CN103236984B - A kind of efficient infection route-caching management strategy method in delay-tolerant network - Google Patents

Abstract

The invention discloses a kind of efficient infection route-caching management strategy method in delay network.Route storage resource is infected high for delay network, cache congestion problem frequently, according to true DTN (delay tolerant network, be called for short DTN) in network node motor pattern predictable, the historical record met by statistics and analysis node, obtain message in nodal cache with the distributed intelligence in this node contact node cluster frequently, assessment is for each message, this node can make its ability infected further in a network, message more dense relative to this Node distribution on select probability abandons, reduce the loss of the infection effectiveness that node causes because of packet loss, the message making distribution the most sparse obtains more repeater-transmitter meeting, that improves message send the rate of passing.

Description

A kind of efficient infection route-caching management strategy method in delay-tolerant network

Technical field

The present invention relates to communication field, particularly relate to a kind of efficient infection route-caching management strategy method in delay-tolerant network.

Background technology

In the face of infecting the serious consumption routeing storage and broadband resource, many research worker propose some heuritic approaches, optimize and improve the forwarding decision of infection route, such as PROPHET Routing Protocol, Spay Routing Protocol.The main restricted copy amount of these methods, based on historical information, probability Estimation, prediction, effectiveness etc..These improve and can reduce the resource consumption infecting route to a certain extent, but can not avoid the possibility of node congestion completely.It practice, in the network that resource is relatively limited, the phenomenon that network node caching is not enough still occurs the most frequently.

When network node inadequate buffer space, in order to accept new message, node needs to carry out packet loss decision-making by cache management strategy, selects a message to abandon, vacate memory space to new information from local message queue.But, tradition cache management strategy, such as Drop-Last (DL), Drop-random, Drop-front (DF), do not take into full account and utilize DTN network to infect the characteristic of route and " store-carry-forward " mechanism, in DTN network environment to node congestion in the case of the raising effect of network performance very limited.Correlational study [14] shows that DF strategy is sending the rate of passing and the performance on network delay will be better than DL strategy.In addition, in packet loss selection course, give source message (message that local node generates) privilege, retained and be not selected and abandon, that can improve message to a certain extent send the rate of passing, but can slow down message propagation rate in a network, because the actual available space of whole network is reduced, information flow is under some influence.Different cache management strategy is combined by Lindgren et al. [18] with Routing Protocol, analyzes its network performance.Result shows, opportunism route can optimize DTN network various aspects of performance with the combination of appropriate cache management strategy.It should be noted that result is emphasized: under infecting route, DF strategy has the highest transport (deliveryrate), and this is also herein by using DF strategy as the reason comparing reference.AmirKrifa et al. [19] proposes to carry out the optimum cache management strategy of packet loss selection based on network global information, but this method has the strictest hypothesis to the motor pattern of network node, and require that in network, the message size of transmission is consistent, and having preferable transfer rate, these assume to be difficult in actual applications all meet.

Summary of the invention

The technical problem to be solved is to provide a kind of cache management strategy method infecting route in delay-tolerant network, with improve in delay-tolerant net infect route-caching congested time message send to rate.

For achieving the above object, the present invention provides a kind of cache management strategy method infecting route in delay-tolerant network, including: history contact history is safeguarded, packet loss decision-making, reduces superinfection.

Described history contact history is safeguarded, is used for: provide initial data for calculating " local infection density ".For message M in arbitrary node i, if node j carries message M, then message M in node j adds message M distribution density around node i, reduces the node i infection effectiveness as " carrier " of message M；Contact probability has directly impact to the infection effectiveness of node, if node i is the highest with the contact probability of node j, message M in the j copy relative to the M in other node in distribution the closer to i, message M distribution density around node i is produced bigger increment, the probability that so message M is continued " infection " by node i is the lowest, infects effectiveness the lowest.

