CN111385853A - Directional diffusion routing method based on improved ant colony algorithm in wireless sensor network - Google Patents

Abstract

The invention claims to protect a directional diffusion routing method based on an improved ant colony algorithm in a wireless sensor network. In this method, an improved ant colony algorithm is mainly introduced for the directed diffusion protocol. The invention introduces the node energy factor and the global average energy factor into the ant colony algorithm, so that the network energy consumption factor is always considered in the process of finding the optimal path. In the gradient establishment stage, the gradient path is established with the pheromone concentration generated by the improved ant colony algorithm as the gradient field. In the path enhancement stage, according to the iterative effect of the ant colony algorithm, the optimal paths generated under each iteration are counted and ranked, and the top three paths are enhanced as optional routing paths. Compared with the original protocol, the improved model is more suitable for scenarios with huge network nodes, and it improves the life cycle of the network and is more reasonable in path selection.

Description

Directed Diffusion Routing Method Based on Improved Ant Colony Algorithm in Wireless Sensor Networks

technical field

The invention belongs to the field of directional diffusion routing algorithms in wireless sensor networks, in particular to a directional diffusion routing method based on improved ant colony algorithm in wireless sensor networks.

Background technique

With the advent of the 5G era, it will inevitably promote the development of the Internet of Things industry. As an important part of the Internet of Things, the wireless sensor network, its routing protocol is an important part of the WSN network design. The routing protocol mainly guides the data packets from the source The node forwards to the destination node through the network. In different scenarios, WSN routing protocols are also very different. The invention proposes an improvement to the problem of finding strategies for optimizing paths and neglecting network energy consumption in the directional diffusion protocol. In the wireless sensor network, the directional diffusion protocol is the most basic and common plane routing protocol, and it is a data-centric query route. protocol. Therefore, in the process of finding the path, the probe data, messages, etc. are sent to the nodes of the entire network by flooding, and the process of path optimization also needs to traverse all the nodes to obtain. At the same time, because it is only suitable for small-scale networks, the node energy consumption is not considered. However, after improving the ant colony algorithm, this method can simulate the optimal path with relatively balanced path distance and energy size, which can be applied in the scene with large network scale. A directional diffusion model based on improved ant colony algorithm is obtained. In this model, the node energy factor and the global average energy factor are introduced into the gradient establishment stage and the path enhancement stage of the directional diffusion protocol, so that the network energy consumption factor is always considered in the process of optimizing the path. Compared with the original protocol, the improved model is more suitable for scenarios with huge network nodes, and it improves the life cycle of the network and is more reasonable in path selection.

SUMMARY OF THE INVENTION

The present invention aims to solve the problems in the prior art. A directed diffusion routing method based on improved ant colony algorithm in wireless sensor network is proposed. The technical scheme of the present invention is as follows:

A directed diffusion routing method based on an improved ant colony algorithm in a wireless sensor network, comprising the following steps:

Step 1: the step of interest flooding; the sink sink node periodically floods the entire wireless sensor network to broadcast interest interest, and the target node determines the source node position after receiving the corresponding interest message;

Step 2: Gradient establishment; starting from the source node, the sink node is used as the destination node to run the improved ant colony algorithm. The improved ant colony algorithm mainly lies in adding the node energy factor and the global average energy factor to the ant colony algorithm, which is reflected in The heuristic function of the transition probability function and the update method of the pheromone update function are improved, the number of iterations and the number of ants are initialized, the initial energy of each node is initialized, and the next node is found according to the improved transition probability. The improvement of transition probability is mainly based on inspiration The improvement of the formula function, in which the node energy factor is added, and the energy factor is sent and received; at the same time, pheromone is left on the path, and the pheromone concentration gradient field is finally iterated through the continuous exploration of all ants, and the gradient field is found in turn in the gradient establishment stage. The node with the largest gradient constitutes the gradient path;

Step 3: Path enhancement; according to the pheromone concentration gradient field in step 2, calculate and count the top three optimal paths under N iterations, and enhance and route them. The enhancement method of the path enhancement process is still sent by the sink node in the reverse direction. Intensifying the message of interest establishes the path in the opposite direction in the direction of maximum pheromone concentration.

Further, the interest flooding step described in step 1 still takes the flooding process in the original directed diffusion protocol, and does not change.

