CN113260011B

CN113260011B - Mine safety monitoring opportunity network routing method based on moving track

Info

Publication number: CN113260011B
Application number: CN202110529284.3A
Authority: CN
Inventors: 江海峰; 朱孝顺; 肖硕; 杨浩文
Original assignee: China University of Mining and Technology CUMT
Current assignee: China University of Mining and Technology CUMT
Priority date: 2021-05-14
Filing date: 2021-05-14
Publication date: 2022-07-22
Anticipated expiration: 2041-05-14
Also published as: CN113260011A

Abstract

A mine safety monitoring opportunity network routing method based on a moving track is based on the following equipment: the mine wireless network monitoring system comprises a switch connected with a mine wired network, the switch is arranged both on the ground and in the ground, a wireless network mobile terminal arranged on a mine car and a portable intelligent mobile terminal carried by miners are mounted everywhere in the ground, sensors with various functions are mounted underground, the sensors have sensing data detection functions and wireless network communication functions, the mine car and the miners carried by the miners are regarded as mobile nodes, and the switch in the mine is regarded as a fixed node. The routing method comprises the following steps: step 1: node initializing message copy, step 2: and 3, multi-copy distribution transmission based on utility values, wherein the method comprises the following steps: a single copy transmission based on the optimal expected energy consumption. The design not only improves the data transmission speed and reliability, but also reduces the transmission energy consumption and network overhead.

Description

Mine safety monitoring opportunity network routing method based on moving track

Technical Field

The invention relates to a mine safety monitoring opportunistic network routing method based on a moving track, in particular to an opportunistic routing method suitable for a wireless coverage blind area under a mine.

Background

With the rapid development of economy in China, the demand for coal mines is increasing, and the pressure for coal resource exploitation is increasing. The underground geological conditions of coal mines in China are complex, the underground working conditions are severe, safety accidents such as gas and water damage happen occasionally, and the underground coal mines need to be monitored in a full-coverage and uninterrupted manner. Along with the continuous deepening of intelligent mine construction, miners are mostly equipped with mine intelligent mobile terminals, safety monitoring data under a mine can be collected through the intelligent mobile terminals, and the collected data are uploaded to a ground monitoring center in time. Due to the underground heterogeneous space, the coverage range of a base station is limited, and a wireless network coverage blind area exists, so that part of sensing information cannot be uploaded, and the traditional wireless network cannot well complete the tasks of collecting and transmitting underground safety monitoring data.

The opportunistic network originates from an early delay tolerant network DTN, is a self-organizing network, generally consists of intelligent mobile terminals carried by people, and adopts a storage-carrying-forwarding mode to realize data forwarding by utilizing encounters between nodes so as to solve the problem caused by intermittent network connection. The opportunistic network has no end-to-end communication link and does not depend on the characteristics of infrastructure, so that the opportunistic network can be well applied to the underground coal mine, and the difficult problems that the network topology structure of a special application scene in the underground coal mine is dynamically changed, a wireless coverage blind area exists, the end-to-end communication link is in a local communication state or a complete non-communication state and the like are solved.

Introduction of the existing algorithm:

in the Direct Delivery algorithm, a source node does not transmit through other nodes, data is sent out only when a destination node is met, and if the source node does not meet the destination node, the data can not be transmitted out forever. There is no message duplication in the network and the network overhead is minimal but performs the worst in terms of delivery rate and latency. Epidemic algorithm is similar to spreading of infectious diseases, and when two nodes meet, data packets carried by each other but not carried by the other are exchanged. Because information such as node energy, cache and the like is not considered in each forwarding, messages can be quickly spread out, and a large number of message copies exist in a network, the delivery rate is high, the transmission delay is low, but the data forwarding is blindness, resource waste is easily caused, even the network congestion problem is caused, and the network performance is sharply reduced. The Spray and Wait algorithm is different from the Epidemic algorithm in the number of backup data which are not limited, and the algorithm spreads a certain amount of backup data in a network to achieve the purpose of reducing transmission delay. And the Spray stage transmits the message copies to all neighbor nodes in a spraying mode, when the message copies are one, the message is not transmitted to the target node, the Wait stage is started, and the Direct Delivery algorithm strategy is adopted in the Wait stage and is not transmitted until the target node is met. The algorithm solves the problem that Epidemic algorithm has too many message copies, but the delivery rate is not high because other information is not utilized. The spread and Focus algorithm is changed on the basis of the spread and Wait algorithm, and a node carrying a message can select a proper node to forward according to the characteristics of a neighbor node, so that compared with the spread and Wait algorithm, the message delivery rate is improved by the algorithm, but the algorithm does not specify the diffusion mode of a copy in a network. The PROPHET algorithm predicts the encounter probability among the nodes by using the encounter times and the encounter duration information among the nodes, transmits the information to the nodes with high encounter probability with the target node, improves the encounter probability of the nodes when the two nodes meet, and otherwise, declines along with time. The algorithm is unreasonable by purely depending on the attribute of the meeting probability. The Bubble Rap algorithm calculates the social status of the node by using the centrality of the traditional social network analysis. If the message does not enter the destination node community, the message is sent to the node with high global centrality; and if the message enters the destination node community, transmitting the message to the node with high local centrality. Because the messages are forwarded to the nodes with high centrality in a centralized way, the nodes with low centrality are idle, and the imbalance of network load is easily caused. QoN-ASW algorithm improves the SprayAndWait algorithm, in the Spray phase, the quality of the nodes is comprehensively evaluated according to the message forwarding capability of the nodes and the connection strength between the nodes, the algorithm does not consider the residual energy of the node equipment when evaluating the node performance, and for the intelligent mobile terminal with limited energy, the frequent forwarding of messages easily causes the early consumption of the energy of the nodes.

Due to heterogeneous environment, a wireless coverage blind area exists in a mine, a traditional wireless network cannot well complete a data transmission task, and an opportunity network does not need to have a complete communication link from end to end and can be well applied to the underground environment to solve the problem of data transmission. The method includes the steps that a document [ Korean Nana ] is a mine opportunity routing research [ D ] based on multiple ferry nodes, the Chinese mining university is 2016] divides the nodes into common nodes and ferry nodes, the common nodes collect data and do not participate in data forwarding, and the ferry nodes are mobile nodes and are responsible for data transmission. The method improves the Spray and Focus algorithm, and selects the node activity as a forwarding index to select a next hop node. The document [ open-sensitive, opportunity network application research in mine safety monitoring [ D ]. China mining university.2016 ] also improves the Spray and Focus algorithm, respectively improves the Spray and Focus stages, considers the residual energy and the node activity of the node when calculating the node utility value, and avoids blind forwarding of data. However, the article only considers the index of the number of the encountered nodes in the calculation of the node liveness and is too single. Aiming at a special application scene of underground coal mine safety monitoring, a routing algorithm MTADR based on mobile track assistant decision is provided on the basis of Spray and Focus, and the MTADR divides a routing decision into a multi-copy distribution stage based on a utility value and a single-copy transmission stage based on optimal expected energy consumption. The utility value-based multi-copy distribution strategy distributes message copies according to the utility values of the nodes and assists in calculating the activity of the nodes according to the historical movement tracks of the nodes, and the activity of the nodes and the residual energy of the nodes are mainly considered in the calculation of the utility values. In the single copy transmission strategy based on the optimal expected energy consumption, the node carrying the message copy predicts the possibility of completing the data transmission task and the expected transmission energy consumption according to the historical movement track of the neighbor node, and selects the next hop node according to the possibility.

Disclosure of Invention

The invention aims to solve the problem of unstable receiving and sending of wireless network coverage blind area monitoring data in the prior art, and provides a mine safety monitoring opportunistic network routing method based on a moving track, which improves the success rate of data receiving and sending.

