CN110636458B - Wireless self-organizing network cooperation time synchronization optimization method - Google Patents

Wireless self-organizing network cooperation time synchronization optimization method Download PDF

Info

Publication number
CN110636458B
CN110636458B CN201910963249.5A CN201910963249A CN110636458B CN 110636458 B CN110636458 B CN 110636458B CN 201910963249 A CN201910963249 A CN 201910963249A CN 110636458 B CN110636458 B CN 110636458B
Authority
CN
China
Prior art keywords
node
master
slave
nodes
time
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910963249.5A
Other languages
Chinese (zh)
Other versions
CN110636458A (en
Inventor
陈桂芬
刘大鹍
王义君
王俊喜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Changchun University of Science and Technology
Original Assignee
Changchun University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Changchun University of Science and Technology filed Critical Changchun University of Science and Technology
Priority to CN201910963249.5A priority Critical patent/CN110636458B/en
Publication of CN110636458A publication Critical patent/CN110636458A/en
Application granted granted Critical
Publication of CN110636458B publication Critical patent/CN110636458B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04J—MULTIPLEX COMMUNICATION
    • H04J3/00—Time-division multiplex systems
    • H04J3/02—Details
    • H04J3/06—Synchronising arrangements
    • H04J3/0635—Clock or time synchronisation in a network
    • H04J3/0638—Clock or time synchronisation among nodes; Internode synchronisation
    • H04J3/0658—Clock or time synchronisation among packet nodes
    • H04J3/0661—Clock or time synchronisation among packet nodes using timestamps
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/06—Selective distribution of broadcast services, e.g. multimedia broadcast multicast service [MBMS]; Services to user groups; One-way selective calling services
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W56/00—Synchronisation arrangements
    • H04W56/001—Synchronization between nodes
    • H04W56/0015—Synchronization between nodes one node acting as a reference for the others
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

A wireless self-organizing network cooperation time synchronization optimization method relates to the technical field of wireless self-organizing networks, and solves the problems that the time synchronization precision, the energy consumption and the number of synchronization packets of the wireless self-organizing network need to be improved; determining a master node and a slave node by referring to node message broadcasting; the reference node broadcasts the message again, the main node carries out clock offset estimation and clock drift estimation, and adjusts the time of the main node; the master node sends a message to the slave nodes, the slave nodes reply to the packet, and the slave nodes perform clock offset estimation and clock drift estimation and adjust the time of the slave nodes. The invention achieves the performance improvement in the aspects of time synchronization precision, energy consumption and synchronization packet quantity.

