CN102638820B - Ad Hoc network link stability prediction method - Google Patents

Abstract

The invention discloses an Ad? Link Stability Prediction Method in Hoc Networks. The steps are: first step, link direct stability assessment; second step, link indirect stability assessment; third step, link comprehensive stability assessment; fourth step, multicast routing protocol based on stability. It solves the problem that the multicast routing protocol that takes the existing link stability as the routing standard cannot well adapt to unstable environments such as high mobility and high interference. The invention has simple calculation, low calculation amount in the link stability evaluation process, stable overall performance of the network, strong adaptability, low requirement on hardware, and low energy consumption. <pb pnum="1"/>

Description

Link Stability Prediction Method in Ad Hoc Network

technical field

The invention belongs to the field of mobile AdHoc network (MANET), and in particular relates to a link stability prediction method in the AdHoc network for link stability.

Background technique

Mobile AdHoc network is a dynamically reconfigurable wireless network, each node acts as a host and a router at the same time. As a host, a node needs to run its own application; as a router, a node is responsible for forwarding data according to a specific routing protocol. AdHoc networks are widely used in disaster relief (such as earthquakes, fires) and distributed collaborative computing.

In practical applications, since a node often needs to send the same data to multiple nodes, multicast routing protocols emerge as the times require. However, due to the characteristics of mobile adhoc network such as random movement of nodes, limited resources, and unreliable channels, traditional multicast protocols cannot be well adapted to MANET network environment.

The random movement of nodes leads to dynamic changes in network topology, which increases the frequency of rerouting and affects network performance. Therefore, there is a need for an evaluation method for link stability, which can be used as an evaluation index to avoid unstable links during route selection, which will undoubtedly greatly improve the overall performance of the network.

At present, there are not many studies on link stability in the adhoc field. Most studies assume that the network environment is constant, and only study the mobility of nodes, such as quantifying the position and velocity vector of nodes with the help of GPS, and using statistics The method models the node's mobile behavior and predicts the next moment's mobile behavior of the node. However, in practical applications, not every node has a GPS module, and the GPS module has a large error in the indoor environment, and the statistical method is based on the probability theory to model the nodes, which has certain limitations and is accurate Sex is debatable. In addition, in the existing research, without exception, they all evaluate the link between the detection node and the node under test from the perspective of the detection node, and do not stand on the point of view of the node under test. Therefore, the existing link stability prediction methods have great limitations and the effect is not ideal. Correspondingly, multicast routing protocols that take link stability as the routing standard cannot well adapt to unstable environments such as high mobility and high interference. Therefore, a more reasonable link stability evaluation method and corresponding Multicast routing protocol to solve the above problems.

Contents of the invention

The purpose of the present invention is to provide a link stability prediction method in an AdHoc network, to solve the problems of frequently disconnected links and unstable routes caused by node movement in the adhoc network, and improve the performance of the network (such as increasing the delivery of packets) Rate, reduce end-to-end delay, etc.) to adapt to high mobility environment.

For this reason, the technical scheme that the present invention adopts is:

A link stability prediction method in an AdHoc network, which realistically approximates the real environment of the network by sampling continuously received signal strength, and evaluates each link in the environment for the detection node and the node under test respectively The stability of the stability status, comprehensively consider the detection node and the node under test, calculate the comprehensive stability value, and use it as the routing standard of the stable multicast routing protocol, the specific steps are as follows:

The first step is to directly assess the stability of the link

The node maintains the neighbor relationship by regularly sending hello messages. When the detection node receives the hello message sent by its neighbor node, the detection node maintains a received signal strength or received power list for each neighbor node, that is, the node under test;

The second step, link indirect stability assessment

The tested node uses the same method as the first step to evaluate the stability of its neighbor nodes; since this stability is aimed at evaluating the environment in which the tested node is located, after removing the nodes with particularly high mobility, other nodes Stability takes the mean value, the lower the mean value, the more stable the environment of the tested node, and vice versa, the more unstable the environment;

The third step is the comprehensive stability assessment of the link, which is to comprehensively calculate the direct and indirect stability assessment values of the link using the method of weighted summation;

The fourth step is a stability-based multicast routing protocol. The stability-based multicast routing protocol uses stability as a route selection criterion.

