can obtain a global probability routing table, if the table is transmitted to each node in the network as a routing decision table, the interior of each node occupies a larger storage space, and if the probability value between each pair of nodes occupies 4 bytes, the routing table occupies at least 4 x (n +1)2And the node queries a routing table at least twice during each message forwarding, searches the probability value from the node to a message target node for the first time, searches the probability value from a encountered node to the message target node for the second time, and then compares the two probability values to determine whether the encountered node is used as the next hop forwarding node of the message, which causes the message forwarding efficiency to be low and wastes a large amount of message forwarding time when each forwarding opportunity occurs.
In order to improve the searching efficiency of the routing table when the node forwards the message and save the space of the routing table occupied by the cache as much as possible, the invention provides a lightweight and personalized routing decision table generation method. Because the main decision-making basis of message forwarding in the opportunistic network is that the next hop forwarding node should have higher transmission probability to the target node of the message, the node with higher forwarding probability to the target node should be selected as the forwarding node in the generation of the routing decision table, and the maximum transmission probability matrix P is used as the forwarding nodenRow vector of
Obtain node viForwarding probability to all other target nodes, from column vector
Can obtain the network all nodes to the target node vjThe forwarding probability of (2). For node viA certain target node v in a row vectorjThen the most suitable next hopForwarding node vkShould satisfy the condition
Wherein v isk∈{v0,v1,v2,…,vn-1AP, i.e. the selection of the next hop forwarding node is to find the set of nodes satisfying the condition from the column vector. However, since there is a certain probability of connection between nodes in the opportunistic network, there is a possibility that a certain node exists to the target node v in the set of nodes that satisfy the conditionjWith high forwarding probability, but associated with node viHas a low probability of encountering, in which case the node is difficult to serve as the node viThe next hop forwarding node (because two nodes are difficult to meet), and therefore, the nodes meeting the condition need to be further screened by combining the meeting probability among the nodes.
S3-4, AP node is all nodes viGenerating a routing decision table:
based on the above analysis, the present invention implements a lightweight, personalized routing decision table generation algorithm: AP node is viGenerating a routing decision table by first generating a maximum transmission probability matrix PnMiddle extraction node viThen each node v in the row vectorjAs target nodes, respectively searching for the satisfied conditions in the corresponding column vectors
Node v ofkFrom node viMeet probability vector table qi＝(qi,0,qi,1,qi,2,…,qi,n-1,qi,AP) Middle search and node vkProbability of encounter qi,kIf the calculated integrated value is not equal to the predetermined value
Delta epsilon (0, 1) is the threshold value, then the node v iskAs a destination node vjThe next hop forwarding node. This would be node viEach target node in the row vector generates a candidate next hop forwarding nodeAggregate to form a very lightweight suitable node viAs shown in table 1.
Table 1 mobile node viRouting decision table of
Message forwarding strategy based on routing decision table: in step S3, after the AP node generates the routing decision table, it also generates a new routing message packet for each node in the network, where the message is a packet of a specific type, and the message content only includes the routing decision table of the node, and then sends the routing message to each node; and when the node receives the routing message sent by the AP node, the routing decision table is stored or updated.
When node viIn moving process with node vjAfter a connection link is established, a routing forwarding strategy of the message is started, and the specific steps are as follows:
s5-1, firstly, searching whether a message to be forwarded is carried in a cache or not, and if a message set M to be forwarded exists, sequencing the messages in the set M from small to large according to a receiving timestamp;
s5-2, circularly detecting each message M belongs to M, if the message detection is completed, finishing the algorithm, otherwise, obtaining the destination node v of the message M from the head field of the message MdGo to step S5-3;
s5-3, according to the target node vdLookup routing decision table TiAcquiring the next hop sending node set of the message mdJudging node vjWhether or not in setdIn, if vj∈setdIndicating node vjIs the appropriate next hop forwarding node for message m, the message m is forwarded to node vjOtherwise, it is not forwarded, and goes to step S5-2.
The nodes of the algorithm can judge whether the message should be forwarded or not according to the forwarding node set only by searching a local routing decision table without exchanging routing information with the encountering nodes, so that the forwarding efficiency of the message is greatly improved, the network flow is reduced, and the network pressure is relieved. The time complexity is at most O (| M | × n × s), where s represents the average number of nodes of the forwarding node set in the routing decision table. With the increase of the number of nodes in the network and the increase of the number of forwarding nodes, the time complexity of the algorithm is increased, but the successful forwarding of the message is facilitated.