Description

Wireless self-organizing network cooperation time synchronization optimization method
Technical Field
The invention relates to the technical field of wireless self-organizing networks, in particular to a method for optimizing the cooperation time synchronization of a wireless self-organizing network.
Background
Wireless Ad Hoc Networks (MANETs) are dynamic topology Networks composed of a plurality of movable nodes, and the Networks adopt a wireless communication mode and can be dynamically networked, have the characteristics of multi-hop communication, mobility and the like, and are different from the traditional cellular Networks. The wireless ad hoc network is widely used in the fields of remote areas, temporary conferences, local communication, military battlefields and the like because of the advantage that the wireless ad hoc network can be automatically networked without infrastructure.
In the wireless ad hoc network, there is no unified management of the infrastructure devices, so that a standard clock for providing a time reference for nodes in the network is lacked, and the time of each node is referenced to its own local clock. The uncertainty of the local clock exacerbates the lack of time synchronization between nodes. The research on the time synchronization technology is to synchronize the time of the terminal node in the wireless ad hoc network under the condition of meeting the requirement.
The wireless self-organizing network can carry out time synchronization in the message transmission process, and factors such as hardware conditions, communication interference, transmission delay and the like can generate great interference on the time synchronization precision, so that the transmitted data is greatly influenced. The self-organizing network node is powered by a battery, the energy is very limited, and the energy consumption is reduced as much as possible on the premise of improving the time synchronization precision. The sending end sends the time through the network card, and for the receiving end 1-receiving end 2 time synchronization method, the time delay of the receiving end is an important source of time synchronization error. The method has the greatest characteristic that the transmission time delay and the access time delay on the critical path are eliminated, and a transmission delay diagram is shown in figure 1. The time synchronization accuracy, power consumption and the number of synchronization packets of the wireless ad hoc network need to be improved to meet the demands and developments of the current networks.
Disclosure of Invention
In order to solve the above problems, the present invention provides a method for optimizing the cooperation time synchronization of a wireless ad hoc network.
The technical scheme adopted by the invention for solving the technical problem is as follows:
a wireless self-organizing network cooperation time synchronization optimization method comprises the following steps:
dividing a network coverage area into topological polygons, dividing each topological polygon into a plurality of monotonous polygons, and introducing nodes into the topological polygons; carrying out triangulation on the monotone polygon to obtain a plurality of triangular areas, wherein the positions of the mass center points of the triangular areas are nodes, and the nodes at the positions of the mass center points of the triangular areas are selected as reference nodes;
step two, the reference node is at the time stamp TM1Performing a round of message packet broadcasting, calculating self normalization quantization values of all nodes and comparing the normalization quantization values with a preset threshold value, taking the nodes with the normalization quantization values higher than the preset threshold value as main nodes, taking the other nodes as slave nodes, sending response packets to reference nodes by the main nodes, and sending the response packets to the reference nodes by the reference nodes at the time stamps TM4Receiving a response packet of the main node;
step three, the reference node performs a round of message packet broadcasting again, and the roundThe message packet includes TM1、TM4And local timestamp of the reference node, the main node receives the message packet broadcasted by the reference node, and the main node is according to TM1And TM4Calculating a master node clock offset estimate
Figure GDA0003004981370000021
And master node clock drift estimate
Figure GDA0003004981370000022
The master node generates a local timestamp based on the reference node,
Figure GDA0003004981370000023
And
Figure GDA0003004981370000024
the self time is adjusted, and the time of the self time is adjusted,
Figure GDA0003004981370000025
Figure GDA0003004981370000026
wherein the content of the first and second substances,
Figure GDA0003004981370000027
Figure GDA0003004981370000028
k is the current round of time synchronization, N is the total round of time synchronization, ω is the master node clock drift value,
Figure GDA0003004981370000029
is the master node clock offset value, tdFor fixed time delays in the cross-propagation process between the reference node and the master node, gammaS2Random noise, Γ, conforming to a Gaussian distribution when receiving a reference node message packet for a master nodeS3Random noise conforming to Gaussian distribution when sending a response packet for a master node;
step four, the main node is at the time stamp TS5Then sending message packet to slave node, the slave node selecting the master node corresponding to the first received message packet as master node of its master-slave synchronous network, the slave node sending response packet to the master node of its master-slave synchronous network, the master node being in time stamp TS8Receiving the response packet sent by the slave node, and sending the local timestamp and T of the master node by the master nodeS5And TS8To TS8Corresponding slave node according to TS5And TS8Calculating slave node clock offset estimates
Figure GDA00030049813700000210
And slave node clock drift estimate ω11The slave node according to the master node local time stamp,
Figure GDA0003004981370000031
And
Figure GDA0003004981370000032
the self time is adjusted, and the time of the self time is adjusted,
Figure GDA0003004981370000033
Figure GDA0003004981370000034
wherein the content of the first and second substances,
Figure GDA0003004981370000035
Figure GDA0003004981370000036
ω1is the slave node clock drift value and,
Figure GDA0003004981370000037
for slave node clock offset value, td1Fixed time delay, Γ, for master and slave node interaction propagation processP6To be a slave nodeRandom noise, Γ, conforming to a gaussian distribution when a point receives a message packetP7Random noise conforming to Gaussian distribution when the slave node sends out a response packet.