Described packet loss decision-making, is used for: judge when network node is congested, preferential which node of deletion.Using the contact probability p of node j and node i as the increment to message M distribution density around node i of message M in node j；By calculate contact with node i node frequently to message M in node i the increment sum of distribution density try to achieve " the local infection density " of message M of node i.For certain node, " the local infection density " of a message is the highest, illustrate node that the copy of this message is distributed and this node contact the most frequent or quantity is the most, when this message is dropped, the minimizing of " infection " chance of this message caused owing to this message is dropped is by relatively small, because contact node frequently with multihome node, to carry the probability of this message higher.So the message deleting " local infection density " high when node occurs congested is preferentially deleted.

Reduce the impact that the message being dropped is propagated by " packet loss " in a network the most on the whole, thus improve message and send the rate of passing.

Described reduction superinfection, is used for: after a node abandons a message, the most fully erased message correlation information, but the identification information of reservation message within a certain period of time, one " gravestone " mark is set for it, directs at end of fixing time, delete message correlation information the most completely.Within this time period, sending, if there being other external nodes to attempt to node, the message identified Bei " gravestone ", node will rejection.

As can be seen from the above scheme, in the present invention, packet loss decision-making reduces the impact that the message being dropped is propagated by " packet loss " in a network on the whole, thus improves message and send the rate of passing.And the method reducing superinfection effectively alleviates storage pressure and the cache congestion that " repeatedly infecting " brings.

Accompanying drawing explanation

In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, the accompanying drawing used required in embodiment or description of the prior art will be briefly described below, apparently, accompanying drawing in describing below is only some embodiments of the present invention, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.

Fig. 1 is DTN network node schematic diagram in the present invention；

Fig. 2 is t1 moment network diagram of the present invention；

Fig. 3 is t2 moment network diagram of the present invention.

Wherein (t2 ＞ t1)

Detailed description of the invention

Understandable for enabling the above-mentioned purpose of the present invention, feature and advantage to become apparent from, the present invention is further detailed explanation with detailed description of the invention below in conjunction with the accompanying drawings.Obviously, described embodiment is only a part of embodiment of the present invention rather than whole embodiments.Based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art are obtained under not making creative work premise, broadly fall into the scope of protection of the invention.

Assume that the DTN network under nonrandom motor pattern is made up of wireless mobile node, the arbitrary node i in network:

N_iT () represents the message count in t node i caching.

LM_iT () represents the message queue in t node i caching

(i, j, t) represent t to Contact, and node i is met with node j's.

Each message in network has a unique destination node, and has a unified lifetime value TTL (time-to-live).After life span terminates, message will be removed from caching.The exchange of internodal information is sent in after two nodes mutually enter the communication context (contact) of the other side and carries out.By transfer rate, effectively transmitting time restriction, the exchange of information is the most sufficient.

1. history contact maintenance of information: in LDI, arbitrary node i is by for one contact history queue CL of self maintained_iRecord and be in contact with it network node frequently, and be CL_iIn each Contact bind one according to time of contact interval with frequent degree variation life span CL_{Contact (i, j, t)}With contact cumulative numberWork as t₁In the moment, node i runs into a CL_iIn when there is no the node j of history contact history, by new record Contact (i, j, t₁) add CL_i, and giveInitial value α, and makeEqual to 1.WhenWhen expiring, i.e. when

${\mathrm{CL}}_{Contact(i,j,t_1)}<t-t_1$

(t is current time)

Node i is by Contact (i, j, t₁) from CL_iMiddle deletion, and remove related status information.If t, the node j that node i runs into, and at CL_iIn there is node j at t₁The history contact history in moment, then update Contact (i, j, t₁) it is Contact (i, j, t₂), and make

${\mathrm{CL}}_{Contact(i,j,t_2)}={\mathrm{CL}}_{Contact(i,j,t_1)}+\mathrm{\&alpha;}-(t-t_1)$

${\mathrm{CA}}_{inew}^j={\mathrm{CA}}_{iold}^j+1$

Front and back being spaced the shortest twice time of contact, contact node the most frequently, its life span can be more and more longer, is more permanently maintained in contact history queue, and its contact cumulative number also can progressively tire out increasing.CL maximum is less than life span TTL of nodes.By such mode, LDI will be higher with node contact probability, and contact node frequently remains stored in the contact history queue of oneself, makes contact few simultaneously, or the node that front and back twice contact interval time is the longest, deleted from queue over time.

