CN110139304B

CN110139304B - MPR set selection method based on isolation degree and non-MPR nodes

Info

Publication number: CN110139304B
Application number: CN201910391162.5A
Authority: CN
Inventors: 裴文江; 王宇飞; 夏亦犁
Original assignee: Southeast University
Current assignee: Southeast University
Priority date: 2019-05-11
Filing date: 2019-05-11
Publication date: 2022-06-10
Anticipated expiration: 2039-05-11
Also published as: CN110139304A

Abstract

The invention provides an MPR set selection method based on isolation and non-MPR nodes, belonging to the field of wireless Mesh network protocols. Aiming at the MPR selection algorithm in the standard OLSR protocol, the problem that the global optimal MPR set can not be solved, and simultaneously, a large number of TC groups are generated under the environment that a large number of nodes are accessed, so that network congestion and load imbalance are caused.

Description

MPR set selection method based on isolation degree and non-MPR nodes

Technical Field

The invention relates to the field of wireless Mesh network communication, in particular to an MPR set selection method based on isolation and non-MPR nodes.

Background

With the rapid development of the Internet (Internet), Internet of Things (IoT), vehicle networking (vanehlar ad-hoc networks, VANETS) and other ad hoc networks, the development of various wireless transmission network technologies is driven, and people's demands and expectations for networks are higher and higher, and not only people want to connect to the Internet at any time and place and stably, but also want to have diversified network services. Various Network developments such as Wireless Local Area Network (WLAN), Wireless Broadband Access Network (WBAN), and Worldwide Interoperability for Microwave access (WiMAX) can meet most of the needs of people, but for the occasions such as major social events in cities, large-scale event guarantees, disaster relief, etc., the standard communication system cannot completely meet the actual emergency deployment and control needs, and the guarantee of the activities is very critical to the normal operation of the society, so that the design and optimization of the Network suitable for the activities are very important.

In the OLSR protocol, broadcast message flooding in a network can be effectively reduced by selecting the MPR node, the defect of high network maintenance cost is overcome, and the method is widely applied to a network environment with high density. Since the number of TC messages generated in the network is proportional to the MPR nodes, many researchers research MPR selection algorithms in the standard OLSR protocol, which use greedy algorithm to find MPR sets, which is a problem difficult in NP, so that the obtained result can be generally close to a global optimal solution rather than a global optimal solution. The selection of the MPR set is very important and has a significant impact on network performance.

In order to reduce the probability of obtaining a local optimal solution and the addition of unnecessary MPR nodes to an MPR set, Li et al propose an nfa (throughput First of selecting algorithm) algorithm to select MPR nodes, and the core is to delete one-hop neighbor nodes covering the minimum number of two-hop neighbor nodes when there is no two-hop neighbor nodes that can only be connected by one-hop neighbor nodes, that is, these nodes cannot obtain the chance of becoming MPR nodes, at this time, the number of the selected one-hop neighbor nodes is reduced, and the necessity of the remaining one-hop neighbor nodes to become the two-hop neighbor nodes is increased.

Compared with the standard OLSR protocol, the algorithm has the greatest difference that one-hop neighbor nodes with the least coverage number are deleted, and the method has the main improvement that MPR nodes are not selected according to the sequence of the capability of the coverage nodes in the standard algorithm, so that the condition of cross among the nodes can be reduced, and the standard algorithm is improved to a certain extent.

Although the NFA algorithm has the advantages described above, when two candidate nodes have the same chance to join the MPR set of the master node, one node is randomly selected to join, and in this case, the selected node does not necessarily have global optimality. Therefore, the invention provides an MPR selection algorithm based on isolation and non-MPR nodes, and the MPR nodes selected each time are ensured to be a global optimal solution.