The link direct stability evaluation value in the first step:

According to the Friis free-space propagation model, the ratio P _TX of the received power P _RX to the transmitted power is inversely proportional to the square of the transmission distance R, namely $\frac{P_{\mathrm{RX}}}{P_{\mathrm{TX}}}\&Proportional;\frac{1}{R^2}$ or $R\&Proportional;\sqrt{\frac{P_{\mathrm{TX}}}{P_{\mathrm{RX}}}};$

Assume that the lower limit of the received power is P _min , that is, when the received power is lower than this value, a higher bit error rate will be generated, resulting in the retransmission of data packets, correspondingly, the transmission distance at this time is R _max ; P _min is initialized to a fixed value. When a neighbor node moves out of its communication range, it is considered that the received power at the time of successful reception at the previous moment is approximately equal to the current P _min at this time, so update the node's P _min with the received power at the previous time ;

Assuming that the nth sampling value of the received power of the detected node j by the detection node i is P _n ^ij , then according to the Friis free space propagation model, we have And because of Among them, P _TX is the transmission power of the node, R _n ^ij is the distance between the sending node and the receiving node at the nth power sampling, P _min ⁱ is the minimum receiving power of the detection node i, and R _max is when the receiving power is P _min ⁱ The corresponding maximum transmission distance, therefore, has

If the relative distance between the nodes keeps changing slowly, it is considered that the link stability between the two is good, so the variance of the above formula is taken as the direct stability evaluation value of the link between the detection node i and the measured node j, namely:

${\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; \_</span>\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; =</span>\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}^{\mathrm{<span\; class="notranslate">\; i</span>}}}\mathrm{<span\; class="notranslate">\; var</span>}<span\; class="notranslate">\; \{</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; ij</span>}}}}<span\; class="notranslate">\; ,</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; ij</span>}}}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}^{\mathrm{<span\; class="notranslate">\; ij</span>}}}}<span\; class="notranslate">\; \}</span>$

Thus, the calculation formula of the direct link stability evaluation value is obtained, and the smaller the evaluation value is, the better the link stability is.

The link indirect stability evaluation value in the second step:

Assuming that the node X under test has n neighbors (Y ₁ , Y ₂ ... Y _n ), the calculation of the stability evaluation value of node X to its neighbor node Y _K is the same as the first step, namely:

${\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; K</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}^{\mathrm{<span\; class="notranslate">\; x</span>}}}\mathrm{<span\; class="notranslate">\; var</span>}<span\; class="notranslate">\; \{</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{{\mathrm{<span\; class="notranslate">\; X\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}}}}<span\; class="notranslate">\; ,</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{{\mathrm{<span\; class="notranslate">\; X\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; K</span>}}}}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}^{{\mathrm{<span\; class="notranslate">\; X\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}}}}<span\; class="notranslate">\; \}</span>$

Since the node X under test has n neighbors, and the indirect link stability is an evaluation of the environment of the node X under test, therefore, node X needs to calculate the stability evaluation value of its n neighbors, that is, S _X (Y ₁ ), S _X (Y ₂ )...S _X (Y _n ); in order to eliminate the influence of the most unstable nodes, sort these n stability nodes and take the smallest 0.5n values, assuming S _X (Y _T1 ), S _X (Y _T2 ), ... S _X (Y _T0.5n ), and then take the average of these 0.5n values as the evaluation value of the environment of the tested node X, that is, the indirect stability of the link The assessed value:

in_S _X ＝avg(S _X (Y _T1 ), S _X (Y _T2 ), ... S _X (Y _T0.5n ))

Thus, the calculation formula of the indirect stability evaluation value of the link is obtained. The smaller the evaluation value is, the more stable the network environment of the node under test is.