The invention has the beneficial effects that:
the method for optimizing the cooperation time synchronization of the wireless self-organizing network divides the network coverage area into topological polygons, namely, the multi-hop self-adaptive topological structure is established, so that the energy consumption of the network is reduced; triangulation is carried out on the network, a reference node is determined through a triangle centroid, then a main node is selected by the reference node, and the whole network node clock synchronization is realized by utilizing the self-adaptive change of a topological structure; the average synchronization error is reduced and the synchronization precision is improved by performing joint estimation through clock offset and clock drift; the overall implementation ensures node communication link time synchronization. With the increase of the number of network nodes, on the basis of clock joint estimation and self-adaptive triangulation topology control, the method improves the network overhead and the system stability in the whole execution and has better convergence. The invention achieves the performance improvement in the aspects of time synchronization precision, energy consumption and 3 synchronous packets.
Drawings
Fig. 1 is a schematic diagram of transmission delay.
Fig. 2 is a flowchart of a method for optimizing coordinated time synchronization of a wireless ad hoc network according to the present invention.
Fig. 3 is a diagram of a network topology.
Fig. 4 is a network topology structure diagram after triangulation.
Fig. 5 is a schematic diagram of reference node selection.
Fig. 6 is a clock skew statistical distribution graph.
Fig. 7 is a clock drift and clock skew model of the interaction of time information between the reference node and the master node.
Fig. 8 is a result of average synchronization error data obtained by simulation when a triangulation topology is employed.
Fig. 9 shows the average synchronization error experimental data of the four algorithms when the conventional clustering topology is adopted.
Fig. 10 shows the results of energy consumption data obtained by simulation when using the triangulation topology according to the present invention.
Fig. 11 is experimental data of energy consumption of the 4-class algorithm in the conventional clustering topology.
Fig. 12 is a simulation result of the number of CTSOA algorithm synchronization packets in two topologies.
Fig. 13 is a simulation result of synchronous convergence time of the CTSOA algorithm under two topologies.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
A method for optimizing the cooperation time synchronization of a wireless self-organizing network is disclosed, the general flow is shown in FIG. 2, and the method comprises the following steps:
dividing a network coverage area into topological polygons, dividing each topological polygon into a plurality of monotonous polygons, and introducing nodes into the topological polygons; and carrying out triangulation on the monotone polygon to obtain a plurality of triangular areas, wherein the positions of the mass center points of the triangular areas are nodes, and the nodes at the positions of the mass center points of the triangular areas are selected as reference nodes.
Dividing the network coverage area into topological polygons is to establish the multi-hop self-adaptive topological structure. Paired node synchronization needs to be extended to global multi-hop range network adaptive synchronization, and if full network synchronization is to be achieved, an effective network topology structure must be established. The invention realizes the self-adaptive synchronization of the nodes by constructing a local tree network based on the reference nodes, and the structural schematic of the self-adaptive tree network is shown in figure 3.
And (4) assuming that the network coverage topological polygonal area is C, and selecting a reference node by adopting a geometric calculation mode. The region C is divided into n monotone polygons in order to eliminate the inflection points caused in the case of polygon irregularities by introducing diagonal lines. The monotonous polygon division of the area C is completed, namely, the monotonous division of the polygon structure outside the wireless self-organizing network is completed, then the nodes in the network are sequentially introduced according to a random order, for example, 100 nodes with numbers of 1-100 are introduced in common, the nodes with the numbers of 1-100 are randomly disordered to obtain a random order, namely a new sequence, for example, the new sequence is '25, 11, 99, 3, 48, 50, 30, 82 … …', and then the nodes are sequentially introduced according to the new sequence (from front to back), namely, the node 25, the node 11, the node 99 and the node 3 … … are sequentially introduced. The execution of the program requires maintenance and updating of a triangulation corresponding to the current set of points. The monotone polygon is triangulated according to the introduced node positions to obtain a plurality of triangular regions, the positions of some of the introduced nodes are the positions of centroid points of the triangular regions obtained by triangulation, and it is further preferable that some of the introduced nodes are vertices of the triangular regions obtained by triangulation. By performing monotonous polygon division and triangulation on the external polygon structure of the coverage area C, fig. 4 shows a triangulation topology structure diagram divided in a planar area when there are 50 nodes in the wireless ad hoc network.
And after the triangulated topological structure is formed, nodes at the vertex of the triangle form a subset. In order to maximize the number of nodes in the subset that can be found as few as possible to accomplish the area coverage task, all nodes in the subset are colored with four colors, red, yellow, blue and green. The dyeing scheme is required to satisfy that the two nodes connected by any edge can not be dyed in the same color. Based on this, the triangulated polygons are dyed, where each triangle has and only has three different color vertices, e.g., a first vertex red, a second yellow, and a third blue. By the method, after the network completes triangulation, the node at the centroid point position in the well-divided triangle is selected as the reference node, the reference node needs to cover the communication distance of the triangle area, and the reference node selection schematic diagram is shown in fig. 5.
And step two, after the Reference node of each triangular area is selected through triangulation and dyeing, the Reference node is a main synchronization node of the area and is also a root node of the tree network, and the local time of the Reference node is taken as the Reference time of the network, so that a Reference node (Reference node) and Master node (Master node) synchronization network is established.