Disclosure of Invention

In order to solve the problem that broadcast message flooding in a network can be effectively reduced by selecting an MPR node in an OLSR protocol, the defect of high network maintenance cost is overcome, and the method is widely applied to a network environment with high density. Since the number of TC messages generated in the network is proportional to the MPR nodes, many researchers in the years research MPR selection algorithms, the MPR selection algorithm in the standard OLSR protocol uses a greedy algorithm to search the MPR set, which belongs to the problem of NP difficulty, the obtained result can be generally close to a global optimal solution rather than the global optimal solution, although some heuristic algorithms are researched and applied to MPR selection, but the heuristic algorithms are complex, if the MPR selection algorithm is complex, the providing of an approximate MPR set optimal solution is difficult, the invention provides an MPR set selection method based on isolation and non-MPR nodes, ensures that the MPR node is always the global optimal solution when being selected each time, thereby reducing the generation of TC messages in the whole network, reducing WMN link conflicts, and achieving the purpose, the invention provides the MPR set selection method based on isolation and non-MPR nodes, the method comprises the following steps:

(1) detecting a one-hop neighbor set N (u) and a two-hop neighbor set N2(u) of the master node, and calculating the connection number of each node in the N (u);

(2) adding the node with the highest N _ willingness and non-zero reachability into the MPR set;

(3) adding nodes connected with a larger number of nodes in N2(u) into the MPR set;

(4) adding D (y) larger nodes to the MPR set;

(5) and adding nodes with larger isolation degree or non-MPR degree into the MPR set.

As a further improvement of the structure of the present invention, in the step (1), a one-hop neighbor set N (u) and a two-hop neighbor set N2(u) of the master node are detected, and the specific steps are as follows:

step 2.1: the relation between the nodes N (u) and N2(u) is checked, and if a node in N (u) is the only node connecting a node in N2(u), the node is added to the MPR set of the master node.

As a further improvement of the structure of the present invention, in the step (2), the node with the highest N _ willingness and non-zero reachability is added to the MPR set, and the specific steps are as follows:

step 3.1: and (3) executing the step 2.1, selecting the N (u) node which is not uniquely connected with the N2(u), selecting the node with the maximum N _ willingness in the neighbor information table, adding the node into the MPR set according to whether the N _ STATUS of the node in the neighbor information table is STATUS _ SYM or not, and deleting the node in the N2(u) covered by the node.

As a further improvement of the structure of the present invention, in the step (3), the nodes connected to a larger number of nodes in N2(u) are added to the MPR set, and the specific steps include:

step 4.1: if all the neighbor nodes N _ willingness are the same, the MPR node cannot be selected, and the nodes connected with the maximum number of N2(u) nodes in N (u) are compared and added into the MPR set.

As a further improvement of the structure of the present invention, in the step (4), a node with a larger size of d (y) is added to the MPR set, and the specific steps are as follows:

step 5.1: if all nodes in N (u) are connected with the same number of N2(u) nodes, the MPR node cannot be selected, and the node with larger value D (y) is added into the MPR set, wherein D (y) refers to the number of nodes which the main node can only be connected with the N2(u) node through the node.

As a further improvement of the structure of the present invention, in the step (5), the node with larger isolation or non-MPR degree is added to the MPR set, and the specific steps are as follows:

step 6.1: if there are multiple same highest values of d (y), MPR node can not be selected, because HELLO message in standard OLSR protocol does not provide source node three-hop neighbor node information, format of HELLO message is modified, Reserved field is divided into two parts, each part is one byte, the first part is still Reserved field, the second part is newly added Degree field for representing isolation or non-MPR Degree;

the Reversed field in the table must be set to '0000000000000', conforming to the compilation specification; the depth field is used to indicate the Degree of isolation or non-MPR Degree; the Htime field is the publication interval of the HELLO message, i.e., the time from the next transmission of the HELLO message; willingness refers to the Willingness of a node to carry and send to other nodes, if the field of one point is WILL _ NEVER which cannot be selected as MPR by other points, if the field of one point is WILL _ ALWAYS, the field of the DEFAULT node is WILL _ DEFAULT; the Link Code field refers to information of links between the sending terminal interface and all the neighbor interfaces of the sending terminal interface and also refers to information of neighbor states, general Link types comprise asymmetric links, symmetric links, failure links and default conditions, and the neighbor types comprise MPR nodes, symmetric neighbor nodes and asymmetric neighbor nodes; the LinkMessage Size field refers to the Size of the Link information, counts in bytes, measures the Size of the Link information by the distance from the beginning of the Link Code domain to the end of the next Link Code domain, and indicates the end of the message if no other Link type exists; the Neighbor Interface Address represents the Interface Address of the Neighbor node;