In order to achieve the above purpose, the technical solution of the invention is as follows:

the mine safety monitoring opportunistic network routing method based on the moving track is based on the following equipment: the mine car wireless network switching system comprises a switch connected with a mine wired network, the switch is arranged both on the mine and in the mine, a wireless network mobile terminal arranged on a mine car and a portable intelligent mobile terminal carried by miners are arranged everywhere in the mine, sensors with various functions are arranged underground, the sensors have a sensing data detection function and a wireless network communication function, in the method, the mine car and the mobile terminal carried by the miners are regarded as mobile nodes, and the switch in the mine is regarded as a fixed node;

step 1: initializing a message copy by a node; initialization of a source node: the data of the sensor collected by the mobile node generates a message to be forwarded, the mobile node becomes a source node of the message data, the source node copies the message to be forwarded into L message copies after generating the message to be forwarded, and the initialization energy of the node is obtained as E_initWhen the mobile node moves along the past track information, the initialization is completed; initialization of a mobile node: when the mobile node obtains the message data, the mobile node carries out initialization to obtain the nodePoint initialization energy of E_initWhen the mobile node moves along the past track information, the initialization is completed; after initialization is completed, different information transmission modes are adopted according to the number of the message copies carried by the nodes, and when the number of the message copies is larger than 1, step 2 is carried out; when the number of the message copies is equal to 1, entering a step 3;

and 2, step: multi-copy allocation transmission based on utility values: when the number of the message copies carried by the node is more than 1, a multi-copy allocation strategy based on the utility value is implemented: the node carrying the message meets the neighbor node, searches whether the neighbor node has a destination node or not, and forwards the message to the destination node if the neighbor node has the destination node; if no destination node exists, calculating a utility value according to the historical movement track information: if the message is urgent data, distributing the message copy number according to the utility value; if the message is non-urgent data, setting a utility value threshold value, selecting nodes higher than the utility value threshold value, and distributing the message copy number according to the utility value by the nodes;

and 3, step 3: single copy transmission based on optimal expected energy consumption: when the number of message copies carried by the node is 1, a single copy transmission strategy based on expected energy consumption is implemented, the node and the neighbor node meet each other, a moving path is obtained according to historical moving track information, and the moving path which can complete a data transmission task in the message survival period is selected to be added into a candidate set R; and selecting a moving path with the optimal expected energy consumption from the candidate set R, if the second node of the path is a neighbor node, forwarding the data copy to the neighbor node, and otherwise, not forwarding the data copy with the message.

The step 2: multi-copy allocation transmission based on utility values: after a source node generates a message to be forwarded, firstly, the source node copies a message copy with the quantity of L in a cache of the source node, a node carrying the message copy meets other nodes in the moving process, and if the meeting node does not have the message copy, a message copy distribution strategy selects whether to distribute the message copy and the quantity of the distributed message copy according to the transmission capacity of the meeting node; the transmission capacity of the node is represented by a utility value, and the utility value is calculated based on the activity and the residual energy of the node; the node with high utility value obtains more message copies, and transmits data to the destination node by using the stronger transmission capability of the node;

number of initial message copies of node i: the node initial message copy number L is calculated according to the following formula:

in the above formula, a is a delay constraint factor, the delay constraint factor is a multiple of the average delay of the network, M is the number of mobile nodes in the network, and H_MFor the harmonic progression, the corresponding calculation formula is as follows:

in the above formula, r is an order, and finally, the initial message copy number L can be obtained by solving the expression (1) and rounding up the solved solution;

II, similarity of moving tracks of nodes: the moving tracks of all the mobile nodes in the last day are obtained through an underground positioning system, and because the moving track of a miner deviates from the historical moving track, similarity calculation needs to be carried out on the current moving track and the historical moving track of the miner so as to measure the usability of the historical moving track information of the nodes; calculating the track similarity by adopting the following method:

acquiring the position information of the node once every t time, wherein the historical track of the node consists of track points according to the time sequence; let the history movement track be { p₁,p₂,...,p_n1The current moving track of the node is { q }₁,q₂,...,q_m1The calculation formula of the matching number is as follows:

in the above formula, L (p, q) is the number of matching points in the two tracks; x is min (m)₁,n₁) In whichm₁The number of moving track points of the current node, n₁The number of the historical moving track points is set;

in the above formula, | p_iq_iI represents the distance between two track points, when the distance between the two track points is less than or equal to alpha, the value of the match function is 1, otherwise, the value is 0; wherein, alpha is the error of the current node moving track point and the historical moving track point, and can be properly adjusted according to the positioning interval time and the moving speed of the node;

the trajectory similarity of node i is calculated as follows:

in the above formula, s (i) is the track similarity of the node i, and L (p, q) is the number of matching points in the two tracks; x is min (m)₁,n₁) Wherein m is₁The number of moving track points of the current node, n₁The number of the historical moving track points is set;

III, node activity degree: calculating the node activity degree by combining the underground application scene of the coal mine and considering the work type attribute of the miner node; when the node i and the node j meet, the activity of the node is calculated according to the following formula (7):

CEN(i,j)＝N_a(i)∪N_a(j) (6)

in the above formula, CEN (i, j) is the information according to the historical movement track, CEN (i, j) is the union of node i and the node j encountered node set, N_a(i) And N_a(j) Respectively node sets encountered by a node i and a node j according to the historical movement track;

in the above formula, N_ac(i) Indicating the liveness of node i, N_l(i) Is expressed according toIn the historical movement track information acquired by the bit system, a node set N which is encountered by the node i next_o(i) Represents a set of nodes previously encountered by node i; wherein b is N_l(i) The number of the types of the jobs in the set, a is the sum of the number of different jobs encountered by the nodes i and j according to the historical track, and s (i) is the track similarity of the nodes i, see formula (5);

and IV, calculating utility value: calculating a node utility value U (i) based on the node activity and the node residual energy:

in the above formula, E_cur(i) Is the current residual energy of node i, E_initIs the initial energy of node i, N_ac(i) The activity of the node i is represented by w, the weight of the attribute is the weight of the attribute, and the weight of the attribute is a set value;

v, copy distribution of different data types: considering that resources of an intelligent mobile terminal are limited under a mine, urgent real-time data and non-urgent non-real-time data exist, and different copy distribution strategies are adopted for different types of data; and selecting whether to set a utility value threshold according to the urgency of the message: for the emergency data, a utility value threshold is not set, and more nodes are distributed to message copies to participate in the forwarding of the message; for non-emergency data, setting utility value threshold, when utility value U (i) of adjacent node is higher than utility value threshold U_th(i) Message copies can be distributed according to the size of the utility value, so that nodes with strong performance participate in copy distribution;

the utility value threshold is calculated as follows:

in the above formula, U_th(i) Distributing a copy to a utility value threshold of a neighbor node for a node i, wherein N (i) is a neighbor node set of the node i, U (j) is a utility value of the neighbor node, and k is the number of the neighbor nodes of the node i;

vi message duplicate assignment: when a node i carrying a message copy meets a neighbor node, distributing the message copy for each neighbor node without the message copy according to the size of a utility value, if the neighbor node is non-urgent data, selecting the neighbor node higher than a utility value threshold, and then distributing the copy according to the utility value; assuming that k neighbor nodes are provided, the neighbor node j calculates the number of distributed message copies according to the following formula:

in the above formula, m (j) is the copy number of the message m in the cache of the node j, and m (i) is the message copy number of the node i;

the node i distributes the message copies to the node j according to the calculated number, and after the node i distributes the message copies, the message copies of the node i are updated to m_up(i)：

In the above formula, m_up(i) Caching the copy number of the message m in the updated node i;

in the process of copy distribution and transmission, a node carrying a message copy meets a neighbor node, if the neighbor node owns the message copy, the node does not participate in the distribution of the step 2 and the transmission of the step 3, and if the neighbor node does not own the message copy, the node participates in the distribution of the step 2 and the transmission of the step 3.

The step 2: current residual energy E of node i in multi-copy distribution transmission based on utility value_cur(i) The calculation method of (2) is as follows: node scanning energy consumption refers to energy consumed by node scanning channel, and then scanning energy consumption E of node i_s(i) Can be expressed as:

in the above formula, e_sThe energy consumed by single scanning of the node i is represented by T, the scanning period of the node is represented by T, and the working time length of the node is represented by T;

the data transmission energy consumption of the node is proportional to the transmitted data quantity, and the energy consumed by the node i for transmitting unit data is e_tThe amount of transmitted data is s_tThen, the transmission energy consumption E of the node i_t(i) Can be expressed as:

E_t(i)＝e_t×s_t (13)

similarly, the energy consumption for receiving data is proportional to the amount of data received by the node, and the energy consumed by the node i for receiving unit data is e_rThe amount of data received by node i is s_rThen reception energy E of node i_r(i) Can be expressed as:

E_r(i)＝e_r×s_r (14)

in summary, the total energy consumption E of node i_c(i) Can be expressed as:

E_c(i)＝E_t(i)+E_r(i)+E_s(i) (15)

the node i residual energy is:

E_cur(i)＝E_init-E_c(i) (16)

wherein E is_initEnergy is initialized for the node.