In the third step, evaluate the comprehensive stability of the link between the detection node i and the measured node j:

S _ij =w ₁ *d_S _ij +w ₂ *in_S _j (w ₁ +w ₂ =1)

The node j under test acts as both sender and receiver.

In the fourth step, the stability-based multicast routing protocol uses stability as the routing criterion. When a node generates a routing request message, a specific field is added to the routing request message to represent the stability evaluation value of the path passed. ; The stability evaluation value of a path is equal to the maximum value of the stability evaluation of all the links it passes through, that is, the stability of a path depends on the most unstable link; when the node receives the routing reply message, from all Among the paths, a path with the smallest stability evaluation value is selected to be activated.

The specific process is:

A. Group members join

A node needs to go through three stages to join a multicast group: route request stage, route reply stage and route activation stage. When broadcasting the route of the tree, it sends a RREQ message; the destination address field dest_addr of the message is the address of the multicast group, and the source address source_addr is the address of the node itself; an additional Path_Stable field is added to the RREQ message to record the node passed The stability evaluation value of the path of ;

When a node receives a RREQ message, it first checks whether the same RREQ has been processed, and if so, discards it, otherwise it calculates the comprehensive stability value of the last hop link according to the link stability evaluation method proposed above; then the calculated Compare the stable value with the Path_Stable field in the RREQ, update the Path_Stable field to the larger value of the two, and then continue to forward the RREQ message;

If a node is a member of a multicast group or has a route leading to a multicast tree, after receiving the RREQ, it will cache the RREQ for a period of time, because the same RREQ message may arrive at the node from different paths during this period, so the The node selects the RREQ message with the smallest Path_Stable field to send the routing reply message RREP;

When an intermediate node receives an RREP message, it creates a multicast routing table entry based on the RREP and marks the entry status as inactive, that is, Activated_Flag is not set, and updates its Path_Stable field, and then continues along the reverse direction created by the RREQ message The path forwards the RREP message;

A node may receive multiple RREP messages, but it only processes the updated RREP messages with larger RREP sequence numbers or smaller RREP messages with more stable Path_Stable fields, and simply discards other RREP messages;

When the RREP message arrives at the source node, that is, the node requesting to join the multicast group, the source node caches it for a period of time, and selects a route activation message with the smallest Path_Stable field from the received multiple RREP messages, that is, the MACT message, that is, it will reply in the route The status of the routing entries created in the stage is marked as activated, that is, the Activated_Flag is set, and other inactive routing entries are deleted by the corresponding nodes due to timeout;

B. Group members leave

When a group member wishes to leave the multicast tree, no action is taken if it is a routing node for other nodes. If it is a leaf node, a prune message Prune is unicast to its previous hop, and its previous hop node deletes the routing entry corresponding to the successor node. If the deletion causes the previous hop node to become a leaf node, Then continue to send pruning messages to its previous hop to perform pruning operations;

C. Link break

When a circuit break occurs on the multicast tree, the branch where the circuit break occurs becomes invalid; at this time, the upstream node deletes the invalid node from its routing entry, and then the downstream node is responsible for link repair; the downstream node sends RREQ and appends a multicast The group reconstruction message is used for circuit break repair; the multicast group reconstruction message contains the multicast group hop field Group_Hop_Cnt to record the hops between the disconnected node and the root node of the tree, and only the nodes whose hop count is not greater than this count can respond to the RREQ message, so as to avoid The routing loop generation is eliminated; when the RREP message is received, the routing activation process is the same as that described in the group member joining section;

D. Maintain the multicast tree

Each node regularly sends hello messages to maintain the neighbor relationship and update the receiving power list of neighbor nodes; in addition, the root node of each multicast tree regularly sends GroupHello messages, and the sequence number of the root node increases by 1; when the group When the broadcast tree members receive the GroupHello message, they send a MACT message to activate the route, thus ensuring that each node can obtain the latest routing information of the multicast tree;

E. Multicast tree merging

In the mobile environment, due to the mobility of nodes, the adhoc network may be divided into several disconnected areas, and each area maintains its own multicast tree; due to the mobility of nodes, two or more disconnected The area may be connected again. To eliminate confusion, the tree root node with the largest sequence number becomes the tree root node of the new multicast tree, while other tree root nodes stop sending GroupHello messages and wait for the reconstruction of the multicast tree.