Further, the number of initialization iterations, the number of ants and the initial energy of each initialized node in the second step are specifically: in the wireless sensor network, assuming that there are n nodes, first initialize, and give these nodes initial energy E ₁ , E ₂ , E ₃ . _{, . _ _ _}

Further, the calculation steps of the pheromone concentration gradient field in the second step are:

(1) The transition probability of the improved ant colony algorithm is shown in formulas (15) and (16):

表示t时刻，蚂蚁k从节点i到节点j的转移概率，τ_ij(t) 表示t时刻残留在路径(i，j)上的信息量，η_ij为启发函数，α为信息启发式因子， β为期望启发式因子，s表示下一跳节点，τ_is(t)表示本次探索i节点到s节点的 信息素，η_is表示本次探索i节点到s节点的启发函数，allowedk表示蚂蚁k下 一步允许选择的节点的集合；In formula (15),

represents the transition probability of ant k from node i to node j at time t, τ _ij (t) represents the amount of information remaining on the path (i, j) at time t, η _ij is the heuristic function, α is the information heuristic factor, β is the expected heuristic factor, s represents the next hop node, τ _is (t) represents the pheromone from node i to node s in this exploration, η _is represents the heuristic function from node i to node s in this exploration, and allowedk represents the ant k set of nodes that are allowed to be selected in the next step;

In formula (16), E _i and E _j are the residual energy values of nodes i and j, respectively, E _Tx is the power consumption of the sensor node to transmit data, and E _Rx is the power consumption of the sensor node to receive data;

The energy consumption of the sensor node sending kbit data:

The energy consumption of the sensor node to receive kbit data:

E _Rx (k)=kE _elec (4)

In formula (17), d ₀ is the distance threshold, d is the Euclidean distance between two nodes, ε _fs and ε _mp are the power consumption of the amplifier, and E _elec is the power consumption per bit of data;

Each ant selects the next hop node according to the transition probability. After each node is completed, it will modify the taboo table and record the remaining pheromone. Until m ants have completed the path and reached the sink node, one iteration is completed. update pheromone.

(2) In the pheromone update stage, the pheromone update is also carried out by the combination of path and energy selection, and the path with a shorter path and a higher average energy is defined as the best path, so that the pheromone on such a path is The concentration will be highlighted compared to other paths, and the update method is shown in formula (19)(20)(21):

τ _ij (t+n)=(1-ρ)τ _ij (t)+Δτ _ij ^best (5)

表示一条路径上所有节点的平均剩余能量的倒数；In formula (19), ρ is the pheromone volatilization factor, Δτ _ij ^best represents the amount of pheromone that needs to be added on the defined optimal path, ω ₀ and 0<ρ<1, represents the pheromone from i to j at time t+n It is equal to the sum of the pheromone retained after volatilization at time t and the pheromone increment on the optimal path under global conditions; Path _best represents the standard of the defined optimal path.

Represents the reciprocal of the average remaining energy of all nodes on a path;

The definition of the optimal path also adds the path average residual energy factor. In Eq. (21), E _avg represents the path average residual energy factor, and its value is equal to the average value of the energy of all nodes on the path, and L represents one of all possible paths. distance, ω ₀ is the weight, used to control the importance between the path length and the average energy when defining the optimal path;

Step 2.2: Establishment of the Gradient Field

According to the improved ant colony algorithm in step 2.1, the gradient field determined by the pheromone concentration will be generated after the operation is completed. During the gradient establishment process, the source node will find the next hop node in turn along the direction with the largest pheromone concentration, and finally Find the sink node, which will constitute the optimal gradient path considering the distance and energy.

Further, the improvement of the enhanced path process in the step 3 specifically includes: according to the iterative process of the ant colony algorithm, each iterative process will generate an optimal path, and under this pheromone update mechanism, when the number of iterations is completed, The optimal 3 paths are selected through the optimal path ranking, that is, the optimal path set is calculated and sorted from small to large Path _best ={Path _best1 ,Path _best2 ,Path _best3 ,...}, and the ranking is selected. For the first three, all nodes passed by the corresponding path are found by recording. In the process of enhancing the path, the source node sends data along the gradient direction, and the sink node will correspondingly strengthen the three optimal paths according to the pheromone level value.

The advantages and beneficial effects of the present invention are as follows:

1. The present invention optimizes the routing of the traditional directed diffusion protocol through the improved ant colony algorithm, reduces the waste of energy consumption and resources in the flood diffusion mode, and uses the improved pheromone gradient field to provide redundant paths for the model, Increases the robustness of the network.