The step 3: single copy transmission based on optimal expected energy consumption: due to the fact that energy of an intelligent mobile terminal carried under a mine is limited, when the number of copies of a message m carried by a node is 1, a single copy transmission strategy based on optimal expected energy consumption is implemented; the goal of this strategy is to transmit the message to the destination node within the message lifetime while selecting the node path that forwards the least energy consumption expected:

i defines the encountered set of nodes: defining an encounter set of single nodes, constructing the encounter set of the nodes according to the moving track of the node history, and when the encounter set M (A) of the node A is { (B, t)_a1),(C,t_a2),...,(D,t_am) Denotes node A and node B at t_a1The time is met, the node A and the node C are at t_a2Meet at all times, node AAnd node D is at t_amMeet at any moment;

II defines the moving path between nodes: according to the meeting time set of the node carrying the message duplicate and other nodes, removing outdated records in the meeting set, and constructing a moving path between nodes by taking the node carrying the message duplicate as an initial node, wherein the nodes in the path meet in sequence according to the time sequence; when the moving path is A-B-C-D, the meeting time of the nodes A and B is earlier than that of the nodes B and C, and the meeting time of the nodes B and C is earlier than that of the nodes C and D;

III, screening a moving path candidate set: screening a mobile path candidate set R capable of completing a message transmission task, wherein the mobile path candidate set R comprises two types of mobile paths: the first type is that a destination node exists on a moving path; the second type is that the nodes on the moving path can upload messages through the wireless network; when the node i carries the message copy m to meet the neighbor nodes, finding out all the moving paths of the node i as the initial node, and setting the moment when the message m is generated as t₀If the message survival time is TLL, the effective time of the message m is [ t ]₀,t₀+TLL]；

The screening process of the moving path candidate set R is as follows: screening of the first type of movement path: checking all the moving paths which are found out in the past and take the node i as the initial node, searching whether a destination node of the message m exists on the moving path, if the destination node of the message m exists on the moving path and the time from the initial node to the destination node is less than t₀+ TLL, adding the moving path from the starting node to the destination node into a moving path candidate set R; and (3) screening of a second type of moving path: checking all the mobile paths which are found out in the prior art and take the node i as an initial node, searching whether the node enters a wireless network coverage area during the message survival period exists on the mobile paths, and if the node exists, adding the mobile path from the initial position to the node into a mobile path candidate set R;

and IV, calculating the expected energy consumption of the moving path: calculating expected energy consumption of each moving path in the set for the screened moving path candidate set R capable of completing the data transmission task, and selecting a moving path capable of completing message transmission and having the least expected energy consumption; the expected energy consumption for the first type of travel path is calculated as follows:

E_exp＝E_t*d (17)

in the above formula, E_expFor expected energy consumption of the movement path, E_tD is the hop number passing from the starting node to the destination node;

the expected energy consumption for the second type of movement path is calculated as follows:

the data transmission power of the adopted 3G, LTE and WIFI network data transmission power model is calculated according to the following formula:

P＝α_ut_u+α_dt_d+β (18)

in the above formula, P is the power of the transmitted data, t_uFor the uplink rate, t_dFor the downlink rate, α_uFor uploading a power parameter, α_dFor receiving power parameters, beta is the basic power under different networks;

the energy consumption for transmitting data under the 3G, LTE and WIFI networks is calculated by adopting the following formula:

E_ud(t)＝P*t (19)

in the above formula, E_ud(t) the energy consumption for transmitting data in the 3G, LTE and WIFI networks, P the calculated power for transmitting data, and t the duration of transmitting data; and calculating the expected energy consumption of the second type of moving paths in the moving path candidate set R by combining the transmission energy consumption of the different networks:

E_exp＝E_t*d+E_ud(t) (20)

in the above formula, E_ud(t) energy consumption for data transmission in 3G, LTE, WIFI networks, E_expFor expected energy consumption of the movement path, E_tRepresenting the energy consumption for transmitting the message to the next hop node on the moving path, and d is the hop number from the starting node to the destination node;

and then sequencing the expected energy consumption of the moving paths in the set R, traversing the set R from the beginning until finding out the optimal moving path with the least expected energy consumption, and then transmitting the single message copy according to the optimal path.

The step 1: initializing a message copy by a node; the sensor classifies the sensed data, judges whether the data is urgent or non-urgent according to the numerical value of the sensed data, and sends the data at the same time as the data sent by the sensor.

Compared with the prior art, the invention has the beneficial effects that:

1. the routing method of the mine safety monitoring opportunistic network based on the moving track introduces the opportunistic network, transmits data to a target node through the encounter of underground moving nodes through a data transmission mode of storage-carrying-forwarding, and effectively improves the reliability of data transmission. Therefore, the data transmission reliability of the design is high.

2. The invention discloses a mine safety monitoring opportunity network routing method based on a moving track, which comprises two stages, namely a multi-copy distribution stage based on a utility value and a single-copy transmission stage based on optimal expected energy consumption: in a multi-copy distribution stage based on utility values, two attribute values of activity and residual energy of nodes are calculated according to historical movement tracks of the nodes, the similarity of the tracks is calculated to measure the reliability of the historical track information, then the utility values are comprehensively calculated, the blind forwarding of message copies is avoided, and the messages complete data transmission tasks through the nodes with high utility values: meanwhile, different allocation strategies are adopted according to urgency and non-urgency of the message, and for urgent data, a utility value threshold is not set, and the data are transmitted to more nodes; for non-urgent data, a utility value threshold is set, and more appropriate nodes are assigned with more copies to complete data transmission: and in the single copy transmission stage based on the optimal expected energy consumption, according to the historical movement track information of the nodes, considering whether the nodes can transmit data to the destination node or whether the nodes enter the coverage area of the wireless network during the survival period of the message, and selecting the node with the optimal energy consumption from the nodes. The success rate of data transmission is effectively improved, the data transmission time is shortened, and the network overhead and the energy consumption of data transmission are reduced. Therefore, the design can improve the data transmission speed and reliability and reduce the data transmission energy consumption and network overhead at the same time.

Drawings

FIG. 1 is a schematic diagram of an underground coal mine opportunity network.

Fig. 2 is a flow chart of the screening of the second type of moving paths based on the optimal expected energy consumption in step 3.

Fig. 3 is a diagram showing an example of the non-urgent data algorithm in embodiment 3.

Fig. 4 is a graph comparing the success rates of the emergency data transmission according to the four algorithms of embodiment 3.

Fig. 5 is a comparison chart of the success rate of non-urgent data transmission of the four algorithms in embodiment 3.

Fig. 6 is a graph comparing the average transmission delay of urgent data of the four algorithms in embodiment 3.

Fig. 7 is a graph comparing the average transmission delay of non-urgent data of the four algorithms in embodiment 3.

Fig. 8 is a network overhead comparison graph of urgent data of the four algorithms in embodiment 3.

Fig. 9 is a network overhead comparison graph of non-urgent data for the four algorithms in example 3.

Fig. 10 is a graph of the average remaining energy of the emergency data for the four algorithms in example 3.

Fig. 11 is a graph of the average remaining energy for non-urgent data for the four algorithms in example 3.

In the figure: MTADR denotes the MTADR algorithm of the present invention; epidemic denotes the Epidemic algorithm; SprayAndFocus represents the SprayAndFocus algorithm; QoN-ASW denotes the QoN-ASW algorithm.

Detailed Description

The invention is described in further detail below with reference to the figures and the detailed description of the invention.