Compared with the prior art, the method adopted in the present invention has the following advantages:

(1) The present invention calculates based on the signal strength of continuously received messages. When a new node joins, only when the new node maintains a neighbor relationship with the detection node for a period of time, the node can be included in the calculation range of link stability. It reduces the calculation instability caused by the addition of new nodes, and has high adaptability and stability.

(2) Compared with the method of simply evaluating the moving speed of nodes, the present invention does not rely on other third-party modules (such as GPS), and has higher accuracy.

(3) Different from the method of only evaluating the stability of the detection node, the present invention evaluates the link stability from two angles of the detection node and the node under test, and the evaluation result is comprehensive, credible and adaptable.

(4) The present invention has simple calculation, low calculation amount in the process of evaluating link stability, low requirement on hardware, and low energy consumption.

(5) Applying the link stability evaluation method to the stability-based multicast routing protocol enables the present invention to be widely used in actual needs, improves network stability, and then improves work efficiency.

Description of drawings

FIG. 1 is a schematic diagram of link stability calculation in an embodiment of the present invention.

Fig. 2 is a flow chart of the group member joining process in the stability-based multicast routing protocol.

detailed description

An embodiment of the present invention is given below and the present invention will be further described in conjunction with the accompanying drawings.

Figure 1 and Figure 2

(1) Direct link stability assessment

According to the Friis free-space propagation model, the ratio of received power (P _RX ) to transmitted power (P _TX ) is inversely proportional to the square of the transmission distance (R), namely $\frac{P_{\mathrm{RX}}}{P_{\mathrm{TX}}}\&Proportional;\frac{1}{R^2}$ or $R\&Proportional;\sqrt{\frac{P_{\mathrm{TX}}}{P_{\mathrm{RX}}}}.$

In the present invention, it is assumed that the lower limit of the received power is P _min , that is, when the received power is lower than this value, a higher bit error rate will be generated, resulting in the retransmission of the data packet. Correspondingly, the transmission distance at this time is R _max . Due to the different environments of different nodes, the value of P _min is different. In the present invention, the P _min of all nodes is initialized to a fixed value. When a neighbor node moves out of its communication range, it can be considered that the last time successfully received The received power at time is approximately equal to the P _min at this time, so the received power at the previous time is used to update the P _min of the node.

As shown in Figure 1(a), assuming that the nth sampling value of the received power of the detected node j by the detection node i is P _n ^ij , then according to the Friis free-space propagation model, we have And because of Among them, P _TX is the transmission power of the node, R _n ^ij is the distance between the sending node and the receiving node at the nth power sampling, P _min ⁱ is the minimum receiving power of the detection node i, and R _max is when the receiving power is P _min ⁱ The corresponding maximum transmission distance, therefore, has

If the relative distance between the nodes changes slowly, it can be considered that the link stability between the two is good, so the variance of the above formula can be taken as the direct stability evaluation value of the link between the detection node i and the measured node j. which is:

${\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; \_</span>\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; =</span>\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}^{\mathrm{<span\; class="notranslate">\; i</span>}}}\mathrm{<span\; class="notranslate">\; var</span>}<span\; class="notranslate">\; \{</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; ij</span>}}}}<span\; class="notranslate">\; ,</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; ij</span>}}}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}^{\mathrm{<span\; class="notranslate">\; ij</span>}}}}<span\; class="notranslate">\; \}</span>$

From this, the calculation formula of the link direct stability evaluation value is obtained. The smaller the evaluation value, the better the link stability.