2. The present invention optimizes the gradient establishment process in step 2. First, the improved ant colony algorithm is used to simulate the data transmission process, and the transition probability is optimized by the distance and the energy of the next hop node. On this basis, according to the global average energy and path The distance optimization pheromone update formula, and finally complete the simulation of the improved ant colony algorithm to form the gradient field network of pheromone. This process makes the path optimization take into account the factors of distance and node energy, which helps to prolong the life cycle of the network. This method is a step of using an improved ant colony algorithm to optimize the directional diffusion route. In the establishment of the directional diffusion gradient, the gradient establishment process in the Central Plains is a process of large-scale data flood exploration, and then the gradient is mainly established at the speed at which the data is sent to the sink. The ant colony algorithm itself has the concept of pheromone concentration, and the improved algorithm takes into account the updated pheromone concentration under the conditions of distance and energy, and has the standard to guide the establishment of the gradient, and thus has the standard to guide the optimal path.

3. The present invention performs redundancy improvement on the path enhancement process in step 3. According to the iterative process of the improved ant colony algorithm in step 2, each iteration generates an optimal path, and counts and calculates the top three optimal paths and strengthens them. This kind of processing helps in the later stage of network formation, when data transmission is in the optimal path for a long time and the energy consumption is too fast. The redundant path improves the network life cycle and also strengthens the robustness of the network. This part uses step 2 In the process of iterative exploration of the ant colony, through iterative action, not only the optimal path can be found, but also a sub-optimal path can be provided for the network. Compared with routing protocols such as HREEMR, although the suboptimal path is also searched for the network in the process of path enhancement, it requires additional cumbersome steps. In this method, the exploration process of the improved ant colony algorithm can be directly used to provide redundant paths and avoid unnecessary steps and energy consumption. In this method, the three redundant paths increase the robustness of the network and improve the overall life cycle of the network.

Description of drawings

FIG. 1 is a schematic flow chart of the directional diffusion routing method based on the improved ant colony algorithm in the wireless sensor network of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be described clearly and in detail below with reference to the accompanying drawings in the embodiments of the present invention. The described embodiments are only some of the embodiments of the invention.

The technical scheme that the present invention solves the above-mentioned technical problems is:

The technical scheme that the present invention solves the above-mentioned technical problems is:

Step 1: Interest flooding;

No changes were made to the original directed proliferation protocol. The sink node periodically floods the entire network to broadcast interest, and the target node determines the location of the source node after receiving the corresponding interest message.

Step 2: Gradient establishment;

According to the special network model of the directed diffusion protocol, the ant colony algorithm is first improved, and the node energy factor and the global average energy factor are added to make it suitable for planning network routing. On this basis, the improved ant colony algorithm is used to optimize the path from the source node to the sink node. Starting from the source node, sink node as the destination node, run the improved ant colony algorithm, initialize the number of iterations and the number of ants, initialize the initial energy of each node, and find the next node according to the improved transition probability, and at the same time in the path The pheromone is left on the ant, and the gradient field of the pheromone is finally iterated through the continuous exploration of all the ants. In the gradient establishment stage, the nodes with the largest gradient are sequentially found to form the gradient path.

Specifically include the following steps:

Step 2.1: Run the modified ant colony algorithm

Initialize first. In the wireless sensor network, assuming that there are _n nodes, first initialize, and give these nodes initial energy E ₁ , E ₂ , E ₃ , . . . Initialize the node pheromones τ ₁ , τ ₂ , τ ₃ . . . , τ _n . The number of initialization iterations is N=0, and the number of ants is m. Run the improved ant colony algorithm.

(1) The transition probability of the improved ant colony algorithm is shown in formulas (15) and (16):

表示t时刻，蚂蚁k从节点i到节点j的转移概率。τ_ij(t) 表示t时刻残留在路径(i，j)上的信息量，η_ij为启发函数。α为信息启发式因子，； β为期望启发式因子，s表示下一跳节点，τ_is(t)表示本次探索i节点到s节点的 信息素，η_is表示本次探索i节点到s节点的启发函数，allowedk表示蚂蚁k下 一步允许选择的节点的集合。In formula (15),

Represents the transition probability of ant k from node i to node j at time t. τ _ij (t) represents the amount of information remaining on the path (i, j) at time t, and η _ij is a heuristic function. α is the information heuristic factor, β is the expectation heuristic factor, s represents the next hop node, τ _is (t) represents the pheromone from the i node to the s node in this exploration, and η _is represents the current exploration from the i node to s The heuristic function of the node, allowedk represents the set of nodes that ant k is allowed to select in the next step.