Referring to fig. 1 to 2, a mine safety monitoring opportunistic network routing method based on a moving track is based on the following equipment: the mine car wireless network switching system comprises a switch connected with a mine wired network, the switch is arranged both on the mine and in the mine, a wireless network mobile terminal arranged on a mine car and a portable intelligent mobile terminal carried by miners are arranged everywhere in the mine, sensors with various functions are arranged underground, the sensors have a sensing data detection function and a wireless network communication function, in the method, the mine car and the mobile terminal carried by the miners are regarded as mobile nodes, and the switch in the mine is regarded as a fixed node;

step 1: initializing a message copy by a node; initialization of a source node: the data of the sensor collected by the mobile node generates a message to be forwarded, the mobile node becomes a source node of the message data, the source node copies the message to be forwarded into L message copies after generating the message to be forwarded, and the initialization energy of the node is obtained as E_initWhen the mobile node moves along the past track information, the initialization is completed; initialization of a mobile node: when the mobile node obtains the message data, initializing to obtain the node initialization energy E_initWhen the mobile node moves along the past track information, the initialization is completed; after initialization is completed, different information transmission modes are adopted according to the number of the message copies carried by the nodes, and when the number of the message copies is larger than 1, a step 2 is carried out; when the number of the message copies is equal to 1, entering a step 3;

and 2, step: multi-copy allocation transmission based on utility values: when the number of the message copies carried by the node is more than 1, a multi-copy allocation strategy based on the utility value is implemented: the node carrying the message meets the neighbor node, searches whether the neighbor node has a destination node or not, and forwards the message to the destination node if the neighbor node has the destination node; if no destination node exists, calculating a utility value according to the historical movement track information: if the message is urgent data, distributing the number of message copies according to the utility value; if the message is non-urgent data, setting a utility value threshold value, selecting nodes higher than the utility value threshold value, and distributing the message copy number according to the utility value by the nodes;

and step 3: single copy transmission based on optimal expected energy consumption: when the number of the message copies carried by the node is 1, a single copy transmission strategy based on expected energy consumption is implemented, the node meets the neighbor node, a moving path is obtained according to historical moving track information, and the moving path which can complete a data transmission task during the message survival period is selected to be added into a candidate set R; and selecting a mobile path with optimal expected energy consumption from the candidate set R, if a second node of the path is a neighbor node, forwarding the data copy to the neighbor node, and if not, forwarding the data copy with the message.

The step 2: multi-copy allocation transmission based on utility values: after a source node generates a message to be forwarded, firstly, the source node copies a message copy with the quantity of L in a cache of the source node, a node carrying the message copy meets other nodes in the moving process, and if the meeting node does not have the message copy, a message copy distribution strategy selects whether to distribute the message copy and the quantity of the distributed message copy according to the transmission capacity of the meeting node; the transmission capacity of the node is represented by a utility value, and the utility value is calculated based on the activity and the residual energy of the node; the node with high utility value can obtain more message copies, and the data can be transmitted to the destination node by utilizing the strong transmission capability of the node;

in the above formula, a is a delay constraint factor, the delay constraint factor is a multiple of the average delay of the network, M is the number of mobile nodes in the network, H_MFor the harmonic progression, the corresponding calculation formula is as follows:

in the above formula, r is the order, and finally the initial message copy number L can be obtained by solving the expression (1) and rounding the solved solution;

II, similarity of moving tracks of nodes: the moving tracks of all the mobile nodes in the last day are obtained through an underground positioning system, and because the moving tracks of miners and the historical moving tracks have deviation, similarity calculation needs to be carried out on the current moving tracks and the historical moving tracks of the miners to measure the usability of the historical moving track information of the nodes; to the trackThe similarity is calculated by the following method: acquiring position information of the nodes once every t time, wherein the historical tracks of the nodes are composed of track points according to a time sequence; let the history movement track be p₁,p₂,...,p_n1The current moving track of the node is { q }₁,q₂,...,q_m1The calculation formula of the matching number is as follows:

in the above formula, L (p, q) is the number of matching points in the two tracks; x is min (m)₁,n₁) Wherein m is₁The number of moving track points of the current node, n₁The number of the historical moving track points is;

in the above formula, | p_iq_iThe | represents the distance between the two track points, when the distance between the two track points is less than or equal to alpha, the value of the match function is 1, otherwise, the value is 0; wherein, alpha is the error of the current node moving track point and the historical moving track point, and can be properly adjusted according to the positioning interval time and the moving speed of the node; the trajectory similarity of node i is calculated as follows:

in the above formula, s (i) is the track similarity of the node i, and L (p, q) is the number of matching points in the two tracks; x is min (m)₁,n₁) Wherein m is₁The number of moving track points of the current node, n₁The number of the historical moving track points is;

node III activity: calculating the node activity degree by combining the underground application scene of the coal mine and considering the work type attribute of the miner node; when the node i meets the node j, the activity of the node is calculated according to the following formula (7):

CEN(i,j)＝N_a(i)∪N_a(j) (6)

in the above formula, CEN (i, j) is the information according to the historical movement track, CEN (i, j) is the union of node i and the node j encountered node set, N_a(i) And N_a(j) Respectively a node set encountered by the node i and the node j according to the historical movement track;

in the above formula, N_ac(i) Indicating the liveness of node i, N_l(i) Represents a node set N which is to be encountered by the node i in the historical moving track information acquired by the positioning system_o(i) Represents a set of nodes previously encountered by node i; wherein b is N_l(i) The number of the varieties in the set, a is the sum of the numbers of different varieties encountered by the nodes i and the nodes j according to the historical track, and s (i) is the track similarity of the nodes i, see formula (5);

v, copy distribution of different data types: considering that resources of an intelligent mobile terminal are limited in a mine, urgent real-time data and non-urgent non-real-time data exist, and different copy distribution strategies are adopted for different types of data; and selecting whether to set a utility value threshold according to the urgency of the message: for the emergency data, a utility value threshold is not set, and more nodes are distributed to message copies to participate in message forwarding; for non-emergency data, setting a utility value threshold, and when the utility value U (i) of the adjacent node is higher than the utility valueThreshold U_th(i) Message copies can be distributed according to the size of the utility value, so that nodes with strong performance participate in copy distribution; the utility value threshold is calculated as follows:

and VI message copy allocation: when a node i carrying a message copy meets a neighbor node, distributing the message copy according to the size of a utility value for each neighbor node without the message copy, if the neighbor node is non-urgent data, selecting the neighbor node higher than the threshold of the utility value, and then distributing the copy according to the utility value; assuming that k neighbor nodes are provided, the neighbor node j calculates the number of distributed message copies according to the following formula:

the node i distributes the message copies to the node j according to the calculated quantity, and after the node i distributes the message copies, the message copies of the node i are updated to m_up(i)：

in the process of copy distribution and transmission, the node carrying the message copy meets the neighbor node, if the neighbor node has the message copy, the node does not participate in the distribution of the step 2 and the transmission of the step 3, and if the neighbor node does not have the message copy, the node participates in the distribution of the step 2 and the transmission of the step 3.

in the above formula, e_sThe energy consumed by single scanning of the node i is represented by T, the scanning period of the node is represented by T, and the working time length of the node is represented by T; the data transmission energy consumption of the node is proportional to the transmitted data quantity, and the energy consumed by the node i for transmitting unit data is e_tThe amount of transmitted data is s_tThen the sending energy consumption E of the node i_t(i) Can be expressed as:

E_t(i)＝e_t×s_t (13)

similarly, the receiving energy consumption of the data is proportional to the data quantity received by the node, and the energy consumed by the node i for receiving the unit data is e_rThe amount of data received by node i is s_rThen reception energy E of node i_r(i) Can be expressed as:

E_r(i)＝e_r×s_r (14)

in summary, the total energy consumption E of node i_c(i) Can be expressed as:

E_c(i)＝E_t(i)+E_r(i)+E_s(i) (15)

the node i residual energy is:

E_cur(i)＝E_init-E_c(i) (16)

wherein E is_initEnergy is initialized for the node.

i defines the encountered set of nodes: defining an encounter set of single nodes, constructing the encounter set of the nodes according to the historical movement track of the nodes, and when the encounter set M (A) of the node A is { (B, t)_a1),(C,t_a2),...,(D,t_am) Denotes node A and node B are at t_a1Meet at a moment, node A and node C meet at t_a2The time is met, the node A and the node D are at t_amMeet at any moment;

II, defining a moving path between nodes: according to the meeting time set of the node carrying the message duplicate and other nodes, removing outdated records in the meeting set, and constructing a moving path between nodes by taking the node carrying the message duplicate as an initial node, wherein the nodes in the path meet in sequence according to the time sequence; when the moving path is A-B-C-D, the meeting time of the nodes A and B is earlier than that of the nodes B and C, and the meeting time of the nodes B and C is earlier than that of the nodes C and D;

III, screening a moving path candidate set: screening a mobile path candidate set R capable of completing a message transmission task, wherein the mobile path candidate set R comprises two types of mobile paths: the first type is that a destination node exists on a moving path; the second type is that the nodes on the moving path can upload messages through the wireless network; when the node i carries the message copy m and meets the neighbor nodes, all moving paths of which the node i is an initial node are found, and the moment when the message m is generated is set as t₀If the message survival time is TLL, the effective time of the message m is [ t₀,t₀+TLL]；