2. calculate " local infection density ": each message M ∈ LM during arbitrary node i is caching in LDI_iT (), safeguards an infected queue IL_M, for any node j ∈ IL_M, node j is at CL_iIn there is contact history, and when the last node j carries out information exchange with node i, the caching of node j carries message M.When contact history is expired, IL synchronization removal corresponding node record wherein.

The contact cumulative number that contact probability is node of definition node and CL_iIn the ratio of sum of contact cumulative number of all nodes:

$P_j={\mathrm{CA}}_i^j/\underset{n\&Element;{\mathrm{CL}}_i}{\&Sigma;}{\mathrm{CA}}_i^n$

(n∈CL_iRepresent that node n is at CL_iIn there is contact history)

Definition Ldi_M(M∈LM_i(t)) represent " the local infection density " of message M in node:

${\mathrm{Ldi}}_M=\underset{n\&Element;{\mathrm{IL}}_M}{\&Sigma;}{P}_n$

(M∈LM_i(t)

" the local infection density " of message M be M infected queue IL in the total value that is added with the contact probability of node i of all nodes.

3. cache management decision-making: when joint occurs congested, abandon the message that Ldi value is maximum.If there is the message that multiple maximum Ldi values are identical, the message selecting wherein cache-time the longest abandons.

Therefore, a kind of based on public IP network the communication system of present invention offer, method and apparatus, have the advantage that.

(1) improve information sends to rate

When the message that local density of infection is high is dropped, the minimizing of " infection " chance of this message caused owing to this message is dropped is by relatively small, because contact node frequently to carry the probability of this message higher with multihome node, reduce the impact that the message being dropped is propagated by " packet loss " in a network the most on the whole, thus improve message and give the rate of passing, the message that those infected degree are low simultaneously also obtains the longer propagation time.

(2) the cache congestion probability of network node is reduced

Packet loss during network node caching, effectively reduces the most congested of caching.

(3) degree of coupling with other modules is reduced

Need not infect route and extra side information is provided, because infecting route itself to require mutually to send between node the message index information in local cache, for quickly finding the difference of the message carried in two nodes of contact.

(4) pressure of network node itself is reduced

Because policy accounting is local infection density, it is not necessary to statistics global information, it is not required that any network topology knowledge.

Description by above embodiment of the method, those skilled in the art is it can be understood that can add the mode of required general hardware platform by software to the present invention and realize, can certainly pass through hardware, but a lot of in the case of the former is more preferably embodiment.Based on such understanding, the part that prior art is contributed by technical scheme the most in other words can embody with the form of software product, this computer software product is stored in a storage medium, including some instructions with so that a computer equipment (can be personal computer, server, or the network equipment etc.) perform all or part of step of method described in each embodiment of the present invention.And aforesaid storage medium includes: the various media that can store program code such as read only memory (ROM), random access memory (RAM), magnetic disc or CDs.

For system embodiment, owing to it essentially corresponds to embodiment of the method, so the part that relevant part sees embodiment of the method illustrates.Device described above or system embodiment are only schematically, the wherein said unit illustrated as separating component can be or may not be physically separate, the parts shown as unit can be or may not be physical location, i.e. may be located at a place, or can also be distributed on multiple NE.Some or all of module therein can be selected according to the actual needs to realize the purpose of the present embodiment scheme.Those of ordinary skill in the art, in the case of not paying creative work, are i.e. appreciated that and implement.