In formula (16), E _i and E _j are the residual energy values of nodes i, j. E _Tx is the power consumption of the sensor node to transmit data, E _Rx is the power consumption of the sensor node to receive data, n is 2 or 4, these two parameters are obtained according to the wireless sensor energy consumption model, as shown in formulas (17) (18) :

The energy consumption of the sensor node sending kbit data:

The energy consumption of the sensor node to receive kbit data:

E _Rx (k)=kE _elec (11)

In formula (17), d ₀ is the distance threshold, and d represents the Euclidean distance between two nodes. ε _fs and ε _mp are amplifier power consumption. E _elec is energy consumption per unit bit of data.

Each ant selects the next hop node according to the transition probability. After each node is completed, it will modify the taboo table and record the remaining pheromone. Until m ants have completed the path and reached the sink node, one iteration is completed. At this time, the pheromone is updated.

(2) In the pheromone update stage, the pheromone update is also carried out by the combination of path and energy selection. And the path with shorter path and higher average energy is defined as the best path, so the pheromone concentration on such path will be prominent compared with other paths. The update method is shown in formula (19)(20)(21):

τ _ij (t+n)=(1-ρ)τ _ij (t)+Δτ _ij ^best (12)

表示一条路径上所有节点的平均剩余能量的倒数。In formula (19), ρ is the pheromone volatilization factor, and 0<ρ<1, Δτ _ij ^best represents the amount of pheromone that needs to be added on the defined optimal path, which means that the pheromone from i to j at time t+n is equal to t The sum of the remaining part of the time pheromone after volatilization and the pheromone increment on the optimal path under global conditions. Path _best represents the criteria for the best path defined.

Represents the reciprocal of the average remaining energy of all nodes on a path.

The definition of the optimal path also incorporates the path average residual energy factor. In formula (21), E _avg represents the average residual energy factor of the path, and its value is equal to the average value of the energy of all nodes on the path. L represents a path distance among all possible paths, and ω ₀ is a weight used to control the importance between path length and average energy when defining the optimal path.

Step 2.2: Establishment of the Gradient Field

According to the improved ant colony algorithm in step 2.1, the gradient field determined by the pheromone concentration will be generated after the operation is completed. During the gradient establishment process, the source node will find the next hop node in turn along the direction with the largest pheromone concentration, and finally Find the sink node, which will constitute the optimal gradient path considering the distance and energy.

Step 3: Path enhancement; according to the iterative process of the ant colony algorithm, each iteration process will generate an optimal path. Under this pheromone update mechanism, when the number of iterations is completed, the optimal 3 paths are selected through the optimal path ranking, that is, the optimal path set is calculated and sorted from small to large Path _best ={Path _best1 ,Path _best2 , Path _best3 ,······}, select the top three, and find all the nodes that the path passes through by recording. In the process of enhancing the path, the source node sends data along the gradient direction, and the sink node will correspondingly enhance the three optimal paths according to the pheromone level value. The above embodiments should be understood as only for illustrating the present invention and not for limiting the protection scope of the present invention. After reading the contents of the description of the present invention, the skilled person can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims (5)

1. a directional diffusion routing method based on improved ant colony algorithm in a wireless sensor network, is characterized in that, comprises the following steps: Step 1: the step of interest flooding; the sink sink node periodically floods the entire wireless sensor network and broadcasts the interest interest, and the target node determines the location of the source node after receiving the corresponding interest message; Step 2: Gradient establishment; starting from the source node, the sink node is used as the destination node to run the improved ant colony algorithm. The improved ant colony algorithm mainly lies in adding the node energy factor and the global average energy factor to the ant colony algorithm, which is reflected in The heuristic function of the transition probability function and the update method of the pheromone update function are improved, the number of iterations and the number of ants are initialized, the initial energy of each node is initialized, and the next node is found according to the improved transition probability. The improvement of transition probability is mainly based on inspiration The improvement of the formula function, in which the node energy factor is added, and the energy factor is sent and received; at the same time, pheromone is left on the path, and the pheromone concentration gradient field is finally iterated through the continuous exploration of all ants, and the gradient field is found in turn in the gradient establishment stage. The node with the largest gradient constitutes the gradient path; Step 3: Path enhancement; according to the pheromone concentration gradient field in step 2, calculate and count the top three optimal paths under N iterations, and enhance and route them. The enhancement method of the path enhancement process is still sent by the sink node in the reverse direction. Intensifying the message of interest establishes the path in the opposite direction in the direction of maximum pheromone concentration. 2. The directed diffusion routing method based on improved ant colony algorithm in the wireless sensor network according to claim 1, is characterized in that, the interest flooding step described in step 1 still takes the flooding process in the original directed diffusion protocol, and does not Change. 3. The directional diffusion routing method based on improved ant colony algorithm in the wireless sensor network according to claim 1, is characterized in that, described step 2 initialization iteration times and the number of ants and initializing each node initial energy are specifically: wireless sensor In the network, suppose there are n nodes, first initialize, give these nodes initial energy E ₁ , E ₂ , E ₃ ,..., E _n , the energy of n nodes at the initial time is the same, initialize the node pheromone τ ₁ , τ ₂ , τ ₃ ... , τ _n , the number of initialization iterations is N=0, and the number of ants is m. 4. The directional diffusion routing method based on improved ant colony algorithm in the wireless sensor network according to one of claims 1-3, is characterized in that, the calculation step of described step 2 pheromone concentration gradient field is: (1) The transition probability of the improved ant colony algorithm is shown in formulas (15) and (16):