The screening process of the moving path candidate set R is as follows: screening of the first type of movement path: checking all the moving paths which are found out in the past and take the node i as an initial node, searching whether a destination node of the message m exists on the moving path, if the destination node of the message m exists on the moving path and the time from the initial node to the destination node is less than t₀+ TLL, adding the moving path from the starting node to the destination node into the moving path candidateSelecting a set R; and (3) screening of a second type of moving path: checking all the mobile paths which are found out in the front and take the node i as an initial node, searching whether the node enters a wireless network coverage area during the message survival period exists on the mobile paths, and if the node exists, adding the mobile path from the initial position to the node into a mobile path candidate set R;

E_exp＝E_t*d (17)

in the above formula, E_expFor expected energy consumption of the movement path, E_tD is the hop number passing from the starting node to the destination node; the expected energy consumption for the second type of movement path is calculated as follows: the adopted 3G, LTE and WIFI network data transmission power model is that the transmission data power is calculated according to the following formula:

P＝α_ut_u+α_dt_d+β (18)

E_ud(t)＝P*t (19)

E_exp＝E_t*d+E_ud(t) (20)

in the above formula, E_ud(t) energy consumption for data transmission in 3G, LTE, WIFI networks, E_expFor expected energy consumption of the movement path, E_tD is the hop number passing from the starting node to the destination node; and then, sequencing the expected energy consumption of the moving paths in the set R, traversing the set R from the beginning until finding out the optimal moving path with the least expected energy consumption, and then transmitting a single message copy according to the optimal path.

The principle of the invention is illustrated as follows:

performance evaluation indexes of the algorithm are as follows: in subsequent simulation experiments, performance evaluation is mainly performed on four aspects of message transmission success rate, average transmission delay, network overhead and average residual energy.

Message transmission success rate: representing the ratio of the number of packets successfully transmitted from the source node to the total number of packets generated by the source node

Calculation of where d_srIndicates the message transmission success rate, p_sNumber of data packets, p, indicating successful transmission_tRepresenting the total number of messages generated by the source node.

Average transmission delay: refers to the average transmission time from the generation of a data packet from the source node to the receipt of the data by the destination node, in terms of

Calculation of where d_avgRepresenting the average transmission delay, p_sNumber of data packets indicating successful transmission, d_tlRepresenting the total transmission delay.

Controlling the overhead: control overhead represents all control in the networkThe ratio of the number of message packets to the total number of data packets, in terms of

Calculating where n_oDenotes the control overhead, p_cIndicating the number of control message packets.

Average residual energy: average remaining capacity of all mobile intelligent terminals:

wherein E is_avgDenotes the average residual energy, E_iRepresenting the remaining energy of node i and n representing the total number of nodes.

In the formula (1), a is a delay constraint factor, the delay constraint factor is a multiple of the average network delay (the average network delay also has a corresponding calculation formula, and if the document is too long, the average network delay is not added), and the copy number L is changed according to the change of a and the node number M.

The MTADR algorithm of the invention: a Mobile track Aided Decision-making based Routing algorithm for mine safety monitoring opportunity network in common mine safety monitoring is disclosed.

According to the mobile characteristics of the nodes, the MTADR algorithm selects a map-based mobile model MBM, which is shown in the following steps: liu string, user movement model research [ D ] in mobile opportunity network, SiAn electronics technology university, 2018. Zhaoyao, Gao Xiu Feng, Chen Li Yun, Li Shi Wei, Mobile Ad hoc network group movement model overview [ J ] flying missile, 2019(05): 68-72.

The step 3: based on single copy transmission of optimal expected energy consumption, the expected energy consumption calculation method of the second type of moving path in the expected energy consumption calculation of the IV moving path adopts a 3G, LTE and WIFI network data transmission power model in documents [ Huang J, Qian F, Gerber A, et al. A close evaluation of performance and power characteristics of 4G LTE network-works.in: Proceedings of the ACM mobilsys, Lake Dis-trict, 25-29 June 2012 ].

Example 1:

step 1: initializing a message copy by a node; initialization of a source node: the data of the sensor collected by the mobile node generates a message to be forwarded, the mobile node becomes a source node of the message data, the source node copies the message to be forwarded into L message copies after generating the message to be forwarded, and the initialization energy of the node is obtained as E_initWhen the mobile node moves along the past track information, the initialization is completed; initialization of a mobile node: when the mobile node obtains the message data, initializing to obtain the node initialization energy of E_initWhen the mobile node is in the underground state, the mobile node is initialized; after initialization is completed, different information transmission modes are adopted according to the number of the message copies carried by the nodes, and when the number of the message copies is larger than 1, a step 2 is carried out; when the number of the message copies is equal to 1, entering a step 3; initializing a message copy by a node; the sensor classifies the sensed data, and judges whether the data is urgent or non-urgent according to the numerical value of the sensed data, and the sensor sends the data at the same time as the data is urgent;

step 2: multi-copy allocation transmission based on utility values: when the number of the message copies carried by the node is more than 1, a multi-copy allocation strategy based on the utility value is implemented: the node carrying the message meets the neighbor node, searches whether the neighbor node has a destination node or not, and forwards the message to the destination node if the neighbor node has the destination node; if no destination node exists, calculating a utility value according to the historical movement track information: if the message is urgent data, distributing the number of message copies according to the utility value; if the message is non-urgent data, setting a utility value threshold value, selecting nodes higher than the utility value threshold value, and distributing the message copy number according to the utility value by the nodes;

and step 3: single copy transmission based on optimal expected energy consumption: when the number of message copies carried by the node is 1, a single copy transmission strategy based on expected energy consumption is implemented, the node and the neighbor node meet each other, a moving path is obtained according to historical moving track information, and the moving path which can complete a data transmission task in the message survival period is selected to be added into a candidate set R; and selecting a mobile path with optimal expected energy consumption from the candidate set R, if a second node of the path is a neighbor node, forwarding the data copy to the neighbor node, and if not, forwarding the data copy with the message.

Example 2:

example 2 is substantially the same as example 1 except that:

in the above formula, a is a delay constraint factor, the delay constraint factor is a multiple of the average delay of the network, M is the number of mobile nodes in the network, H_MFor the harmonic series, the corresponding calculation formula is as follows:

II, similarity of moving tracks of nodes: the moving tracks of all the mobile nodes in the last day are obtained through an underground positioning system, and because the moving track of a miner deviates from the historical moving track, similarity calculation needs to be carried out on the current moving track and the historical moving track of the miner so as to measure the usability of the historical moving track information of the nodes; calculating the track similarity by adopting the following method: acquiring the position information of the node once every t time, wherein the historical track of the node consists of track points according to the time sequence; let the history movement track be p₁,p₂,...,p_n1The current moving track of the node is q₁,q₂,...,q_m1The calculation formula of the matching number is as follows:

in the above formula, s (i) is the track similarity of the node i, and L (p, q) is the number of matching points in the two tracks;

x＝min(m₁,n₁) Wherein m is₁The number of moving track points of the current node, n₁The number of the historical moving track points is set;

III, node activity degree: calculating the node activity degree by combining the underground application scene of the coal mine and considering the work type attribute of the miner node; when the node i meets the node j, the activity of the node is calculated according to the following formula (7):

CEN(i,j)＝N_a(i)∪N_a(j) (6)

in the above formula, CEN (i, j) is the information according to the historical movement track, CEN (i, j) is the union of the node sets encountered by node i and node j, and N is_a(i) And N_a(j) Respectively a node set encountered by the node i and the node j according to the historical movement track;

in the above formula, N_ac(i) Indicating the liveness of node i, N_l(i) Represents a node set N which is to be encountered by the node i in the historical movement track information obtained by the positioning system_o(i) Represents a set of nodes previously encountered by node i; wherein b is N_l(i) The number of the varieties in the set, a is the sum of the numbers of different varieties encountered by the nodes i and the nodes j according to the historical track, and s (i) is the track similarity of the nodes i, see formula (5);

in the above formula, E_cur(i) Is the current remaining energy of node i, E_initIs the initial energy of node i, N_ac(i) The activity of the node i is shown, w is the weight of the attribute, and the weight of the attribute is a set value;

v, copy distribution of different data types: considering that resources of an intelligent mobile terminal are limited in a mine, urgent real-time data and non-urgent non-real-time data exist, and different copy distribution strategies are adopted for different types of data; and selecting whether to set a utility value threshold according to the urgency of the message: for the emergency data, a utility value threshold is not set, and more nodes are distributed to message copies to participate in the forwarding of the message; for non-emergency data, setting utility value threshold, when utility value U (i) of adjacent node is higher than utility value threshold U_th(i) Message copies can be distributed according to the size of the utility value, so that nodes with strong performance participate in copy distribution; the utility value threshold is calculated as follows:

vi message duplicate assignment: when a node i carrying a message copy meets a neighbor node, distributing the message copy according to the size of a utility value for each neighbor node without the message copy, if the neighbor node is non-urgent data, selecting the neighbor node higher than the threshold of the utility value, and then distributing the copy according to the utility value; assuming that there are k neighbor nodes, the neighbor node j calculates the number of distributed message copies according to the following formula:

in the above formula, m (j) is the number of copies of the message m in the cache of the node j, and m (i) is the message copy of the node iThe number of the components; the node i distributes the message copies to the node j according to the calculated number, and after the node i distributes the message copies, the message copies of the node i are updated to m_up(i)：

In the above formula, m_up(i) Caching the copy number of the message m in the updated node i; in the process of copy distribution and transmission, the node carrying the message copy meets the neighbor node, if the neighbor node has the message copy, the node does not participate in the distribution of the step 2 and the transmission of the step 3, and if the neighbor node does not have the message copy, the node participates in the distribution of the step 2 and the transmission of the step 3.