Reference node at time stamp TM1Carry out the firstAnd in turn, the nodes can receive one or more than one message packet by broadcasting the message packets, and the nodes calculate the normalization quantization values of the nodes according to the first message packet received by the nodes. (the normalization quantization value of the node is calculated by the node according to the time delay weight, the self link energy and the packet error rate.) the node compares the normalization quantization value with a preset threshold value, the preset threshold value is a uniform threshold value of all the nodes, the node with the normalization quantization value higher than the preset threshold value is taken as a main node, the node with the normalization quantization value not higher than the preset threshold value is taken as a slave node, and the main node packs a response packet in TS3The time for the master node to receive the message packet sent by the round of reference nodes is TS2Reference node at time stamp TM4And receiving a response packet of the main node.
Step three, the reference node performs a round of message packet broadcasting again, wherein the round of message comprises TM1、TM4And a local timestamp of the reference node (namely the local timestamp of the reference node when the information packet in the round is broadcasted), the main node receives the information packet broadcasted by the reference node, records the local timestamp of the reference node, and records the local timestamp of the reference node according to the TM1And TM4Calculating a master node clock offset estimate and a master node clock drift estimate,
Figure GDA0003004981370000061
Figure GDA0003004981370000062
wherein the content of the first and second substances,
Figure GDA0003004981370000063
is an estimate of the clock offset of the master node,
Figure GDA0003004981370000064
is a main node clock drift estimated value, k is the current round of time synchronization, N is the total round of time synchronization, omega is a main node clock drift value, and omega' is a main node auxiliary frequency offset factor
Figure GDA0003004981370000065
Figure GDA0003004981370000066
For the master node clock offset value,
Figure GDA0003004981370000067
k is the current round of time synchronization, N is the total round of time synchronization, tdFor fixed time delays in the cross-propagation process between the reference node and the master node, gammaS2Random noise, Γ, conforming to a Gaussian distribution when receiving a reference node message packet for a master nodeS3Random noise conforming to Gaussian distribution when sending a response packet for a master node;
and the master node adjusts the self time according to the local timestamp of the reference node, the master node clock offset estimation and the master node clock drift estimation.
And step four, establishing a synchronous network of the master node and the Slave node (Slave node), which is called a master-Slave synchronous network.
Master node at timestamp TS5The slave node selects the master node corresponding to the first received message packet as the master node of the master-slave synchronous network, the slave node establishes the master-slave synchronous network, and the time of receiving the first master node message packet by the slave node is recorded as TP6And the slave node records the message packet of the master node and the slave node of the synchronous network. The slave node sends a response packet to the master node of the master-slave synchronous network, and the time of sending the response packet by the slave node is recorded as TP7At time stamp T by the master nodeS8And receiving the response packet sent from the node. The master node sends the local timestamp, T, of the master nodeS5And TS8To TS8Corresponding slave node according to TS5And TS8Calculating slave node clock offset estimates
Figure GDA0003004981370000068
And slave node clock drift estimate value ω11，
Figure GDA0003004981370000069
Figure GDA0003004981370000071
Wherein the content of the first and second substances,
Figure GDA0003004981370000072
ω′1is assisted by frequency deviation factors from nodes
Figure GDA0003004981370000073
ω1Is the slave node clock drift value and,
Figure GDA0003004981370000074
for slave node clock offset value, td1Fixed time delay, Γ, for master and slave node interaction propagation processP6For random noise conforming to a Gaussian distribution when receiving message packets from nodes, ΓP7In order to conform to the gaussian distribution of random noise when the reply packet is sent from the node,
and the slave node adjusts the self time according to the local timestamp of the master node, the clock offset estimation value of the slave node and the clock drift estimation value of the slave node.
And step three and step four are completed once, namely one round of time synchronization, and step three and step four are circulated until all rounds are completed, namely N times of synchronization are completed.
After the slave node in step four determines, if the slave node is the edge of the communication network at the master node of its master-slave synchronous network and the slave node is able to receive multiple reference node message packets and multiple master node message packets simultaneously, such a node is called a synchronous slave node. The synchronous auxiliary node can send the received multiple reference node message packets and the multiple main node message packets to the main node of the slave node master-slave synchronous network, the synchronous auxiliary node sends the received message packets sent by all other nodes to the main node of the slave node master-slave synchronous network, the slave node can carry out bidirectional communication with the main node to realize synchronization, the main function of the synchronous auxiliary node is used as a supplement of the main node synchronization, and the synchronous precision among the main nodes can be further coordinated. The synchronous auxiliary node can also send the received message packets of the multiple reference nodes and the message packets of the multiple main nodes to the reference nodes, the synchronous auxiliary node sends the received message packets sent by all other nodes to the reference nodes of the slave node master-slave synchronous network, the slave node can carry out two-way communication with the reference nodes to realize synchronization, and the synchronous auxiliary node has the main function of supplementing the synchronization of the reference nodes and can further coordinate the synchronization precision among the reference nodes.