Claims (6)
1. An opportunistic network routing forwarding method based on a computable AP is characterized in that:
there are n mobile nodes v in the opportunistic networkiAnd a fixed AP node; each mobile node maintains a vector table for recording the probability of encountering all other nodes;
the method comprises the following steps:
s1, when the mobile node enters the signal coverage of the AP node in the moving process, the mobile node uploads the encounter probability vector table to the AP node;
s2, the AP node constructs an uncertain graph representing the connection relation between the nodes of the whole network according to the historical encounter situation among all the nodes in the network according to the encounter probability vector table uploaded by all the mobile nodes;
s3, the AP node calculates and generates a global routing decision table by using the uncertain graph, further generates routing decision tables of all nodes, and issues the routing decision table to the mobile node while uploading the encounter probability vector table each time by the mobile node;
s4, after receiving the route decision list, the mobile node detects the ID of the target node of the message in sequence, finds the forwarding node set corresponding to the target node, and first judges whether the AP node is in the forwarding node set of the message:
if the AP node is in the set, the mobile node uploads the message to the AP node, and the AP node replaces the forwarding; if the AP node is not in the set, the mobile node carries the message to be forwarded to a proper opportunity;
s5, in the moving process, when the mobile node meets other nodes, judging whether the node belongs to the forwarding node set of the message, if so, forwarding the message to the meeting node, and if not, carrying the message continuously.
2. The opportunistic network routing forwarding method based on a computable AP according to claim 1, wherein: each mobile node maintains a vector table for recording the encounter probability of the mobile node and all other nodes in the network, and the calculation and maintenance process of the encounter probability specifically comprises the following steps:
(1) any pair of nodes (v) in a computing networki，vj) Average encounter interval therebetween:
wherein the content of the first and second substances,
denotes vi，vjIn the time interval of meeting at m and m +1 times,
representing a node vi，vjThe moment of the m-th encounter, where m is 0,
then
Denoted as node vi，vjN consecutive encounter interval time series;
(2) calculating two nodes vi，vjThe probability of meeting in time T:
any pair of nodes vi，vjAverage meeting in betweenInterval all obeys a parameter of
The probability density function of the exponential distribution of (1) is:
node vi，vjThe probability of meeting over time T is:
(3) with the operation of the network, an n + 1-dimensional vector table q recording the probability of its encounter with all nodes in time T is generated in each mobile node by using the two calculation formulasi＝(qi,0,qi,1,qi,2,…,qi,n-1,qi,AP) Wherein q isi,i1 denotes the probability of the node itself, if node vi，vjWithout having historically met qi,j0 denotes a node vi，vjThe encounter probability of 0.
3. The opportunistic network routing forwarding method based on a computable AP according to claim 1, wherein: in the steps S4 and S5, the mobile node detects the time-to-live of the carried message in each timeslot, discards the message whose time-to-live has been over-timed out from the buffer, and when a new message forwarding request is received, if the remaining space is insufficient, discards the message received earliest in the buffer to make enough free space for storing the new message.