The step 2: current residual energy E of node i in multi-copy distribution transmission based on utility value_cur(i) The calculation method of (2) is as follows: node scanning energy consumption refers to energy consumed by the node to scan channels, and then scanning energy consumption E of the node i_s(i) Can be expressed as:

E_t(i)＝e_t×s_t (13)

similarly, the energy consumption for receiving data is proportional to the amount of data received by the node, and the energy consumed by the node i for receiving unit data is e_rThe amount of data received by node i is s_rThen, the receiving energy consumption E of the node i_r(i) Can be expressed as:

E_r(i)＝e_r×s_r (14)

in summary, the total energy consumption E of node i_c(i) Can be expressed as:

E_c(i)＝E_t(i)+E_r(i)+E_s(i) (15)

the node i residual energy is:

E_cur(i)＝E_init-E_c(i) (16)

wherein E is_initEnergy is initialized for the node.

i defines the encountered set of nodes: defining an encounter set of single nodes, constructing the encounter set of the nodes according to the moving track of the node history, and when the encounter set M (A) of the node A is { (B, t)_a1),(C,t_a2),...,(D,t_am) Denotes node A and node B at t_a1Meet at a moment, node A and node C meet at t_a2Meet at a moment, node A and node D meet at t_amMeet at any moment;

II defines the moving path between nodes: according to the meeting time set of the node carrying the message copy and other nodes, removing outdated records in the meeting set, and constructing a moving path between nodes by taking the node carrying the message copy as an initial node, wherein the nodes in the path meet in sequence according to the time sequence; when the moving path is A-B-C-D, the meeting time of the nodes A and B is earlier than that of the nodes B and C, and the meeting time of the nodes B and C is earlier than that of the nodes C and D;

III, screening a moving path candidate set: screening a mobile path candidate set R capable of completing a message transmission task, wherein the mobile path candidate set R comprises two types of mobile paths: the first type is that a destination node exists on a moving path; the second type is that the nodes on the moving path can upload messages through the wireless network; when in useThe node i carries the message copy m to meet the neighbor nodes, finds out all moving paths of which the node i is an initial node, and sets the moment generated by the message m as t₀If the message survival time is TLL, the effective time of the message m is [ t ]₀,t₀+TLL]；

The screening process of the moving path candidate set R is as follows: screening of the first type of movement path: checking all the moving paths which are found out in the past and take the node i as an initial node, searching whether a destination node of the message m exists on the moving path, if the destination node of the message m exists on the moving path and the time from the initial node to the destination node is less than t₀+ TLL, adding the moving path from the starting node to the destination node into a moving path candidate set R; and (3) screening of a second type of moving path: checking all the mobile paths which are found out in the front and take the node i as an initial node, searching whether the node enters a wireless network coverage area during the message survival period exists on the mobile paths, and if the node exists, adding the mobile path from the initial position to the node into a mobile path candidate set R;

E_exp＝E_t*d (17)

in the above formula, E_expFor the expected energy consumption of the movement path, E_tRepresenting the energy consumption for transmitting the message to the next hop node on the moving path, and d is the hop number from the starting node to the destination node; the expected energy consumption for the second type of movement path is calculated as follows: the data transmission power of the adopted 3G, LTE and WIFI network data transmission power model is calculated according to the following formula:

P＝α_ut_u+α_dt_d+β (18)

in the above formula, P is the power of the transmitted data, t_uFor the uplink rate, t_dFor the downlink rate, α_uIn order to upload the power parameter(s),α_dfor receiving power parameters, beta is the basic power under different networks;

E_ud(t)＝P*t (19)

in the above formula, E_ud(t) the energy consumption for transmitting data in 3G, LTE and WIFI networks, P the calculated power for transmitting data, and t the duration of transmitting data; and calculating the expected energy consumption of the second type of moving paths in the moving path candidate set R by combining the transmission energy consumption under different networks:

E_exp＝E_t*d+E_ud(t) (20)

in the above formula, E_ud(t) energy consumption for data transmission in 3G, LTE, WIFI networks, E_expFor expected energy consumption of the movement path, E_tEnergy consumption for transmitting the message to the next hop node on the moving path, and d is the number of hops from the starting node to the destination node; and sequencing the expected energy consumption of the moving paths in the set R, traversing the set R from the beginning until finding out the optimal moving path with the least expected energy consumption, and then transmitting the single message copy according to the optimal path.

Example 3:

an example of the algorithm: step 1: initializing a message copy by a node; initialization of a source node: the data of the sensor collected by the mobile node generates a message to be forwarded, the mobile node becomes a source node of the message data, the source node copies the message to be forwarded into L message copies after generating the message to be forwarded, and the initialization energy of the node is obtained as E_initWhen the mobile node is in the underground state, the mobile node is initialized; initialization of a mobile node: when the mobile node obtains the message data, initializing to obtain the node initialization energy E_initWhen the mobile node is in the underground state, the mobile node is initialized; after initialization is completed, different information transmission modes are adopted according to the number of the message copies carried by the nodes, and when the number of the message copies is larger than 1, a step 2 is carried out; when the number of the message copies is equal to 1, entering a step 3;

and step 3: single copy transmission based on optimal expected energy consumption: when the number of the message copies carried by the node is 1, a single copy transmission strategy based on expected energy consumption is implemented, the node meets the neighbor node, a moving path is obtained according to historical moving track information, and the moving path which can complete a data transmission task during the message survival period is selected to be added into a candidate set R; and selecting a moving path with the optimal expected energy consumption from the candidate set R, if the second node of the path is a neighbor node, forwarding the data copy to the neighbor node, and otherwise, not forwarding the data copy with the message. The node parameter settings are shown in the following table:

an example of the algorithm for non-urgent data is shown in fig. 3, and it is assumed that the message copy number of the source node S is m (S), the destination node is a node H, and the neighbor nodes of the source node S are an a node and a B node. The threshold value of the message at this time is:

so the source node S distributes

A copy of the message is given to node a. Node A continues to move with the message copy at this point, and when nodes E, C and D are encountered, the threshold value of the message at this point

Node A will allocate

A message copy to node E, distribute

A copy of the message is given to node C. And then, the node A carries the message copies to continue moving, and when the number of the message copies of the node A is one, the node A enters a single copy transmission stage based on the optimal expected energy consumption. At this time, when the node a encounters the node F and the node G, a node on a moving path with the node a as an initial node reaches the wireless network coverage area, the 3G network used for uploading data, the energy consumption for transmitting the data from the node a to the node Q entering the wireless network coverage area is 500J, the energy consumption for uploading the data by the node Q through the 3G network is 1700J, and the total expected energy consumption is 2200J. A node arrives at a wireless network coverage area on a moving path with a node A as an initial node, a second node is a neighbor node G, Wifi is used for uploading data, the energy consumption of transmitting the data from the node A to a node P entering the wireless network coverage area is 1000J, the energy consumption of uploading the data by the node P through the Wifi is 1000J, and the expected energy consumption is 2000J. The node a is used as a destination node on a moving path of the initial node, the second node is a neighbor node F, and the energy consumption for transmitting data from the node a to the destination node H is 1200J. And transmitting the message copy to the neighbor node F according to the single copy transmission strategy. And finally, the node F meets the destination node H, and the data transmission task is smoothly completed. The algorithm flow of the urgent data is similar to that of the non-urgent data, and only in the multi-copy distribution stage based on the utility value, the message copy is distributed to each neighbor node according to the utility value without considering the threshold of the utility value.