(2) Link indirect stability assessment

As shown in Figure 1(b), the node X under test has n neighbors (Y ₁ , Y ₂ ...Y _n ). The calculation of _K stability evaluation value of node X to its neighbor node Y is the same as (1), that is:

${\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; K</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}^{\mathrm{<span\; class="notranslate">\; x</span>}}}\mathrm{<span\; class="notranslate">\; var</span>}<span\; class="notranslate">\; \{</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{{\mathrm{<span\; class="notranslate">\; X\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}}}}<span\; class="notranslate">\; ,</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{{\mathrm{<span\; class="notranslate">\; X\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; K</span>}}}}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}^{{\mathrm{<span\; class="notranslate">\; X\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}}}}<span\; class="notranslate">\; \}</span>$

Since the node X under test has n neighbors, and the indirect link stability is an evaluation of the environment of the node X under test, therefore, node X needs to calculate the stability evaluation value of its n neighbors, that is, S _X (Y ₁ ), S _X (Y ₂ )...S _X (Y _n ). In order to eliminate the influence of the most unstable node (link), the present invention sorts the n stability and takes the smallest 0.5n values, assuming S _X (Y _T1 ), S _X (Y _T2 ), .. .S _X (Y _T0.5n ), and then take the mean value of these 0.5n values as the evaluation value of the environment of the tested node X, that is, the indirect stability evaluation value of the link:

in_S _X ＝avg(S _X (Y _T1 ), S _X (Y _T2 ), ... S _X (Y _T0.5n ))

Thus, the calculation formula of link indirect stability evaluation value is obtained. The smaller the evaluation value, the more stable the network environment where the tested node is located.

(3) Comprehensive link stability assessment

The direct and indirect stability value of comprehensive link, the present invention adopts following method to evaluate the comprehensive stability of link between detection node i and measured node j:

S _ij =w ₁ *d_S _ij +w ₂ *in_S _j (w ₁ +w ₂ =1)

It can be seen that when the present invention calculates the stability evaluation value, the measured node j acts as both a sender and a receiver.

(4) Multicast routing protocol based on stability

A. Group members join

It can be seen from Fig. 2 that a node needs to go through three stages to join a multicast group: a route request stage, a route reply stage and a route activation stage.

When a node wants to join a multicast group or has a message to send to a multicast group, but it is not a member of the group and has no route to the multicast tree, it sends a RREQ message. The destination address field dest_addr of the message is the address of the multicast group, and the source address source_addr is the address of the node itself. In the present invention, an additional Path_Stable field is added to the RREQ message to record the stability evaluation value of the path that the node passes.

When a node receives a RREQ message, it first checks whether the same RREQ has been processed, and discards it if so, otherwise calculates the comprehensive stability value of the last-hop link according to the link stability evaluation method proposed above. Then compare the calculated stable value with the Path_Stable field in the RREQ, update the Path_Stable field to the larger value of the two, and then continue to forward the RREQ message.

If a node is a member of a multicast group or has a route leading to a multicast tree, after receiving the RREQ, it will cache the RREQ for a period of time, because the same RREQ message may arrive at the node from different paths during this period, so the The node selects the RREQ message with the smallest Path_Stable field to send the routing reply message RREP.

When an intermediate node receives the RREP message, it creates a multicast routing table entry based on the RREP and marks the entry state as inactive (Activated_Flag is not set), and updates its Path_Stable field at the same time, and then continues along the reverse path created by the RREQ message Forward the RREP message to the path.

A node may receive multiple RREP messages, but it only processes the updated (larger RREP sequence number) or more stable (smaller Path_Stable field) RREP messages, and simply discards other RREP messages.

When the RREP message arrives at the source node (the node requesting to join the multicast group), the source node caches it for a period of time, and selects a Path_Stable field minimum for routing activation (MACT message) from multiple RREP messages received. The status of the routing entries created in the routing reply stage is marked as activated (Activated_Flag is set), and other inactive routing entries are deleted by the corresponding nodes due to timeout.