4. The opportunistic network routing forwarding method based on a computable AP according to claim 1, wherein: in step S3, the specific steps of the AP node generating the routing decision table for each node are:
s3-1, after the AP node receives the encounter probability vector tables uploaded by each mobile node, first summarizing the vector tables of all the mobile nodes to form an encounter probability matrix Q:
wherein q isi,i＝1；
S3-2, based on the meeting probability matrix Q, the AP node dynamically constructs an uncertain graph G (V, E, p) describing a network global topology;
wherein V ═ { V ═ V0,v1,v2,…,vn-1AP is the set of all nodes, E ═ Ei,jIs the set of all edges, if two nodes vi，vjHistorically there is one edge e if a connection is establishedi,j，p:E→(0,1]Is a function of assigning a weight to each edge, the weight having a magnitude of the probability of two nodes meeting, i.e., p (e)i,j)＝qi,j；
S3-3, calculating the maximum path transmission probability between each pair of nodes in the network based on the uncertainty map G ═ V, E, p, and obtaining an inter-node maximum path transmission probability matrix representing the global probability routing table:
s3-3-1, for slave node viTo vjN +1 trials are performed, first the path v is determinedi，v0，vjIf it is present, v is comparedi，vjAnd vi，v0，vjThe path probability of (1) is taken as the slave v with a higher probabilityiTo vjThe maximum probability path with the sequence number of the middle vertex not more than 0; adding a node v on the path1And so on, after n +1 times of comparison, the slave node v is finally obtainediTo vjThe maximum forwarding probability path of (a);
s3-3-2, node vi，vjThe intermediate nodes on the path are all taken from the set v0,v1,v2,…,vkSummary of maximum Forwarding probability paths of }Value of rate
The iterative formula of (a) is:
s3-3-3, calculating in increasing order using the modified Flouard algorithm according to the recursive formula above
The input to the algorithm is an encounter probability matrix Q whose elements represent the weights of the edges of the uncertainty map G ═ V, E, p, where:
then initializing: let matrix P(-1)Is an encounter probability matrix Q; then the triple nested loop generates the matrix P of the maximum forwarding probabilityn: the variable k of the first layer loop increases from 0 to n, wherein the variable k indicates that the maximum value of the number of path intermediate nodes does not exceed k, and a new matrix of (n +1) × (n +1) is created each time the first layer loop goes through
The variable i of the second layer loop is increased from 0 to n, wherein the variable i represents the number of the source node, the variable j of the third layer loop is also increased from 0 to n, wherein the variable j represents the number of the destination node, and the comparison is carried out every time the third layer loop goes through
And
the magnitude of the value, larger value being assigned
The return result of the final algorithm is the maximum forwarding probability matrix Pn：
S3-4, AP node is all nodes viGenerating a routing decision table:
firstly, a maximum transmission probability matrix P is usednMiddle extraction node viThen each node v in the row vectorjAs target nodes, respectively searching for the satisfied conditions in the corresponding column vectors
Node v ofkFrom node viMeet probability vector table qi＝(qi,0,qi,1,qi,2,…,qi,n-1,qi,AP) Middle search and node vkProbability of encounter qi,kIf the calculated integrated value is not equal to the predetermined value
Is a threshold value, then the node v is connectedkAs a destination node vjNext hop forwarding node of (1); thus being node viEach target node in the row vector generates a candidate next hop forwarding node set, thereby forming a very light weight suitable for the node viThe routing decision table of (1).
5. The opportunistic network routing forwarding method based on a computable AP according to claim 1, wherein: in step S3, after the AP node generates the routing decision table, it also generates a new routing message packet for each node in the network, where the message content only includes the routing decision table of the node, and then sends the routing message to each node; and when the node receives the routing message sent by the AP node, the routing decision table is stored or updated.
6. The opportunistic network route forwarding method based on a computable AP according to claim 1 or 5, wherein: in the step S5, when the node viIn moving process with node vjAfter a connection link is established, a routing forwarding strategy of the message is started, and the specific steps are as follows:
s5-1, firstly, searching whether a message to be forwarded is carried in a cache or not, and if a message set M to be forwarded exists, sequencing the messages in the set M from small to large according to a receiving timestamp;
s5-2, circularly detecting each message M belongs to M, if the message detection is completed, finishing the algorithm, otherwise, obtaining the destination node v of the message M from the head field of the message MdGo to step S5-3;