Simulation analysis: in order to evaluate the performance of the MTADR algorithm, the performance of the algorithm is evaluated mainly from four aspects of message transmission success rate, average transmission delay, network overhead and average residual energy of nodes, and the result is compared with Epidemic, SprayAndFocus and QoN-ASW. The experiment is based onone simulation experiment platform [ 2 ]

A,Ott J,

T.TheONE simulator for DTN protocol evaluation[C]//Proceedings of the 2nd international conference on simulation tools and techniques.2009:1-10.]An MBM movement model is adopted. The simulation scene is a coal mine underground scene, two types of mobile nodes and wireless base station nodes of employees and mine cars are mainly arranged in the coal mine underground, and the parameter settings of the nodes in the network are shown in the following table:

node type	Staff member	Mine car	Radio base station
				Moving speed (m/s)	[0.5,1.5]	[4,5]	0
Transmission speed (kbps)	300	300	1024
				Communication radius (m)	50	50	200

The global parameter settings for the ONE platform are shown in the following table:

categories	Value of
		Simulation time (h)	12
Message size (B)	[25,100]
		Message generation period(s)	[10,50]
Message lifetime (min)	200

According to the data simulation, the performance evaluation comparison graph sets of the four algorithms are obtained through comparison (shown in figures 4-11).

Referring to fig. 4 and 5, as the number of mobile nodes increases, the success rate of message transmission of the four algorithms increases. Because the Epidemic algorithm is a flooding-based routing algorithm, no screening is adopted for the relay nodes, the number of times of node meeting is increased along with the increase of the number of the nodes, the energy is easily exhausted prematurely, the data cannot be transmitted to the target node, and the success rate of message transmission is lowest. The spark and focus and QoN-ASW algorithm is improved on the spark and wait algorithm, and the blind forwarding of the message is avoided, so the success rate of message transmission is improved compared with the Epidemic algorithm. The MTADR algorithm distributes the copies according to the utility values of the nodes by implementing a multi-copy distribution strategy, thereby avoiding the blind forwarding of the message; the screening of the relay nodes is carried out by implementing a single copy transmission strategy, so that the MTADR algorithm is optimal on the performance index of the success rate of message transmission. As can be seen from comparison between fig. 4 and fig. 5, for urgent data, the MTADR algorithm does not set a threshold value, more nodes participate in data forwarding, and the energy is not exhausted early as in the Epidemic algorithm due to the limitation of the number of message copies. Compared with the non-emergency data setting threshold value, the message transmission success rate for sending the emergency data is higher.

Referring to fig. 6 and 7, as the number of mobile nodes increases, the average transmission delay of the four algorithms decreases. Because the Epidemic algorithm does not screen the relay nodes, the energy is exhausted too early along with the increase of the number of the nodes, and the message cannot be forwarded to the destination node in time. The SprayAndFocus algorithm and the QoN-ASW algorithm screen the relay nodes, so that the message transmission delay is reduced compared with the Epidemic algorithm. The MTADR algorithm distributes message copies according to the size of the utility value when a multi-copy distribution strategy is implemented; the relay node is strictly screened by implementing a single copy transmission strategy, and the message copy is forwarded to the neighbor node which can complete message delivery and is expected to have optimal energy consumption, so that the MTADR algorithm is optimal in performance index of message average transmission delay. As can be seen from comparison between fig. 6 and fig. 7, for the emergency data, more nodes are involved in forwarding, although there is an increase in energy consumption, the average transmission delay for delivering the emergency data is lower than that for delivering the non-emergency data.

Referring to fig. 8 and 9, as the number of mobile nodes increases, the network overhead of all four algorithms decreases. Because the Epidemic algorithm does not screen the relay nodes, the number of copies in the network is excessive, and the control data is the most, the network overhead is the highest. The QoN-ASW algorithm and the SprayAndFocus algorithm control the number of copies, so the normalized network overhead is reduced compared with the Epidemic algorithm. The MTADR algorithm performs the screening of the relay nodes in the implementation of a multi-copy distribution strategy and a single-copy transmission strategy, the success rate of message transmission is high, and the node for distributing the message copies is the least compared with the first three algorithms, so the network overhead is the least. As can be seen from the comparison between fig. 8 and fig. 9, the emergency data has less overhead than the non-emergency data network because the transmission success rate of the emergency data is higher and the number of successfully transmitted data packets is greater.

Referring to fig. 10 and 11, as the simulation time increases, the residual energy of all four algorithms decreases. In the Epidemic algorithm, a node carrying a message forwards a copy when encountering a node not carrying a copy of the message, and frequent forwarding causes faster energy consumption. The QoN-ASW algorithm and the SprayAndWait algorithm limit the number of copies, and compared with Epidemic, the energy consumption caused by the unlimited message copy forwarding is reduced, so the average residual energy of all nodes is more than that of the Epidemic algorithm. The MTADR algorithm considers the residual energy of the nodes in the utility value evaluation of the multi-copy stage, and selects the node with the optimal expected energy consumption as the relay node in the single-copy stage, so that the average residual energy of all the nodes in the network is the highest. As can be seen from comparison between fig. 10 and fig. 11, since the threshold value is not set for delivering the urgent data, more nodes participate in message forwarding, which causes more energy consumption, and therefore, the average residual energy of the nodes delivering the non-urgent data is higher than that of the nodes delivering the urgent data.

Claims

1. A mine safety monitoring opportunity network routing method based on a moving track is characterized by comprising the following steps:

the opportunistic network routing method is based on the following equipment:

the mine car wireless network switching system comprises a switch connected with a mine wired network, the switch is arranged both on the mine and in the mine, a wireless network mobile terminal arranged on a mine car and a portable intelligent mobile terminal carried by miners are arranged everywhere in the mine, sensors with various functions are arranged underground, the sensors have a sensing data detection function and a wireless network communication function, in the method, the mine car and the mobile terminal carried by the miners are regarded as mobile nodes, and the switch in the mine is regarded as a fixed node;

step 1: initializing a message copy by a node;

initialization of a source node: the data of the sensor collected by the mobile node generates a message to be forwarded, the mobile node becomes a source node of the message data, the source node copies the message to be forwarded into L message copies after generating the message to be forwarded, and the initialization energy of the node is obtained as E_initWhen the mobile node moves along the past track information, the initialization is completed;

initialization of a mobile node: when the mobile node obtains the message data, initializing to obtain the node initialization energy of E_initWhen the mobile node moves along the past track information, the initialization is completed;

after initialization is completed, different information transmission modes are adopted according to the number of the message copies carried by the nodes, and when the number of the message copies is larger than 1, a step 2 is carried out; when the number of the message copies is equal to 1, entering a step 3;

and 2, step: multi-copy allocation transmission based on utility values:

when the number of the message copies carried by the node is more than 1, a multi-copy allocation strategy based on the utility value is implemented: the node carrying the message meets the neighbor node, searches whether the neighbor node has a destination node, and forwards the message to the destination node if the neighbor node has the destination node; if no destination node exists, calculating a utility value according to the historical movement track information: if the message is urgent data, distributing the number of message copies according to the utility value; if the message is non-urgent data, setting a utility value threshold value, selecting nodes higher than the utility value threshold value, and distributing the message copy number according to the utility value by the nodes;

and 3, step 3: single copy transmission based on optimal expected energy consumption:

when the number of the message copies carried by the node is 1, a single copy transmission strategy based on expected energy consumption is implemented, the node meets the neighbor node, a moving path is obtained according to historical moving track information, and the moving path which can complete a data transmission task during the message survival period is selected to be added into a candidate set R;

and selecting a mobile path with optimal expected energy consumption from the candidate set R, if a second node of the path is a neighbor node, forwarding the data copy to the neighbor node, and if not, forwarding the data copy with the message.