step 6.2: meanwhile, a new field N _ degree needs to be added in the neighbor information table to store the isolation degree and non-MPR degree information of the neighbor nodes, when a node receives a HELLO message from its neighbor node, the node updates the N _ degree field of the neighbor node stored in the neighbor information table, when a node has to select its MPR, the node selects the node with higher N _ degree value in the neighbor node as the MPR, and all symmetric two-hop neighbor nodes can be reached through the MPR;

n _ neighbor _ main _ addr in the table represents the main address of the neighbor node; n _ STATUS represents the state of a neighbor node and indicates whether links among the nodes are symmetrical or NOT, the symmetry in the protocol is represented as STATUS _ SYM, and the symmetry is NOT referred to as STATUS _ NOT _ SYM; n _ willingness is represented by an integer between 0 and 7, and represents willingness of neighbor nodes to forward TC packets for other nodes in the network, OLSR _ wide _ NEVER is 0, OLSR _ wide _ DEFAULT is 3, and OLSR _ wide _ ALWAYS is 7 in the protocol specification; n _ degree to hold the isolation and non-MPR degree information of the neighbor nodes.

As a further improvement of the structure of the invention, the NS3 network simulator uses randomly distributed network routers in a 1500 × 1500 square area, the simulation has 50 routers at the beginning, the number of the routers is continuously increased along with the increase of time until 100 routers are reached, the size of a data packet is 512bytes, and the flow model adopts CBR and UDP protocols.

The beneficial effect of this application is as follows:

the invention carries out simulation test on NS3 network simulator, uses the randomly distributed network routers in 1500 x 1500 square region, 50 routers are available at the beginning of simulation, the number of routers is continuously increased along with the time increase until 100 routers are reached, the specific setting parameter shows that when 100 router nodes are all added into the network, the number of TC data packets is reduced by about 12.5% by adopting the independence degree algorithm compared with the standard algorithm, and the number of TC data packets is reduced by about 13.7% by adopting the non-MPR degree algorithm compared with the standard algorithm, on the other hand, the total throughput of the algorithm based on the isolation degree and the non-MPR nodes is respectively improved by 16.5% and 14.6% by adopting the algorithm based on the isolation degree and the non-MPR nodes compared with the standard MPR selection algorithm, and the packet delivery success rate is improved by about 6% compared with the standard OLSR protocol MPR selection algorithm.

Drawings

FIG. 1 is an overall process based on isolation and non-MPR node algorithms;

FIG. 2 is an exemplary graph based on an isolation algorithm;

fig. 3 is an exemplary diagram based on a non-MPR node algorithm.

Detailed Description

The invention is described in further detail below with reference to the following detailed description and accompanying drawings:

the invention provides an MPR set selection method based on isolation and non-MPR nodes, which ensures that MPR nodes selected each time are always the global optimal solution, thereby reducing the generation of TC messages in the whole network and reducing WMN link conflicts.

As a specific embodiment of the present invention, an MPR set selection method based on isolation and non-MPR nodes is shown in fig. 1 of the overall flowchart specification, and includes the following steps:

step 1: detecting a one-hop neighbor set N (u) and a two-hop neighbor set N2(u) of the main node, and calculating the connection number of each node in the N (u).

The relation between the nodes N (u) and N2(u) is checked, and if a node in N (u) is the only node connecting a node in N2(u), the node is added to the MPR set of the master node.

Step 2: and (3) selecting the node with the maximum N _ willingness in the neighbor information table from the nodes after the step 1 is executed, adding the node into the MPR set if the N _ STATUS of the node in the neighbor information table is the STATUS _ SYM, and deleting the node in the N2(u) covered by the node.

And step 3: if all the neighbor nodes N _ willingness in the step 2 are the same, the MPR node cannot be selected, and the nodes connected with the maximum number of N2(u) nodes in N (u) need to be added into the MPR set continuously.

And 4, step 4: if all the N (u) nodes in step 3 are connected to the same number of N2(u) nodes, the MPR node cannot be selected, and the node with the larger value of d (y) is added to the MPR set, where d (y) refers to the number of nodes through which the master node can only connect to the N2(u) node.

And 5: if there are multiple same highest d (y) values in step 4, the MPR node cannot be selected, because the HELLO message in the standard OLSR protocol does not provide the source node three-hop neighbor node information, the format of the HELLO message is modified, the Reserved field is divided into two parts, each part is one byte, the first part is still the Reserved field, and the second part is a newly added Degree field for indicating the isolation or the non-MPR Degree. The format of the extended HELLO message is shown in table 1.