B. Group members leave

When a group member wishes to leave the multicast tree, no action is taken if it is a routing node for other nodes. If it is a leaf node, a prune message Prune is unicast to its previous hop, and its previous hop node deletes the routing entry corresponding to the successor node. If the deletion causes the previous hop node to become a leaf node, Then continue to send a pruning message to the previous hop to perform the pruning operation.

C. Link break

When a disconnection occurs on the multicast tree, the branch where the disconnection occurs becomes invalid. At this point, the upstream node removes the invalid node from its routing entry, and then the downstream node is responsible for link repair. The downstream node sends RREQ and attaches a multicast group reconstruction message to repair the broken circuit. Include the multicast group hop count field (Group_Hop_Cnt) in the multicast group reconstruction message to record the hop count of the disconnected node from the root node of the tree, and only the nodes whose hop count is not greater than the count can respond to the RREQ message, thus avoiding the routing loop produce. After receiving the RREP message, the routing activation process is the same as that described in the group member joining section.

D. Maintain the multicast tree

In this protocol, each node regularly sends hello messages to maintain the neighbor relationship and update the receiving power list of neighbor nodes. In addition, the root node of each multicast tree regularly sends GroupHello messages, and the sequence number of the root node increases by 1 at the same time. When the members of the multicast tree receive the GroupHello message, they send a MACT message to activate the route, thereby ensuring that each node can obtain the latest routing information of the multicast tree.

E. Multicast tree merging

In a mobile environment, due to the mobility of nodes, the adhoc network may be divided into several disconnected areas, and each area maintains its own multicast tree. Due to the mobility of nodes, two or more disconnected areas may be connected again. In order to eliminate confusion, the tree root node with the largest sequence number becomes the tree root node of the new multicast tree, while other tree root nodes stop sending GroupHello message and wait for the rebuilding of the multicast tree.

Thus, the network node has completed the establishment and maintenance of the multicast routing tree according to the link stability evaluation method of the present invention, and the simulation results of the embodiments of the present invention show that, compared with other existing technologies, the method of the present invention adopts the method of the present invention. In the mobile environment, it has the highest packet delivery rate, low end-to-end delay and high network throughput, and the overall performance of the network is stable and adaptable.

Claims (5)