2. The mine safety monitoring opportunity network routing method based on the moving track as claimed in claim 1, wherein:

the step 2: multi-copy allocation transmission based on utility values:

after a source node generates a message to be forwarded, firstly, the source node copies a message copy with the quantity of L in a cache of the source node, a node carrying the message copy meets other nodes in the moving process, and if the meeting node does not have the message copy, a message copy distribution strategy selects whether to distribute the message copy and the quantity of the distributed message copy according to the transmission capacity of the meeting node; the transmission capacity of the node is represented by a utility value, and the utility value is calculated based on the activity and the residual energy of the node; the node with high utility value obtains more message copies, and transmits data to the destination node by using the stronger transmission capability of the node;

number of initial message copies of node i:

the node initial message copy number L is calculated according to the following formula:

II, similarity of moving tracks of nodes:

the moving tracks of all the mobile nodes in the last day are obtained through an underground positioning system, and because the moving track of a miner deviates from the historical moving track, similarity calculation needs to be carried out on the current moving track and the historical moving track of the miner so as to measure the usability of the historical moving track information of the nodes; calculating the track similarity by adopting the following method:

acquiring the position information of the node once every t time, wherein the historical track of the node consists of track points according to the time sequence; let the history movement track be { p₁,p₂,...,p_n1The current moving track of the node is { q }₁,q₂,...,q_m1And the calculation formula of the matching number is as follows:

in the above formula, L (p, q) is the number of matching points in the two tracks; x is min (m)₁,n₁) Wherein m is₁The number of moving track points of the current node, n₁The number of the historical moving track points is set;

the trajectory similarity of node i is calculated as follows:

in the above formula, s (i) is the track similarity of the node i, and L (p, q) is the number of matching points in the two tracks; x is min (m)₁,n₁) Wherein m is₁As a moving rail of the current nodeNumber of tracing points, n₁The number of the historical moving track points is set;

III, node activity degree:

calculating the node activity degree by combining the underground application scene of the coal mine and considering the work type attribute of the miner node; when the node i and the node j meet, the activity of the node is calculated according to the following formula (7):

CEN(i,j)＝N_a(i)∪N_a(j) (6)

in the above formula, N_ac(i) Indicating the liveness of node i, N_l(i) Represents a node set N which is to be encountered by the node i in the historical moving track information acquired by the positioning system_o(i) Represents a set of nodes previously encountered by node i; wherein b is N_l(i) The number of the types of the jobs in the set, a is the sum of the number of different jobs encountered by the nodes i and j according to the historical track, and s (i) is the track similarity of the nodes i, see formula (5);

and IV utility value calculation:

calculating a node utility value U (i) based on the node activity and the node residual energy:

v, copy distribution of different data types:

considering that resources of an intelligent mobile terminal are limited in a mine, urgent real-time data and non-urgent non-real-time data exist, and different copy distribution strategies are adopted for different types of data; and selecting whether to set a utility value threshold according to the urgency of the message: for the emergency data, a utility value threshold is not set, and more nodes are distributed to message copies to participate in message forwarding; for non-emergency data, setting utility value threshold, when utility value U (i) of adjacent node is higher than utility value threshold U_th(i) Message copies can be distributed according to the size of the utility value, so that nodes with strong performance participate in copy distribution;

the utility value threshold is calculated as follows:

and VI message copy allocation:

when a node i carrying a message copy meets a neighbor node, distributing the message copy for each neighbor node without the message copy according to the size of a utility value, if the neighbor node is non-urgent data, selecting the neighbor node higher than a utility value threshold, and then distributing the copy according to the utility value; assuming that there are k neighbor nodes, the neighbor node j calculates the number of distributed message copies according to the following formula:

3. The mine safety monitoring opportunity network routing method based on the moving track as claimed in claim 2, wherein:

the step 2: current residual energy E of node i in multi-copy distribution transmission based on utility value_cur(i) The calculation method of (2) is as follows:

node scanning energy consumption refers to energy consumed by node scanning channel, and then scanning energy consumption E of node i_s(i) Can be expressed as:

the data transmission energy consumption of the node is proportional to the transmitted data quantity, and the energy consumed by the node i for transmitting unit data is e_tThe amount of transmitted data is s_tThen the sending energy consumption E of the node i_t(i) Can be expressed as:

E_t(i)＝e_t×s_t (13)

similarly, the energy consumption for receiving data is proportional to the amount of data received by the node, and the energy consumed by the node i for receiving unit data is e_rReceived by node iData quantity is s_rThen reception energy E of node i_r(i) Can be expressed as:

E_r(i)＝e_r×s_r (14)

in summary, the total energy consumption E of node i_c(i) Can be expressed as:

E_c(i)＝E_t(i)+E_r(i)+E_s(i) (15)

the node i residual energy is:

E_cur(i)＝E_init-E_c(i) (16)

wherein, E_initEnergy is initialized for the node.

4. The mine safety monitoring opportunity network routing method based on the moving track as claimed in claim 3, wherein:

the step 3: single copy transmission based on optimal expected energy consumption:

due to the fact that energy of an intelligent mobile terminal carried under a mine is limited, when the number of copies of a message m carried by a node is 1, a single copy transmission strategy based on optimal expected energy consumption is implemented; the goal of this strategy is to transmit the message to the destination node within the message lifetime while selecting the node path that forwards the least energy consumption expected:

i defines the encountered set of nodes:

defining an encounter set of single nodes, constructing the encounter set of the nodes according to the moving track of the node history, and when the encounter set M (A) of the node A is { (B, t)_a1),(C,t_a2),...,(D,t_am) Denotes node A and node B are at t_a1The time is met, the node A and the node C are at t_a2The time is met, the node A and the node D are at t_amMeet at any moment;

II defines the moving path between nodes:

according to the meeting time set of the node carrying the message copy and other nodes, removing outdated records in the meeting set, and constructing a moving path between nodes by taking the node carrying the message copy as an initial node, wherein the nodes in the path meet in sequence according to the time sequence; when the moving path is A-B-C-D, the meeting time of the nodes A and B is earlier than that of the nodes B and C, and the meeting time of the nodes B and C is earlier than that of the nodes C and D;

III, screening a moving path candidate set:

screening a mobile path candidate set R capable of completing a message transmission task, wherein the mobile path candidate set R comprises two types of mobile paths: the first type is that a destination node exists on a moving path; the second type is that the nodes on the moving path can upload messages through the wireless network; when the node i carries the message copy m to meet the neighbor nodes, finding out all the moving paths of the node i as the initial node, and setting the moment when the message m is generated as t₀If the message survival time is TLL, the effective time of the message m is [ t ]₀,t₀+TLL]；

The screening process of the moving path candidate set R is as follows:

screening of the first type of movement path: checking all the moving paths which are found out in the past and take the node i as an initial node, searching whether a destination node of the message m exists on the moving path, if the destination node of the message m exists on the moving path and the time from the initial node to the destination node is less than t₀+ TLL, adding the mobile path from the starting node to the destination node into a mobile path candidate set R;

and (3) screening of a second type of moving path: checking all the mobile paths which are found out in the prior art and take the node i as an initial node, searching whether the node enters a wireless network coverage area during the message survival period exists on the mobile paths, and if the node exists, adding the mobile path from the initial position to the node into a mobile path candidate set R;

and IV, calculating the expected energy consumption of the moving path:

calculating expected energy consumption of each moving path in the set for the screened moving path candidate set R capable of completing the data transmission task, and selecting a moving path capable of completing message transmission and having the least expected energy consumption;

the expected energy consumption for the first type of travel path is calculated as follows:

E_exp＝E_t*d (17)

the adopted 3G, LTE and WIFI network data transmission power model is that the transmission data power is calculated according to the following formula:

P＝α_ut_u+α_dt_d+β (18)

E_ud(t)＝P*t (19)

in the above formula, E_ud(t) the energy consumption for transmitting data in 3G, LTE and WIFI networks, P the calculated power for transmitting data, and t the duration of transmitting data;

and calculating the expected energy consumption of the second type of moving paths in the moving path candidate set R by combining the transmission energy consumption of the different networks:

E_exp＝E_t*d+E_ud(t) (20)

in the above formula, E_ud(t) energy consumption for data transmission in 3G, LTE, WIFI networks, E_expFor the expected energy consumption of the movement path, E_tRepresenting the energy consumption for transmitting the message to the next hop node on the moving path, and d is the hop number from the starting node to the destination node;

and then, sequencing the expected energy consumption of the moving paths in the set R, traversing the set R from the beginning until finding out the optimal moving path with the least expected energy consumption, and then transmitting a single message copy according to the optimal path.

5. The mine safety monitoring opportunity network routing method based on the moving track as claimed in claim 4, wherein:

the step 1: initializing a message copy by a node; the sensor classifies the sensed data, judges whether the data is urgent or non-urgent according to the numerical value of the sensed data, and sends the data and the type of the data urgent.