In the above method, every time the reference node sends a message, all non-reference nodes within the range of the reference node communication network can receive the message of the reference node.
Clock offset estimation is a very important part of the time synchronization algorithm. In the prior art, experiments prove that the clock offset of the local time difference between any two nodes at the receiving end conforms to gaussian distribution of μ ═ 0 and σ ═ 11.1 μ s. The result of the optimal parameters obtained by simulation fitting of the invention is as follows: μ ═ 0.048 and σ ═ 11.221 μ s, and the statistical distribution plot is shown in fig. 6. The experimental result has an important guiding function for realizing the following calculation.
Estimation of clock offset and clock drift estimation:
fig. 7 shows a clock drift and clock offset model of the interaction of time information between the reference node and the master node. As shown in the figure, M is a reference node, S is a main node, and the reference node records a timestamp T of the kth round of information interactionM1、TM4And the main node records the timestamp T of the kth round of information interactionS2、TS3. The reference node transmits a message packet to the master node, which is at TS2The message packet is received and at TS3Sending out a response packet, the reference node is at TM4The response packet is received. In this process, the time model of the master node may be defined as follows:
Figure GDA0003004981370000081
Figure GDA0003004981370000082
wherein, tdFor fixed time delays in the cross-propagation process between the reference node and the master node, gammaS2Random noise, Γ, conforming to a Gaussian distribution when receiving a reference node message packet for a master nodeS3Random noise conforming to Gaussian distribution when a master node sends a response packet, omega is a master node clock drift value,
Figure GDA0003004981370000083
is the master node clock offset value.
From the above model, the present invention provides a general likelihood function form of exponential model estimation:
Figure GDA0003004981370000084
wherein the auxiliary function L is defined as follows:
Figure GDA0003004981370000085
the above a represents the total number of nodes in the topological polygon, and λ is a parameter of poisson distribution.
As can be seen from this functional form, if random noise is ignored, the only parameter related to the clock estimation is the fixed time delay tdThis parameter can be calculated by exchanging message packets, but it inevitably increases network power consumption. Therefore, in this section, based on the above model, we can estimate ω and ω under the condition that we consider it as an unknown quantity without calculating a fixed delay
Figure GDA0003004981370000086
The value is obtained.
Because of tdAnd
Figure GDA0003004981370000087
are unknown, so their domain of definition can be limited to:
Figure GDA0003004981370000091
Figure GDA0003004981370000092
Figure GDA0003004981370000093
the statistical factor E of the time from when the master node receives the reference node message packet to when it sends the reply packet can be expressed as (k is the current round of time synchronization, N is the total round of time synchronization):
Figure GDA0003004981370000094
order to
Figure GDA0003004981370000095
The primary node auxiliary frequency offset factor ω' may be obtained as follows:
Figure GDA0003004981370000096
order to
Figure GDA0003004981370000097
And ω' is the master node clock offset estimate
Figure GDA0003004981370000098
Figure GDA0003004981370000099
Order to
Figure GDA00030049813700000910
And is
Figure GDA00030049813700000911
Then the master node clock drift estimate may be obtained
Figure GDA00030049813700000912
Figure GDA00030049813700000913
The clock drift and clock offset model of time information interaction between the master node and the slave node can be the same as that in fig. 7, that is, the clock drift and clock offset model of time information interaction between the reference node and the master node are the same, and the slave node clock offset estimation and the slave node clock drift estimation can be obtained by adopting the general likelihood function form of exponential model estimation, the master node clock offset estimation and the master node clock drift estimation process.
S is a main node, P is a slave node, and the main node records a timestamp T of the kth round of information interactionS5、TS8Recording the time stamp T of the kth round of information interaction by the slave nodeP6、TP7. Master node at TS5Transmitting the message packet to the slave node, the slave node being at TP6The message packet is received and at TP7Sending out a response packet, the reference node is at TS8The response packet is received. In this process, the time model of the slave node can be defined as follows:
Figure GDA0003004981370000101
Figure GDA0003004981370000102
the invention provides a general likelihood function form of exponential model estimation:
Figure GDA0003004981370000103
as can be seen from this functional form, if random noise is ignored, the only parameter related to the clock estimation is the fixed time delay td1This parameter can be calculated by exchanging message packets, but it inevitably increases network power consumption. Therefore, in this section, based on the above model, without calculating a fixed delay, we consider it as an unknown quantity, under which we estimate ω1And
Figure GDA0003004981370000104
the value is obtained.
td1And
Figure GDA0003004981370000105
are unknown, so their domain of definition can be limited to:
Figure GDA0003004981370000106
Figure GDA0003004981370000107
Figure GDA0003004981370000108
statistical factor E of time from receiving master node message packet to sending response packet by slave node1：
Figure GDA0003004981370000109
Order to
Figure GDA00030049813700001010
Then the secondary frequency offset factor from the node is obtainedω1' the following:
Figure GDA00030049813700001011
order to
Figure GDA00030049813700001012
And omega1＝ω1', then the slave node clock offset estimate can be obtained
Figure GDA00030049813700001013
Figure GDA0003004981370000111
Order to
Figure GDA0003004981370000112
And is
Figure GDA0003004981370000113
Then the slave node clock drift estimate ω can be obtained11：
Figure GDA0003004981370000114
td1Fixed time delay, Γ, for master and slave node interaction propagation processP6For random noise, Γ, conforming to a Gaussian distribution when a slave node receives a master node message packetP7For random noise, omega, conforming to a Gaussian distribution when a reply packet is sent from a node1Is the slave node clock drift value and,
Figure GDA0003004981370000115
is the slave node clock offset value.
