KR20130034524A - Method of routing protocol based on aodv - Google Patents

Abstract

PURPOSE: An AODV(Ad-hoc On-demand Distance Vector) based routing method is provided to remarkably reduce RREQ(Route REQuest) flooding by broadcasting a selective RREQ. CONSTITUTION: When a node N(10) broadcasts an RREQ message to a destination node(30), a node M(20) receiving the RREQ message from the node N broadcasts an RREQ according to a condition satisfaction state. A node including path information for a destination node D for the RREQ received from the node N generates an RREP(Route REPly). When path information for the destination node D is not existed, nodes existed in a domain broadcast the RREQ.

Description

AODV-based routing method {method of routing protocol based on AODV}

The present invention relates to an AODV-based routing method for communicating in an ad-hoc manner, and specifically, to minimize Route REQuest (RREQ) flooding of an Ad-hoc On-demand Distance Vector (AODV) protocol. The present invention relates to an AODV-based routing method in which a path is geometrically close to a straight line from a source node to a destination, and selectively broadcasts an RREQ.

Existing routing protocols flood the control message to find the route and allow it to be transmitted throughout the network. However, flooding of such control messages can lead to severe transmission yield degradation at slow transmission rates. This is especially the case in the marine communications environment.

Currently, ships use satellite communication to communicate with land, and satellite communication is very expensive. In order to reduce such communication costs, researches on marine communication technology using the VHF band have been actively conducted. However, the VHF band has a relatively narrow bandwidth and a limitation in transmission speed. Therefore, it is necessary to minimize the control traffic in the network protocol to minimize the transmission yield degradation. In addition, there are many difficulties in installing and operating a fixed infrastructure at sea. The installation of structures in the sea requires much larger installation and maintenance costs than on land, the power supply is not easy, and there are difficulties in compensation negotiations with fishermen on the coast. Therefore, the Ad-hoc communication method, in which the nodes involved in the communication act as a relay rather than the communication method for establishing the main network, is suitable in the ocean.

In ad-hoc communication environment, it is essential to find a route to deliver data to the destination. It is very important to improve the network transmission yield by minimizing the control message transmission in such a path discovery in a slow data transmission environment.

In order to reduce the RREQ flooding of the AODV protocol, the present invention has been designed to solve the above requirements, and the RREQ flooding is performed by selectively broadcasting the RREQ so as to establish a geometrically straight line from the source node to the destination. An object of the present invention is to provide an AODV-based routing method that can be greatly reduced.

In order to achieve the above object, the AODV-based routing method according to the present invention is a method for establishing a path between nodes using an AODV routing protocol between nodes in a wireless ad hoc network. Broadcasting an RREQ, which is a routing request message, to a neighboring node to query the route information from any node N to the destination node D; I. Any node M receiving the RREQ message from the node N is greater than the minimum radius value R_IN of the distance between the node N and the node M and is smaller than the communication radius R_OUT of the modem of the node N, and node M And broadcasting the routing request message when the distance between the destination D and the destination D is less than the value of the distance between the node N and the destination node D minus the minimum radius value.

According to the AODV-based routing method according to the present invention, it is possible to significantly reduce the path search time and improve the transmission yield of the entire network.

1 is a view for explaining an AODV-based routing method according to the present invention,
2 is a flowchart illustrating a RREQ generation and processing procedure of a source node according to the present invention;
3 is a flowchart illustrating an RREQ processing procedure of a relay node according to the present invention;
4 is a diagram illustrating an example of an RREQ message format according to the present invention.

Hereinafter, an AODV based routing method according to a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.

In the present invention, it is assumed that all nodes on the network know their own location coordinates and that the location information of the destination is known to all nodes in the network. Thus, a path search for passing data through multiple hops to the destination where the location information is known is performed. Consider the process. For example, as shown in FIG. 1, a ship on the sea may transmit data to a coastal base station. Here, the node N 10 may be a source node or a relay node as any node on the wireless ad hoc network for querying route information from the source node to the destination node D 30 for data transmission.

In general, the position information of coastal base stations is known to all ships in the sea, and since the ships in the sea are equipped with GPS, their position is also known.