In several embodiments provided herein, it should be understood that disclosed system and method, in being not above spirit and scope, can realize in other way.Current embodiment is a kind of exemplary example, should not be taken as limiting, and given particular content should in no way limit the purpose of the application.Such as, described unit or the division of subelement, it is only a kind of logic function and divides, actual can have other dividing mode, the most multiple unit or multiple subelement to combine when realizing.It addition, multiple unit can or assembly can in conjunction with or be desirably integrated into another system, or some features can be ignored, or does not performs.

It addition, described system and method and the schematic diagram of different embodiment, without departing from scope of the present application, can be with other system, module, techniques or methods combine or integrated.Another point, shown or discussed coupling each other or direct-coupling or communication connection can be the INDIRECT COUPLING by some interfaces, device or unit or communication connection, can be electrical, machinery or other form.

The above is only the detailed description of the invention of the present invention; it should be pointed out that, for those skilled in the art, under the premise without departing from the principles of the invention; can also make some improvements and modifications, these improvements and modifications also should be regarded as protection scope of the present invention.

Claims (2)

1. a delay-tolerant network infects route-caching management strategy method, it is characterized in that, by the predictable and historic state information of the node that contacts effectively utilizing nodes to move, for each one this message of message construction in each node copy with this node contact node frequently in distribution equivalence value " local infection density ", i.e. LDI, packet loss decision-making is carried out according to the size of LDI value, the circular of LDI is: history contact maintenance of information: in LDI, arbitrary node i is by for one contact history queue CL of self maintained_iRecord and be in contact with it network node frequently, and be CL_iIn each Contact bind one according to time of contact interval with frequent degree variation life span CL_{Contact (i, j, t)}With contact cumulative number(i, j, t) represent t to Contact, and node i is met with node j's, worked as t₁In the moment, node i runs into a CL_iIn when there is no the node j of history contact history, by new record Contact (i, j, t₁) add CL_i, and giveInitial value α, and makeEqual to 1, whenWhen expiring, i.e. whenT is current time, and node i is by Contact (i, j, t₁) from CL_iMiddle deletion, and remove related status information, if t, the node j that node i runs into, and at CL_iIn there is node j at t₁The history contact history in moment, then update Contact (i, j, t₁) it is Contact (i, j, t₂), and make Front and back it is spaced the shortest twice time of contact, contact node the most frequently, its life span can be more and more longer, more permanently it is maintained in contact history queue, its contact cumulative number also can progressively tire out increasing, CL maximum is less than life span TTL of nodes, by such mode, LDI will be higher with node contact probability, contact node frequently remains stored in the contact history queue of oneself, make contact few simultaneously, or the node that front and back twice contact interval time is the longest, deleted from queue over time；Calculate local infection density: in LDI, arbitrary node i is each message M ∈ LM in caching_iT (), safeguards an infected queue IL_M, LM_iT () represents the message queue in t node i caching For any node j ∈ IL_M, node j is at CL_iIn there is contact history, and when the last node j and node i carry out information exchange, the caching of node j carries message M, when contact history is expired, IL synchronization removal corresponding node record wherein, the contact cumulative number that contact probability is node of definition node and CL_iIn the ratio of sum of contact cumulative number of all nodes: n∈CL_iRepresent that node n is at CL_iIn there is contact history, define Ldi_MThe local infection density of message M in expression node:M∈LM_i(t), the local infection density of message M be M infected queue IL in the total value that is added with the contact probability of node i of all nodes, when node occurs congested, abandon the message that LDI value is maximum, if there is the message that multiple maximum LDI values are identical, the message selecting wherein cache-time the longest abandons.

2. the method based on the minimizing message superinfection being effectively retained the time being applied to method described in claim 1, it is characterized in that, after a node abandons a message, the most fully erased message correlation information, but the identification information of reservation message within a certain period of time, for it, one " gravestone " mark is set, direct at end of fixing time, delete message correlation information the most completely, within this time period, sending, if there being other external nodes to attempt to node, the message identified Bei " gravestone ", node will rejection.