Table 1 extended HELLO message format

Step 6: a new field N _ degree needs to be added in the neighbor information table to store the isolation and non-MPR degree information of the neighbor nodes based on step 5, when a node receives HELLO message from its neighbor node, it will update the N _ degree field of the neighbor node stored in the neighbor information table, when a node has to select its MPR, it will select the higher N _ degree value of the one-hop neighbor node as MPR, based on the isolation algorithm as shown in fig. 2, based on the non-MPR degree algorithm as shown in fig. 3, through which all symmetric two-hop neighbor nodes can be reached. The extended neighbor information is shown in table 2.

Table 2 extended neighbor information table

N_neighbor_main_addr

N_status

N_willingness

N_degree

The parameter environment settings are shown in table 3.

Table 3 simulation environment parameter settings

The simulation results are shown in table 4.

Thus, the training and reasoning process of the MPR set selection method based on the isolation degree and the non-MPR nodes is completed.

The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention in any way, but any modifications or equivalent variations made according to the technical spirit of the present invention are within the scope of the present invention as claimed.

Claims

1. An MPR set selection method based on isolation and non-MPR nodes is characterized in that: the method comprises the following steps:

in the step (3), the nodes connected to a larger number of nodes in N2(u) are added to the MPR set, and the specific steps include:

step 4.1: if all the neighbor nodes N _ willingness are the same, the MPR node cannot be selected, and nodes connected with the maximum number of N2(u) nodes in N (u) are compared and added into the MPR set;

(4) adding D (y) larger nodes to the MPR set;

in the step (4), adding a node with a larger size of d (y) to the MPR set, specifically including the steps of:

step 5.1: if all nodes in N (u) are connected with the same number of N2(u) nodes, the MPR node cannot be selected, and the node with the larger value of D (y) is added into the MPR set, wherein D (y) refers to the number of nodes which can be connected to the N2(u) node through the node;

(5) adding nodes with larger isolation or non-MPR degrees into an MPR set;

in the step (5), the nodes with larger isolation or non-MPR degree are added into the MPR set, and the specific steps are as follows:

the Reversed field in the table must be set to '0000000000000', conforming to the compilation specification; the depth field is used to indicate the Degree of isolation or non-MPR Degree; the Htime field is the publication interval of the HELLO message, i.e., the time from the next transmission of the HELLO message; willingness refers to the Willingness of a node to carry and send to other nodes, if the field of one point is WILL _ NEVER which cannot be selected as MPR by other points, if the field of one point is WILL _ ALWAYS, the field of the DEFAULT node is WILL _ DEFAULT; the Link Code field indicates information of links between the sending end interface and all the neighbor interfaces of the sending end interface and also indicates information of neighbor states, general Link types comprise asymmetric links, symmetric links, failure links and default conditions, and the neighbor types comprise MPR nodes, symmetric neighbor nodes and asymmetric neighbor nodes; the Link Message Size field refers to the Size of the Link information, counts in bytes, measures the Size of the Link information by the distance from the beginning of the Link Code domain to the end of the next Link Code domain, and indicates the end of the Message if no other Link type exists; the Neighbor Interface Address represents the Interface Address of the Neighbor node;

2. The method of MPR set selection based on orphan and non-MPR nodes of claim 1, wherein: in the step (1), a one-hop neighbor set N (u) and a two-hop neighbor set N2(u) of the master node are detected, and the specific steps are as follows:

3. The method of MPR set selection based on orphan and non-MPR nodes of claim 1, wherein: adding the node with the highest N _ willingness and non-zero reachability into the MPR set in the step (2), wherein the method specifically comprises the following steps:

4. The method of MPR set selection based on orphan and non-MPR nodes of claim 1, wherein: an experimental platform of a model established by an MPR set selection method based on an isolation degree and a non-MPR node is as follows: the NS3 network simulator uses randomly distributed network routers in 1500 × 1500 square area, the simulation has 50 routers at the beginning, the number of routers increases continuously with time until 100 routers are reached, the size of data packet is 512bytes, the flow model adopts CBR and UDP protocol.