Ad-hoc routing protocols can be divided into two types, namely, Reactive and Proactive. In case of limited transmission speed and mobility, routes can be routed whenever necessary rather than Proactive, which updates the routing table through periodic message flooding. Reactive method is suitable, and AODV protocol is widely used among Reactive methods. RREQ is the only message that is flooded in the AODV protocol. The present invention proposes a method for minimizing RREQ flooding of the AODV protocol in such a network topology.

The evaluation indicator used by the AODV protocol to determine the path is the number of hops. The optimal routing path that minimizes the number of hops consists of nodes close to the straight line connecting the source and the destination. Also, to minimize the number of hops, one hop must be close to the modem's communication radius. However, in order to increase the probability of finding a route, the AODV protocol broadcasts an RREQ message at least once by a node receiving an RREQ message. Especially for dense network topologies where many nodes are located within the communication radius of the modem, this RREQ flooding of AODV causes severe network transmission yield degradation and very long path seek times due to unnecessary RREQ flooding.

In order to minimize such RREQ flooding, the present invention allows the routing path to consist of nodes close to the straight line connecting the source and the destination, with one hop close to the communication radius of the modem.

In FIG. 1, when node N 10 broadcasts an RREQ message to a destination node (D) 30, node M 20 receiving this RREQ message from node N 10 receives Equation 1 and Equation 2 below. Only broadcast the RREQ when it is satisfied.

The region satisfying the equations (1) and (2) is a hatched area in Fig. 1, denoted by A.

In the above equation, d (x, y) is the distance between node x and node y. R_OUT is a communication radius of the modem of the node N 10, and R_IN is a value smaller than R_OUT, which should be determined in consideration of the density of the network, which will be described later. If the value of R_IN is too large, the node may not be found because there is no node in area A. The smaller the value of R_IN, the larger the hop count of the found path. On the other hand, the area A is calculated by the following equation (3).

In Equation 3 above, the trigonometric function may be expressed by Equation 4 below.

In Equations 3 and 4, d (N, D), R_IN, and R_OUT are represented by R, r ₀ and r ₁ for convenience. Since the area A is very small compared to the area having the communication radius R_OUT around the node N 10, it can be seen that the RREQ flooding is greatly reduced. In order to determine whether Equation 2 is satisfied, M must know d (N, D). This may include the location information of the node N 10 in the RREQ message or payload, or alternatively include the d (N, D) value.

In particular, in the marine communication, as shown in the example of the Automatic Identification System (AIS), since ship position information is one of the most important information, the payload position information is usually included in the payload.

The node having the route information for the destination D with respect to the RREQ received from the node N 10 generates a RREP (Route Reply). If there is no route information for the destination node D 30, only nodes existing in the area A broadcast the RREQ. At this time, in order to improve fairness and further reduce RREQ flooding, additional selective RREQ broadcasting is performed as follows. That is, when a node in the A area first broadcasts an RREQ, other nodes in the A area that listen to it do not broadcast the RREQ, that is, delete the corresponding RREQ message from the queue. This ideally ensures that only one node with the least traffic broadcasts the RREQ. However, to increase the probability of successful path finding, it is necessary to increase the number of nodes broadcasting the RREQ. First, among the nodes of the area A, an event for determining whether to broadcast an RREQ is called STAR.

The method of judging the occurrence of STAR is that the nodes in the area A have already received the RREQ message of the node N (10), so they know this information, which is the same as the RREQ of the node N (10). If a received RREQ is received within a certain distance or at a certain level or more, it can be regarded as a STAR.

Suppose that node M1 and node M2 are in area A, and that M2 has received an RREQ message from M1 with a hop number one greater than the RREQ message received from N. If the location information of M1 can be known, it is determined that STAR has occurred when the following Equation 5 is satisfied.

In Equation 5, STAR_R_THR may be appropriately set as a threshold value.

If the STAR_R_THR value increases, the number of nodes broadcasting RREQ in the area A decreases. On the contrary, if the STAR_R_THR value decreases, the number of nodes broadcasting RREQ in the area A increases.

If the location information of M1 is unknown, it is determined that STAR has occurred if the following Equation 6 is satisfied.