In formula (15),

represents the transition probability of ant k from node i to node j at time t, τ _ij (t) represents the amount of information remaining on the path (i, j) at time t, η _ij is the heuristic function, α is the information heuristic factor, β is the expected heuristic factor, s represents the next hop node, τ _is (t) represents the pheromone from node i to node s in this exploration, η _is represents the heuristic function from node i to node s in this exploration, and allowedk represents the ant k set of nodes that are allowed to be selected in the next step; In formula (16), E _i and E _j are the residual energy values of nodes i and j, respectively, E _Tx is the power consumption of the sensor node to transmit data, and E _Rx is the power consumption of the sensor node to receive data; The energy consumption of the sensor node sending kbit data:

The energy consumption of the sensor node to receive kbit data: E _Rx (k)=kE _elec (4) In formula (17), d ₀ is the distance threshold, d is the Euclidean distance between two nodes, ε _fs and ε _mp are the power consumption of the amplifier, and E _elec is the power consumption per bit of data; Each ant selects the next hop node according to the transition probability. After each node is completed, it will modify the taboo table and record the remaining pheromone. Until m ants have completed the path and reached the sink node, one iteration is completed. update pheromone. (2) In the pheromone update stage, the pheromone update is also carried out by the combination of path and energy selection, and the path with a shorter path and a higher average energy is defined as the best path, so that the pheromone on such a path is The concentration will be highlighted compared to other paths, and the update method is shown in formula (19)(20)(21): τ _ij (t+n)=(1-ρ)τ _ij (t)+Δτ _ij ^best (5)

In formula (19), ρ is the pheromone volatilization factor, Δτ _ij ^best represents the amount of pheromone that needs to be added on the defined optimal path, ω ₀ and 0<ρ<1, represents the pheromone from i to j at time t+n It is equal to the sum of the pheromone retained after volatilization at time t and the pheromone increment on the optimal path under global conditions; Path _best represents the standard of the defined optimal path.

Represents the reciprocal of the average remaining energy of all nodes on a path; The definition of the optimal path also adds the path average residual energy factor. In Eq. (21), E _avg represents the path average residual energy factor, and its value is equal to the average value of the energy of all nodes on the path, and L represents one of all possible paths. distance, ω ₀ is the weight, used to control the importance between the path length and the average energy when defining the optimal path; Step 2.2: Establishment of the Gradient Field According to the improved ant colony algorithm in step 2.1, the gradient field determined by the pheromone concentration will be generated after the operation is completed. During the gradient establishment process, the source node will find the next hop node in turn along the direction with the largest pheromone concentration, and finally Find the sink node, which will constitute the optimal gradient path considering the distance and energy. 5. The directional diffusion routing method based on improved ant colony algorithm in wireless sensor network according to claim 4, is characterized in that, the improvement of the enhanced path process in the described step 3 specifically comprises: according to the iterative process of ant colony algorithm, Each iteration process will generate an optimal path. Under this pheromone update mechanism, when the number of iterations is completed, the optimal 3 paths are selected through the optimal path ranking, that is, the optimal path set is calculated and sorted from small to large. Sort Path _best = {Path _best1 , Path _best2 , Path _best3 ,...}, select the top three, and find all the nodes that the path passes through by recording. In the process of enhancing the path, the source node sends data along the gradient direction, The sink node will strengthen the three optimal paths correspondingly according to the pheromone level value.

Images