s5-3, according to the target node vdLookup routing decision table TiAcquiring the next hop sending node set of the message mdJudging node vjWhether or not in setdIn, if vj∈setdIndicating node vjIs the appropriate next hop forwarding node for message m, the message m is forwarded to node vjOtherwise, it is not forwarded, and goes to step S5-2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910756850.7A CN110461018B (en) | 2019-08-16 | 2019-08-16 | Opportunistic network routing forwarding method based on computable AP |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910756850.7A CN110461018B (en) | 2019-08-16 | 2019-08-16 | Opportunistic network routing forwarding method based on computable AP |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110461018A CN110461018A (en) | 2019-11-15 |
| CN110461018B true CN110461018B (en) | 2021-03-19 |
Family
ID=68487066
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910756850.7A Active CN110461018B (en) | 2019-08-16 | 2019-08-16 | Opportunistic network routing forwarding method based on computable AP |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110461018B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113098770B (en) * | 2020-01-08 | 2024-04-16 | 华为技术有限公司 | Message sending method, route table item generating method, device and storage medium |
| CN113518009B (en) * | 2021-07-12 | 2023-07-21 | 山东建筑大学 | Underwater acoustic communication route forwarding method based on uncertain graph |
| CN114079969B (en) * | 2021-08-16 | 2023-11-28 | 珠海市杰理科技股份有限公司 | Data transmission method and device, readable storage medium and node equipment |
| CN114726434B (en) * | 2022-03-18 | 2023-09-19 | 电子科技大学 | Millisecond-level rapid path-finding method suitable for large-scale optical network |
| CN115134374B (en) * | 2022-06-29 | 2023-08-01 | 中国人民解放军国防科技大学 | Directory data synchronization method and device for opportunistic network |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101977226B (en) * | 2010-10-28 | 2012-12-05 | 北京航空航天大学 | Novel opportunity network data transmission method |
| CN104661275B (en) * | 2014-12-11 | 2018-01-30 | 北京邮电大学 | Data transmission method in a kind of opportunistic network |
| GB2548798A (en) * | 2016-03-17 | 2017-10-04 | Virtuosys Ltd | Wireless communication unit and method for sharing delay tolerant content |
| CN108391300B (en) * | 2018-03-15 | 2021-03-26 | 东北大学 | Credibility-based credible routing algorithm in opportunity network |
-
2019
- 2019-08-16 CN CN201910756850.7A patent/CN110461018B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110461018A (en) | 2019-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110461018B (en) | Opportunistic network routing forwarding method based on computable AP | |
| CN107333312B (en) | Route selection method in opportunistic network | |
| Tabatabaei et al. | A novel energy-aware clustering method via Lion Pride Optimizer Algorithm (LPO) and fuzzy logic in wireless sensor networks (WSNs) | |
| CN114423061B (en) | Wireless route optimization method based on attention mechanism and deep reinforcement learning | |
| CN111510956B (en) | Hybrid routing method based on clustering and reinforcement learning and ocean communication system | |
| Gnana Prakasi et al. | Decision tree based routing protocol (DTRP) for reliable path in MANET | |
| Tang | dao 2: Overcoming overall storage overflow in intermittently connected sensor networks | |
| Yin et al. | Reinforcement Learning‐Based Routing Algorithm in Satellite‐Terrestrial Integrated Networks | |
| Khodaparas et al. | A multi criteria cooperative caching scheme for internet of things | |
| CN108632785B (en) | Ant colony self-adaptive Internet of vehicles routing method based on link quality | |
| CN113727408A (en) | Unmanned aerial vehicle ad hoc network improved AODV routing method based on speed and energy perception | |
| CN110417572B (en) | Method for predicting message transfer node based on target node meeting probability | |
| CN116782330A (en) | Method and medium for selecting path in SD-UANET by considering load and link change | |
| Chen et al. | An efficient neural network-based next-hop selection strategy for multi-hop VANETs | |
| CN1731763A (en) | Distributed Qos multicast routing method based on information entropy in Ad Hoc network | |
| CN110121185A (en) | A kind of power distribution communication network route optimization method | |
| KR102308799B1 (en) | Method for selecting forwarding path based on learning medium access control layer collisions in internet of things networks, recording medium and device for performing the method | |
| Hua et al. | Path selection algorithms in homogeneous mobile ad hoc networks | |
| Wu et al. | On-demand Intelligent Routing Algorithms for the Deterministic Networks | |
| CN113708953A (en) | Underwater acoustic sensor network anti-damage method based on node importance balance | |
| Du et al. | A data collection approach based on mobile sinks for heterogeneous sensor networks | |
| CN111432447A (en) | Message transmission method, system, terminal device and storage medium | |
| CN113778620B (en) | Large-scale cluster storage system architecture based on cooperation of multiple SDN controllers and software and hardware | |
| Li et al. | A MANET accessing Internet routing algorithm based on dynamic gateway adaptive selection | |
| CN103346965A (en) | Light multicast route method based on coding subgraph optimized coding cost |
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 |