In Equation 6, SNR_M2 is a signal-to-noise ratio of the RREQ message M2 receives from M1, and STAR_SNR_THR may be appropriately set as a threshold value.

When the STAR_SNR_THR value decreases, the number of nodes broadcasting RREQ in the area A decreases. On the contrary, when the STAR_SNR_THR value increases, the number of nodes broadcasting RREQ in the area A increases.

Such selective RREQ flooding has a higher probability of failing to find a path because the RREQ transmission fails in the middle due to a collision in the MAC layer and an increase in BER due to the deterioration of the channel environment of the physical layer than the conventional AODV. In order to overcome this phenomenon, the present invention uses an effective indirect ACK method. If a node broadcasts an RREQ, firstly, if my 1-hop neighbor node has route information for D, it will receive an RREP. Second, if my 1-hop neighbor node has no route information for D, Since one of the 1-hop neighbor nodes broadcasts the RREQ, it receives this RREQ, which is considered to have received an indirect ACK. In other words, it is determined that the RREQ was transmitted successfully without collision. The event that this indirect ACK holds is named SHARP. The specific process of determining whether SHARP is generated is as follows.

1) received a RREP message for the broadcasted RREQ, or

2) Let me be myself. When a duplicate RREQ is received from a node M ', the "Hop Count" of the RREQ that is repeatedly received is greater than 1 by the "Hop Count" of the RREQ broadcasted by M, and the following Equation 7 or Equation 8 is satisfied. , It is determined that SHARP has occurred.

In Equation 8, SNR_IN and SNR_OUT are signal-to-noise ratios of the receiver, which are received when the distance between R_IN and R_OUT is propagated in the path loss model used in the physical layer, respectively.

A detailed process for implementing the method for minimizing the RREQ flooding of the AODV protocol will be described with reference to FIG. 2 with respect to a process for the case where the node N described above is a source node that is an RREQ originator.

Here, the RREQ originator means a source node on a routing path. If there is no valid routing information (RE; Routing Entry) to the destination D node 30, the source node generates and broadcasts an RREQ in order to find a delivery path. 110).

The parameter W_MAX, which is set during the initialization process, is the allowable time for waiting for the RREP to be received after broadcasting the RREQ, and the TTL is the maximum number of hops for which the RREQ is propagated. TTL is calculated by Equation 9 below.

Where x is the function to find the minimum integer greater than or equal to x. In addition, TTL_OFFSET is a constant value for adding up the calculated TTL, and it is preferable to apply 2 in marine communication. TTL_MAX is the maximum number of hops determined by the network size and can be set appropriately. min (A, B) is an operator that selects the smaller of A and B. W_MAX is calculated by Equation 10 below.

Here, the W_SH_MAX value is the average time for one hop round trip of the R_OUT distance.

EVENT_STAR is a variable indicating whether STAR is applied and is a field present in the RREQ message. When the EVENT_STAR value is 1, it is determined whether or not a STAR has occurred to selectively broadcast the RREQ, and the initial value is set to "1". In the present invention, if RREP is not received despite RREQ retries as much as REP_RREQ_MAX, the probability of successful path finding is improved by reducing the intensity of selective RREQ flooding by decreasing R_IN and increasing the size of region A. At this time, REP_RREQ_DEPTH is a variable that determines the R_IN value, is a field present in the RREQ message, the initial value is "0". REP_RREQ is the number of times RREQ is attempted again because RREP was not received in the same REP_RREQ_DEPTH. The initial value is "0". REP_RREQ_NO_STAR is the number of RREQ attempts without applying STAR, and the initial value is "0".

After the variable initialization, the RREQ is broadcast (step 120).

An example of an RREQ message format for realizing the present invention is shown in FIG. Referring to FIG. 4, the RREQ message 400 is defined as a new message Type 410 for compatibility with the existing AODV, and the value of Type is set to “5”. Compared to the existing RREQ message format, four fields (420 to 450) have been added, and a lower number of "Hop Count" is the same as the existing RREQ. The EVENT_STAR FID 420 and the REP_RREQ_DEPTH FILTER 430 are always listed, and the REP_RREQ FILTER 440 and the REP_RREQ_NO_STAR FILTER 450 are optional to determine the life time of the reverse route according to the network traffic situation. It can be utilized.