1. A method for predicting link stability in an AdHoc network, which realistically approaches the real environment of the network by sampling continuously received signal strengths, and evaluates each item of its environment at a detection node and a measured node respectively The stability of the link, comprehensively consider the detection node and the measured node pair, calculate the comprehensive stability value, and use this as the routing standard of the stable multicast routing protocol, it is characterized in that the specific steps are: The first step is to maintain the neighbor relationship by regularly sending hello messages to link stability assessment nodes. When the detection node receives the hello message sent by its neighbor node, the detection node maintains a receiving node for each neighbor node, that is, the node under test. Signal strength is the received power list; In the second step, link indirect stability assessment The node under test uses the same method as the first step to evaluate the stability of its neighbor nodes; since this stability is aimed at evaluating the environment in which the node under test is located, moving After the node with particularly high stability, take the average value of the stability of other nodes, the lower the average value, the more stable the environment of the tested node, and vice versa, the more unstable the environment; The third step is the comprehensive stability assessment of the link, which is to comprehensively calculate the direct and indirect stability assessment values of the link using the method of weighted summation; In the fourth step, the stability-based multicast routing protocol uses stability as the routing criterion; The link direct stability evaluation value in the first step: According to the Friis free-space propagation model, the ratio of received power (P _RX ) to transmitted power (P _TX ) is inversely proportional to the square of the transmission distance (R), namely or Assume that the lower limit of the received power is P _min , that is, when the received power is lower than this value, a higher bit error rate will be generated, resulting in the retransmission of the data packet, correspondingly, the transmission distance at this time is R _max ; due to different nodes The environment is different, so the value of P _min is different. The P _min of all nodes is initialized to a fixed value. When a neighbor node moves out of its communication range, it can be considered that the received power at the time of successful reception at the previous moment is approximately equal to P _min of the node, so update the P _min of the node with the received power at the last moment; Assume that the nth sampling value of the received power of the detection node i to the measured node j is According to the Friis free space propagation model, we have And because of Among them, P _TX is the transmission power of the node, R _n is the distance between the sending node and the receiving node at the nth power sampling, P _min ⁱ is the minimum receiving power of the detection node i, and R _max is the corresponding when the receiving power is P _min ⁱ The maximum transmission distance, therefore, have $\frac{R_{\mathrm{no}}}{R_{\max }}=\sqrt{\frac{P_{\min }^i}{P_{\mathrm{no}}^{\mathrm{ij}}}}$ If the relative distance between the nodes keeps changing slowly, it is considered that the link stability between the two is good, so the variance of the above formula is taken as the direct stability evaluation value of the link between the detection node i and the measured node j, namely: $\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; \_</span>{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; =</span>\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}^{\mathrm{<span\; class="notranslate">\; i</span>}}}\mathrm{<span\; class="notranslate">\; var</span>}<span\; class="notranslate">\; \{</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; ij</span>}}}}<span\; class="notranslate">\; ,</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; ij</span>}}}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}^{\mathrm{<span\; class="notranslate">\; ij</span>}}}}<span\; class="notranslate">\; \}</span>$ Thus, the calculation formula of the direct link stability evaluation value is obtained, and the smaller the evaluation value is, the better the link stability is. 2. the method for predicting link stability in the AdHoc network as claimed in claim 1, is characterized in that, in the second step, link indirect stability evaluation value: Assuming that the node X under test has n neighbors (Y ₁ , Y ₂ ... Y _n ), the calculation of the stability evaluation value of node X to its neighbor node Y _K is the same as the first step, namely: ${\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; K</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}^{\mathrm{<span\; class="notranslate">\; x</span>}}}\mathrm{<span\; class="notranslate">\; var</span>}<span\; class="notranslate">\; \{</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{{\mathrm{<span\; class="notranslate">\; X\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; K</span>}}}}}<span\; class="notranslate">\; ,</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{{\mathrm{<span\; class="notranslate">\; X\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; K</span>}}}}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}^{{\mathrm{<span\; class="notranslate">\; X\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; K</span>}}}}}<span\; class="notranslate">\; \}</span>$ In order to detect the minimum received power of node X, since the node X under test has n neighbors, and the indirect link stability is an evaluation of the environment where the node X under test is located, node X needs to calculate the stability of its n neighbors Evaluation values, that is, S _X (Y ₁ ), S _X (Y ₂ )…S _X (Y _n ); in order to eliminate the influence of the most unstable nodes, sort these n stability and take the smallest 0.5n values , assuming S _X (Y _T1 ),S _X (Y _T2 ),…S _X (Y _T0.5n ), and then take the average of these 0.5n values as the evaluation value of the environment where the node X under test is located, that is, the chain The indirect stability evaluation value of the road: in_S _X ＝avg(S _X (Y _T1 ),S _X (Y _T2 ),…S _X (Y _T0.5n )) Thus, the calculation formula of the indirect stability evaluation value of the link is obtained; the smaller the evaluation value is, the more stable the network environment of the node under test is. 3. link stability prediction method in AdHoc network as claimed in claim 1, is characterized in that, in the described 3rd step, evaluates the comprehensive stability of link between detection node i and measured node j: S _ij =w ₁ *d_S _ij +w ₂ *in_S _j (w ₁ +w ₂ =1) The node j under test acts as both a sender and a receiver, and in_S _j is the indirect stability evaluation value of the link of node j. 4. link stability prediction method in AdHoc network as claimed in claim 1, it is characterized in that, in the described 4th step, based on the multicast routing protocol of stability, take stability as routing standard, when node produces route request When sending a message, a specific field is added to the routing request message to represent the stability evaluation value of the path passed; the stability evaluation value of a path is equal to the maximum value of the stability evaluation of all the links passed through, that is, the stability evaluation value of a path The stability depends on the most unstable link; when the node receives the routing reply message, it selects a path with the smallest stability evaluation value from all paths to activate. 5. link stability prediction method in AdHoc network as claimed in claim 4, is characterized in that, concrete process is: A. Group members join Nodes need to go through three stages to join the multicast group: route request stage, route reply stage and route activation stage; When a node wants to join a multicast group or has a message to send to a multicast group, but it is not a member of the group and has no route to the multicast tree, it sends a RREQ message; the destination address field dest_addr of the message is the multicast group address, and the source address source_addr is the address of the node itself; an additional Path_Stable field is added to the RREQ message to record the stability evaluation value of the path passed by the node; When a node receives a RREQ message, it first checks whether the same RREQ has been processed, and if so, discards it, otherwise it calculates the comprehensive stability value of the last hop link according to the link stability evaluation method proposed above; then the calculated Compare the stable value with the Path_Stable field in the RREQ, update the Path_Stable field to the larger value of the two, and then continue to forward the RREQ message; If a node is a member of a multicast group or has a route leading to a multicast tree, after receiving the RREQ, it will cache the RREQ for a period of time, because the same RREQ message may arrive at the node from different paths during this period, so the The node selects the RREQ message with the smallest Path_Stable field to send the routing reply message RREP; When an intermediate node receives an RREP message, it creates a multicast routing table entry based on the RREP and marks the entry status as inactive, that is, Activated_Flag is not set, and updates its Path_Stable field, and then continues along the reverse direction created by the RREQ message The path forwards RREP messages; a node may receive multiple RREP messages, but it only processes RREP messages with larger RREP sequence numbers or smaller Path_Stable fields, and simply discards other RREP messages; When the RREP message arrives at the source node, that is, the node requesting to join the multicast group, the source node caches it for a period of time, and selects a route activation message with the smallest Path_Stable field from the received multiple RREP messages, that is, the MACT message, that is, it will reply in the route The status of the routing entries created in the stage is marked as activated, that is, the Activated_Flag is set, and other inactive routing entries are deleted by the corresponding nodes due to timeout; B. Group members leave When a group member wishes to leave the multicast tree, if it is a routing node of other nodes, no action will be taken; if it is a leaf node, a prune message Prune will be unicast to its previous The jump node deletes the routing entry corresponding to the successor node. If the previous hop node also becomes a leaf node after deletion, it will continue to send a pruning message to the previous hop to perform the pruning operation; C. Link break When a circuit break occurs on the multicast tree, the branch where the circuit break occurs becomes invalid; at this time, the upstream node deletes the invalid node from its routing entry, and then the downstream node is responsible for link repair; the downstream node sends RREQ and appends a multicast The group reconstruction message carries out circuit breakage repair; Include multicast group hop count field Group_Hop_Cnt in the multicast group reconstruction message to record the hop count of the disconnected node from the root node of the tree, only the node whose hop count is not greater than the hop count can respond to the RREQ message, like this Avoiding the generation of routing loops; after receiving the RREP message, the routing activation process is the same as that described in the group member joining section; D. Maintain the multicast tree Each node regularly sends hello messages to maintain the neighbor relationship and update the receiving power list of neighbor nodes; in addition, the root node of each multicast tree regularly sends GroupHello messages, and the sequence number of the root node increases by 1; when the group When the broadcast tree members receive the GroupHello message, they send a MACT message to activate the route, thus ensuring that each node can obtain the latest routing information of the multicast tree; E. Multicast tree merging In the mobile environment, due to the mobility of nodes, the adhoc network may be divided into several disconnected areas, and each area maintains its own multicast tree; due to the mobility of nodes, two or more disconnected The area may be connected again. To eliminate confusion, the tree root node with the largest sequence number becomes the tree root node of the new multicast tree, while other tree root nodes stop sending GroupHello messages and wait for the reconstruction of the multicast tree.