In the invention, the whole network is divided into areas by triangulation, a reference node is determined by a mass center, the reference node selects a main node, and the slave node and the main node realize synchronization. The time of the reference node is the reference time, and whether the network standard time can be accurately synchronized to the whole network is very important. The topological structure prevents factors such as sudden failure and interference of a single node and the like, thereby influencing the synchronization performance of the whole network. The invention achieves the performance improvement in the aspects of time synchronization precision, energy consumption and 3 synchronous packets. According to the method, the exponential delay model is utilized, the time synchronization likelihood function is constructed, the clock skew and the clock drift of the paired nodes are jointly estimated, the average synchronization error is reduced, and the synchronization precision is improved. Dividing the network coverage area into topological polygons is to establish a multi-hop self-adaptive topological structure, so that the energy consumption of the network is reduced. Triangulation is carried out on the network, a reference node is determined through a triangle centroid, then a main node is selected by the reference node, and the whole network node clock synchronization is achieved through self-adaptive change of a topological structure. The overall implementation ensures node communication link time synchronization. With the increase of the number of network nodes, on the basis of clock joint estimation and self-adaptive triangulation topology control, the method improves the network overhead and the system stability in the whole execution and has better convergence.
The simulation experiment of the invention verifies the time synchronization algorithm by arranging 500 self-organizing cooperative network nodes in a 500m multiplied by 500m area, wherein the specific simulation setting parameters are shown in table 1. The method for optimizing the Synchronization of the Wireless self-organizing network collaboration Time is called CTSOA (collaborative Time Synchronization Optimization Algorithm for Wireless Ad Hoc networks), and in order to verify the effectiveness of the CTSOA Algorithm provided by the invention, the CTSOA Algorithm is compared with an ESTSP (efficient Single Hop Time Synchronization protocol) Algorithm, an ERBS (Energy-efficiency references Synchronization) Algorithm and an RBS (RBS) (resource blocks Synchronization) Algorithm, on the premise that 4 algorithms are simultaneously placed in the simulation environment shown in the table 1 for experiment, and the triangulation topology provided by the invention is compared with the Synchronization parameters under the traditional clustering topology. Data processing and analysis are mainly performed in terms of average synchronization error, power consumption and the number of synchronization packets being 3.
TABLE 1 simulation parameters
Figure GDA0003004981370000121
Firstly, the invention carries out simulation verification on the average synchronization error of the 4-class algorithm, and the simulation result is shown in figures 8-9. Fig. 8 shows the result of the average synchronization error data obtained by simulation when the triangulation topology proposed by the present invention is adopted. The longer the synchronization period, the more accurate the synchronization timestamp data, and the smaller the synchronization error. Because the CTSOA algorithm adopts the joint estimation of clock drift and clock offset, compared with the other three algorithms, the average synchronization error is the lowest and is less than 100 mu s. The average synchronization error of the ESTSP algorithm is about 200-100 mus, the ERBS algorithm is about 300-200 mus, and the average synchronization error of the RBS algorithm is the largest and is larger than 600 mus. Fig. 9 is experimental data of average synchronization errors of four algorithms in a conventional clustering topological structure, and it can be known by comparing with fig. 8 that the average synchronization errors are increased compared with those in a triangulation topological structure. The average synchronization error obtained by simulation is the lowest when the triangulation topological structure provided by the invention is adopted.
Next, the energy consumption of the class 4 algorithm is verified by simulation, and the simulation result is shown in fig. 10. Fig. 10 shows the result of energy consumption data obtained by simulation when the triangulation topology proposed by the present invention is adopted. The larger the number of reference nodes, the larger the network size, and the higher the energy consumption. Because the CTSOA algorithm adopts self-adaptive topology control, compared with other three algorithms, the energy consumption is the lowest and is about 200 mW. The energy consumption of the ESTSP algorithm is slightly higher than that of the CTSOA algorithm, the energy consumption of the ERBS algorithm is about 400mW, and the energy consumption of the RBS algorithm is the highest and is between 500mW and 600 mW. Fig. 11 is experimental data of energy consumption of the 4-class algorithm in the conventional clustering topology, and it can be known by comparing with fig. 10 that the energy consumption is increased compared with the triangulation topology. The CTSOA algorithm and the espsp algorithm do not work much, but their energy consumption is relatively low.
Thirdly, the number of the synchronous packets of the CTSOA algorithm is subjected to simulation verification under the two topological structures, and the simulation result is shown in FIG. 12. With the increase of the number of network nodes, on the basis of clock joint estimation and self-adaptive triangulation topology control, the number of synchronous packets of the CTSOA algorithm is obviously less than that of the synchronous packets of a clustering topology structure, and the result proves that the overall execution of the algorithm is improved for both network overhead and system stability.
Finally, simulation verification is carried out on the synchronous convergence time of the CTSOA algorithm under the two topological structures, and the simulation result is shown in FIG. 13. With the increase of the number of network nodes, on the basis of clock joint estimation and self-adaptive triangulation topology control, the average convergence time of the CTSOA algorithm is about 3.28s, and the average convergence time of the clustering topology structure is about 6.21s, so that the result proves that the algorithm is greatly improved in convergence time and has better convergence.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.