After broadcasting the RREQ, it is determined whether the RREP has not been received (step 130), and if waiting for W_MAX fails to receive the RREP, retry the RREQ while increasing the REP_RREQ value by one (step 150) but doubling the W_MAX value. . In Figure 2, W is a variable indicating the time to wait for receiving the RREP after transmitting the RREQ. If no RREP is received after REP_RREQ_MAX RREQ retries, increase the REP_RREQ_DEPTH value by 1 (step 170), calculate the TTL value using Equation 9 according to the reduced R_IN value according to the increased REP_RREQ_DEPTH (step 190). ), REP_RREQ is initialized to "0" and retry RREQ by doubling the W_MAX value. A method of calculating R_IN according to the REP_RREQ_DEPTH value is represented by Equation 11 when the EVENT_STAR value is "1", and is represented by Equation 12 when the EVENT_STAR value is "0".

In Equation 11, R_IN_DEF is a default value of R_IN. R_IN_DEF is an R_IN value used when broadcasting RREQ for the first time. R_DELTA is the size of R_IN that decreases whenever REP_RREQ_DEPTH increases.

If the REP_RREQ_DEPTH value reaches REP_RREQ_DEPTH_MAX in step 180 but does not receive an RREP, EVENT_STAR is set to "0" (step 190), and the RREQ is attempted to prevent selective RREQ flooding using STAR. In this case, retry as much as REP_RREQ_NO_STAR corresponding to the maximum setting value by going through the judgment process of step 210 until receiving the RREP with TTL as TTL_MAX and increasing the W_MAX value twice each time.

Next, the RREQ processing of the relay node that has received the RREQ described above will be described with reference to FIG. 3.

First, the relay node performs initialization (step 310).

In the RODQ of the existing AODV, all duplicated RREQs are discarded. However, in the present invention, whether STAR and SHARP are generated is determined through the duplicated RREQ messages.

FLAG_DISCARD is a variable indicating whether or not duplicate RREQ messages have been processed. The initial value is "0". When this value reaches "1", subsequent RREQ messages are discarded. FLAG_STAR and FLAG_SHARP are variables indicating whether STAR and SHARP are generated, respectively, and an event occurs when a value is "1", and all initial values are "0". PASS_STAR is set to "1" when the RREQ is broadcast without a STAR, and the initial value is "0". The REP_ACK is a variable indicating the number of times the RREQ is broadcasted again because the indirect ACK is not received, that is, no SHARP occurs after the broadcast of the RREQ. The initial value is "0".

If the RREQ is received after such initialization (step 320), it is determined whether a duplicate reception is received (step 330). In the case of duplicate reception, it is determined whether STAR and SHARP are generated by using the duplicated RREQ only when FLAG_DISCARD is "0" and EVENT_STAR is "1" through the determination process of steps 340 and 350. At this time, if PASS_STAR is "0", it is determined whether STAR has occurred, and when STAR occurs, both FLAG_STAR and FLAG_DISCARD are set to "1". If PASS_STAR is "1", it is determined whether SHARP is generated. If SHARP is generated, both FLAG_SHARP and FLAG_DISCARD are set to "1".

Unlike this, if the RREQ received in step 330 is not a duplicate reception, it is checked whether there is routing information (RE) for the destination D. If there is an RE for D, send RREP with FLAG_DISCARD set to "1". If there is no RE for D and the above-described equations (1) and (2) are not satisfied at the same time through the determination in step 370, FLAG_DISCARD is set to "1" and ends. If there is no RE for D and both Equations 1 and 2 are satisfied at the same time, the processing method depends on the EVENT_STAR value. If EVENT_STAR is "0", the RREQ is broadcast unconditionally. However, if EVENT_STAR is "1", the RREQ to be broadcasted is enqueued. If the FLAG_STAR value is set to "1" before broadcasting while in the queue, the RREQ message is deleted from the queue and terminated. Otherwise, the RREQ is broadcasted.

After broadcasting the RREQ, if EVENT_STAR is " 0 ", the procedure ends immediately. If EVENT_STAR is " 1, " If FLAG_SHARP does not become "1" during W_SH_MAX time after the RREQ broadcast, REP_ACK is increased by 1 and the RREQ is rebroadcasted. The number of rebroadcasts is limited by REP_ACK_MAX. If FLAG_SHARP does not become "1" even after re-broadcasting by REP_ACK_MAX, the operation ends with FLAG_DISCARD "1".

10: node N 20: node M
30: Node D

Claims (9)

In the method of routing between nodes using the AODV routing protocol between nodes in a wireless ad hoc network,
end. Broadcasting an RREQ, which is a routing request message, to a neighboring node to query the route information from any node N to the destination node D;
I. And receiving an RREQ message from the node N, broadcasting the RREQ when the node M simultaneously satisfies Equation 1 and Equation 2 below.
[Equation 1]
R_IN <d (N, M) <R_OUT
&Quot; (2) &quot;
d (M, D) <d (N, D)-R_IN
In the above equation, the function d (x, y) is a distance between x and y, the R_OUT is the communication radius of the modem of the node N, the R_IN is a value that is set smaller than the R_OUT. The method of claim 1, wherein in step b),
Any node M receiving the RREQ message from the node N includes a node M1 and a node M2 satisfying the conditions of Equation 1 and Equation 2, and the node M2 hops more than the RREQ message received from the node N. When the number of RREQ messages increased by 1 is received from the node M1, when the following Equation 5 or 6 is satisfied, the node M2 does not broadcast the RREQ.
&Quot; (5) &quot;
d (M1, M2) <STAR_R_THR
&Quot; (6) &quot;
SNR_M2> STAR_SNR_THR
In Equation 5, STAR_R_THR is a set first threshold value, in Equation 6, SNR_M2 is a signal-to-noise ratio of an RREQ message received by the node M2 from the node M1, and STAR_SNR_THR is a set second threshold value. Based routing method. The number of hops of the RREQ broadcasted by the node M according to claim 1, wherein the node M has received an RREP message for the broadcasted RREQ, or has repeatedly received an RREQ from another node M ', When it is larger than 1 and satisfies Equation 7 or Equation 8 below, the node M determines that the ACK is a ACK for the RREQ broadcast.
&Quot; (7) &quot;
R_IN <d (M, M ') <R_OUT
&Quot; (8) &quot;
SNR_OUT <d (M, M ') <SNR_IN
In Equation 8, SNR_IN and SNR_OUT are the signal-to-noise ratios of the receivers received when propagating the distance between R_IN and R_OUT in the path loss model used in the physical layer, respectively. The method of claim 1, wherein in the step A, when the node N is a small node which does not have a valid routing information (RE) to a destination and generates and broadcasts an RREQ to find a delivery path,
A-1. As an initialization process, W_MAX and TTL are set, EVENT_STAR is initialized to "1" and included in the RREQ message, and REP_RREQ_DEPTH, which is a parameter for determining the R_IN value, is initialized to "0" and included in the RREQ message, and the same REP_RREQ_DEPTH Initializing REP_RREQ, which is a parameter for the number of times the RREQ message is transmitted with respect to a value, to be included in the RREQ message, and initializing REP_RREQ_NO_STAR to "0" to include in the RREQ message for broadcasting;
A-2. If the RREP is not received even after waiting for the W_MAX after step A-1, if the RREQ is retried after doubling the value of W_MAX but not receiving the RREP after the RREQ retry of the set REP_RREQ_MAX, the REP_RREQ_DEPTH value is increased by one. Making a step;
A-3. Calculating the TTL according to the increased R_IN value according to the increased REP_RREQ_DEPTH, resetting the REQ by initializing the REP_RREQ to "0" and doubling the W_MAX value;
A-4. If the REP_RREQ_DEPTH value does not receive the RREP so that the REP_RREQ_DEPTH_MAX is exceeded, retry as much as REP_RREQ_NO_STAR until the RENT is received with the EVENT_STAR as "0" and the TTL as the TTL_MAX, and double the value of W_MAX for each retry. Including;