Claims (9)

1. A wireless self-organizing network cooperation time synchronization optimization method is characterized by comprising the following steps:
dividing a network coverage area into topological polygons, dividing each topological polygon into a plurality of monotonous polygons, and introducing nodes into the topological polygons; carrying out triangulation on the monotone polygon to obtain a plurality of triangular areas, wherein the positions of the mass center points of the triangular areas are nodes, and the nodes at the positions of the mass center points of the triangular areas are selected as reference nodes;
step two, the reference node is at the time stamp TM1Performing a round of message packet broadcasting, calculating self normalization quantization values of all nodes and comparing the normalization quantization values with a preset threshold value, taking the nodes with the normalization quantization values higher than the preset threshold value as main nodes, taking the other nodes as slave nodes, sending response packets to reference nodes by the main nodes, and sending the response packets to the reference nodes by the reference nodes at the time stamps TM4Receiving a response packet of the main node;
step three, the reference node performs a round of message packet broadcasting again, wherein the round of message packet comprises TM1、TM4And local timestamp of the reference node, the main node receives the message packet broadcasted by the reference node, and the main node is according to TM1And TM4Calculating a master node clock offset estimate
Figure FDA0003004981360000011
And master node clock drift estimate
Figure FDA0003004981360000012
The master node generates a local timestamp based on the reference node,
Figure FDA0003004981360000013
And
Figure FDA0003004981360000014
the self time is adjusted, and the time of the self time is adjusted,
Figure FDA0003004981360000015
Figure FDA0003004981360000016
wherein the content of the first and second substances,
Figure FDA0003004981360000017
Figure FDA0003004981360000018
k is the current round of time synchronization, N is the total round of time synchronization, ω is the master node clock drift value,
Figure FDA0003004981360000019
is the master node clock offset value, tdIs propagated for the interaction of the reference node and the main nodeFixed time delay of stroke, ΓS2Random noise, Γ, conforming to a Gaussian distribution when receiving a reference node message packet for a master nodeS3Random noise conforming to Gaussian distribution when sending a response packet for a master node;
step four, the main node is at the time stamp TS5Then sending message packet to slave node, the slave node selecting the master node corresponding to the first received message packet as master node of its master-slave synchronous network, the slave node sending response packet to the master node of its master-slave synchronous network, the master node being in time stamp TS8Receiving the response packet sent by the slave node, and sending the local timestamp and T of the master node by the master nodeS5And TS8To TS8Corresponding slave node according to TS5And TS8Calculating slave node clock offset estimates
Figure FDA0003004981360000021
And slave node clock drift estimate ω11The slave node according to the master node local time stamp,
Figure FDA0003004981360000022
And ω11The self time is adjusted, and the time of the self time is adjusted,
Figure FDA0003004981360000023
Figure FDA0003004981360000024
wherein the content of the first and second substances,
Figure FDA0003004981360000025
Figure FDA0003004981360000026
ω1is the slave node clock drift value and,
Figure FDA0003004981360000027
for slave node clock offset value, td1Fixed time delay, Γ, for master and slave node interaction propagation processP6For random noise conforming to a Gaussian distribution when receiving message packets from nodes, ΓP7Random noise which accords with Gaussian distribution when a slave node sends a response packet;
and the normalization quantization value of the node in the step two is calculated by the node according to the link energy, the packet error rate and the time delay weight.
2. The method for optimizing the coordinated time synchronization of the wireless ad hoc network as claimed in claim 1, wherein in the first step, the nodes are firstly sorted according to a random order and then sequentially introduced into the topological polygon according to the sort.
3. The method as claimed in claim 1, wherein the vertex position of the triangular region in the first step is a node.
4. The method as claimed in claim 3, wherein the step one of obtaining a plurality of triangular regions and then selecting a node at a centroid position of the triangular region as a reference node further comprises a step of dyeing the vertices of the triangular region, wherein three vertices of the triangular region have different colors.
5. The method as claimed in claim 1, wherein in step two, all nodes calculate their normalized quantized values according to the first message packet received by themselves.
6. The method for optimizing the coordinated time synchronization of the wireless ad hoc network according to claim 1, wherein said step four further comprises: if the slave node is located at the edge of the communication network of the master node of the master-slave synchronous network, and the slave node can simultaneously receive a plurality of reference node message packets and a plurality of master node message packets, the slave node sends the received plurality of reference node message packets and the plurality of master node message packets to the master node of the master-slave synchronous network of the slave node.
7. The method for optimizing the coordinated time synchronization of the wireless ad hoc network according to claim 1, wherein said step four further comprises: if the slave node is the edge of the communication network of the master node of the master-slave synchronous network, and the slave node can simultaneously receive a plurality of reference node message packets and a plurality of master node message packets, the slave node sends the received plurality of reference node message packets and the plurality of master node message packets to the reference node.
8. The method for optimizing wireless ad hoc network cooperation time synchronization according to claim 1, wherein the method comprises the following steps:
Figure FDA0003004981360000031
Figure FDA0003004981360000032
Figure FDA0003004981360000033
Figure FDA0003004981360000034
Figure FDA0003004981360000035
Figure FDA0003004981360000036
and order:
Figure FDA0003004981360000037
ω ═ ω' and
Figure FDA0003004981360000038
can obtain the product
Figure FDA0003004981360000039
Figure FDA0003004981360000041
Wherein, tdFor fixed time delays in the cross-propagation process between the reference node and the master node, gammaS2Random noise, Γ, conforming to a Gaussian distribution when receiving a reference node synchronization packet for a master nodeS3The method comprises the steps of obtaining random noise which accords with Gaussian distribution when a master node sends a response packet, obtaining a statistical factor of time from receiving a reference node message packet to sending the response packet by the master node, obtaining the total node number in a topological polygon, and obtaining a parameter of Poisson distribution by lambda.
9. The method for optimizing wireless ad hoc network cooperation time synchronization according to claim 1, wherein the method comprises the following steps:
Figure FDA0003004981360000042
Figure FDA0003004981360000043
Figure FDA0003004981360000044
Figure FDA0003004981360000045
Figure FDA0003004981360000046
and order
Figure FDA0003004981360000047
And
Figure FDA0003004981360000048
can obtain the product
Figure FDA0003004981360000049
Figure FDA00030049813600000410
Wherein t isd1Fixed time delay, Γ, for master and slave node interaction propagation processP6For random noise, Γ, conforming to a Gaussian distribution when a slave node receives a master node message packetP7For random noise conforming to the Gaussian distribution when a reply packet is sent from a node, E1Is a statistical factor of the time from the slave node receiving the master node message packet to sending the reply packet.
CN201910963249.5A 2019-10-11 2019-10-11 Wireless self-organizing network cooperation time synchronization optimization method Active CN110636458B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910963249.5A CN110636458B (en) 2019-10-11 2019-10-11 Wireless self-organizing network cooperation time synchronization optimization method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910963249.5A CN110636458B (en) 2019-10-11 2019-10-11 Wireless self-organizing network cooperation time synchronization optimization method