The W_MAX is the allowable time for waiting until the RREP is received after broadcasting the RREQ, the TTL is the maximum number of hops to which the RREQ is propagated, and the EVENT_STAR is among nodes satisfying Equations 1 and 2 The occurrence of STAR, which is an event for determining whether or not to broadcast RREQ, and when the value is 1, the STAR is generated. When the value is 0, the STAR is not generated. The REP_RREQ_DEPTH is a variable for determining the R_IN, and the REP_RREQ is In the same REP_RREQ_DEPTH, the number of times that RREP is retried, REP_RREQ_NO_STAR is the number of RREQ attempts without applying STAR, and the REP_RREQ_MAX is the maximum value set for the REP_RREQ, and the REP_RREQ_DEPTH_MAX is the maximum value set for the REP_RREQ_DEPTH. AODV-based routing method. The method of claim 2,
As a relay node receiving the RREQ from the node N in step b is a selective RREQ broadcast method,
B-1. Initializing the FLAG_DISCARD, FLAG_STAR, FLAG_SHARP, PASS_STAR, and REP_ACK to "0";
B-2. When the FLAG_DISCARD is "0" and the EVENT_STAR is "1" when the RREQ is repeatedly received, when the PASS_STAR is "0", the FLAG_STAR and the FLAG_DISCARD when the duplicate RREQ message satisfies Equation 5 or 6 Set both to "1" and set both FLAG_SHARP and FLAG_DISCARD to "1" when PASS_STAR satisfies Equation 7 or Equation 8 when PASS_STAR is "1";
B-3. If the received RREQ is not duplicate reception, if there is routing information (RE; routing entry) for the destination node D, transmitting RREP with FLAG_DISCARD as "1";
B-4 if the received RREQ is not a duplicate reception, if there is no RE for the destination node D and does not satisfy Equation 1 and Equation 2 at the same time, then ends FLAG_DISCARD as “1”;
B-5. If the received RREQ is not duplicate reception, if there is no RE for the destination node D and satisfies Equation 1 and Equation 2 simultaneously, if the EVENT_STAR is "0", the RREQ is unconditionally broadcasted. If the EVENT_STAR is "1", Deleting the RREQ message from the queue and ending the broadcast if the FLAG_STAR value is set to "1" before broadcasting while queued in the queue and broadcasting in the queue; otherwise, broadcasting the RREQ;
B-6. After broadcasting the RREQ, if the EVENT_STAR is "0", ending FLAG_DISCARD with "1" and ending;
B-7. After broadcasting RREQ, if EVENT_STAR is "1", for indirect ACK, if FLAG_SHARP does not become "1" for W_SH_MAX time after RREQ broadcasting, increase REP_ACK by 1 and re-broadcast RREQ, and perform RREQ re-broadcast by REP_ACK_MAX, If the number of re-broadcast, even if FLAG_SHARP does not become "1" AODV-based routing method characterized by ending with FLAG_DISCARD "1". The method of claim 4, wherein the TTL is
Calculated by TTL = min ("d (N, D) / R_IN_ + TTL_OFFSET, TTL_MAX),
Here, xx is the minimum integer greater than or equal to x, TTL_OFFSET is a constant set to reserve TTL, TTL_MAX is the maximum number of hops set according to the network size, and min (A, B) is A, B. AODV-based routing method characterized in that the operator selects the smaller of the two. The method of claim 4, wherein the W_MAX is
Calculated by W_MAX = TTL × W_SH_MAX,
The W_SH_MAX value is an average time taken for one hop round trip of the R_OUT distance. 5. The method of claim 4, wherein the RREQ message includes the EVENT_STAR and the REP_RREQ_DEPTH values. The method of claim 8, wherein when the EVENT_STAR value is "1", the R_IN is calculated as shown in Equation 11 below. If the EVENT_STAR value is "0", the R_IN is calculated as shown in Equation 12 below.
[Equation 11]
R_IN = R_IN_DEF-R_DELTA × REP_RREQ_DEPTH
[Equation 12]
R_IN = 0
In Equation 11, R_IN_DEF is a default value set for R_IN, and R_IN_DEF is an R_IN value used when broadcasting the RREQ for the first time, and the R_DELTA is a constant that determines a decreasing rate when the REP_RREQ_DEPTH increases. AODV-based routing method, characterized in that.

Images