Publications (2)

Publication Number Publication Date
CN110636458A CN110636458A (en) 2019-12-31
CN110636458B true CN110636458B (en) 2021-08-03

Family

ID=68976295

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910963249.5A Active CN110636458B (en) 2019-10-11 2019-10-11 Wireless self-organizing network cooperation time synchronization optimization method

Country Status (1)

Country Link
CN (1) CN110636458B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111682919A (en) * 2020-06-25 2020-09-18 安徽理工大学 Time synchronization method for underground sensing equipment based on time sensitive network
CN111954242B (en) * 2020-08-12 2023-04-18 赛特斯信息科技股份有限公司 OFDM indoor positioning method based on 5G signal
CN113098674B (en) * 2021-03-12 2022-09-27 中科南京移动通信与计算创新研究院 Time synchronization method and system for data link terminal of high-speed carrier self-organizing network
CN113935469B (en) * 2021-10-26 2022-06-24 城云科技（中国）有限公司 Model training method based on decentralized federal learning
CN114615676B (en) * 2022-02-25 2024-04-26 河南大学 Gateway deployment method of self-organizing network

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106162800A (en) * 2015-03-10 2016-11-23 邵贵阳 A kind of implementation method of Wireless Ad Hoc network topological structure
CN110099439A (en) * 2019-04-16 2019-08-06 新华三技术有限公司 Method for synchronizing time and device

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106162800A (en) * 2015-03-10 2016-11-23 邵贵阳 A kind of implementation method of Wireless Ad Hoc network topological structure
CN110099439A (en) * 2019-04-16 2019-08-06 新华三技术有限公司 Method for synchronizing time and device

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
无线传感器网络跨层服务优化时间同步机制;陈桂芬，王义君;《光学精密工程》;20131231;全文 *
无线自组织网络协作时间同步优化算法;刘大鹍 等;《兵工学报》;20200430;全文 *

Also Published As

Publication number Publication date
CN110636458A (en) 2019-12-31

Similar Documents

Publication Publication Date Title
CN110636458B (en) Wireless self-organizing network cooperation time synchronization optimization method
CN106804058B (en) clustering time synchronization method combined with topology maintenance
US6829222B2 (en) Clusterhead selection in wireless ad hoc networks
Shen et al. CLTC: a cluster-based topology control for ad hoc networks
Stojmenovic et al. Dominating sets and neighbor elimination-based broadcasting algorithms in wireless networks
Law et al. A new Bluetooth scatternet formation protocol
WO2022100191A1 (en) Data fusion method and system for distributed sensor network
CN110278048B (en) Hierarchical Ad Hoc network time synchronization method based on clustering algorithm
Dandapat et al. Smart association control in wireless mobile environment using max-flow
Iacoboaiea et al. SON Coordination for parameter conflict resolution: A reinforcement learning framework
Demir et al. Maintaining connectivity in ad hoc networks through wifi direct
Guan et al. To transmit or not to transmit? Distributed queueing games in infrastructureless wireless networks
AbdelSalam et al. Toward efficient task management in wireless sensor networks
Li et al. Joint computation offloading and service caching for MEC in multi-access networks
CN104394534A (en) Two-stage central cooperative frequency spectrum sharing mechanism based on priority queue
CN110602718B (en) Heterogeneous cellular network power distribution method and system based on alternative direction multiplier method
Alnawafa et al. EDMHT-LEACH: Enhancing the performance of the DMHT-LEACH protocol for wireless sensor networks
Manisekaran et al. Energy efficient hierarchical clustering for sensor networks
Qin et al. Mobility weakens the distinction between multicast and unicast
CN104270479B (en) communication implementation method of next generation wireless network
Wang et al. GreenLink: An energy efficient scatternet formation for BLE devices
CN108811024B (en) Cluster splitting method in cluster network structure, module and microcontroller thereof
Yang et al. A decentralized slot synchronization algorithm for TDMA-based ad hoc networks
Hadi et al. Group mobility based clustering scheme for mobile wireless sensor networks
Liu et al. Diffusion clustering routing protocol (DCRP) in wireless sensor networks

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant