CN103686924A - Directed Diffusion Routing Protocol Based on Return Delay for Wireless Sensor Networks - Google Patents

Abstract

The invention discloses an improved oriented diffusion routing protocol in a wireless sensor network. A Sink node broadcasts Interest at a fixed period in a flooding way. A network node establishes gradients between nodes for broadcasting the Interest according to the sequence of the received Interest, and the gradients of the earlier reached nodes are smaller. After a gradient field of the whole network is established, one of the network nodes with the minimal gradient is selected as a next node; and if a plurality of nodes with same minimal gradients are provided, one node is randomly selected. The rest energy of the nodes are periodically detected, a threshold value is reduced by a half if the rest energy is lower than the threshold value; and meanwhile, the information of increasing the gradient is broadcasted to a neighboring node; and the gradient between the node which receives the information and the node is increased by the node which receives the information, and the gradients between the node which receives the information and other neighbours are reduced simultaneously. Results are compared to show that the protocol provided by the invention has longer network lifetime and more balanced network node energy consumption than the orientated diffusion protocol; and the improved oriented diffusion routing protocol is simpler.

Description

Directed Diffusion Routing Protocol Based on Return Delay for Wireless Sensor Networks

technical field

The invention relates to wireless sensor network routing technology.

Background technique

Routing technology is the most researched field in Wireless Sensor Network (WSN). The existing routing methods can be divided into two categories: hierarchical (hierarchical) and flat (flat). LEACH (Low Energy Adaptive Clustering Hierarchy) and Directed Diffusion (DD) are the most typical representatives of these two types. .

The data-centric DD protocol is considered a milestone result of flat routing. The DD protocol includes three parts: spreading interest, establishing gradient and path reinforcement. It selects one or more destination nodes for the next hop by inquiring sensor information. The DD protocol can find the path between the Discovery Sink and the sensor nodes containing the information it is interested in without any sensor node location information. In the process of path establishment, a node only needs to know the situation of its adjacent nodes, which is a routing technology based on local control. The DD protocol can be applied to the acquisition of static information and the tracking of dynamic targets at the same time.

The DD protocol provides two gradients. The first one uses the path delay as the gradient, and the current node randomly selects the next-hop node from multiple adjacent nodes with the probability corresponding to the gradient value. The random selection is based on the consideration of robustness and fairness of node energy consumption, but it also brings that the path of each data transmission is not the shortest, and the energy consumption and delay are not the smallest. The second gradient in the DD protocol is the transmission data rate, but it requires a path reinforcement process. This process is actually a network-wide flood initiated by the source node, which will bring a large transmission overhead.

By improving the DD protocol, the present invention provides a return delay-based directed diffusion routing protocol IEAF-DD (Interest Earliest Arrival First Directed Diffusion). The comparison results show that the IEAF-DD protocol has better energy saving and energy balance effects than the DD protocol, and greatly reduces the complexity.

Contents of the invention

The purpose of the invention is to provide a directed diffusion routing protocol with better energy consumption performance and simpler wireless sensor network.

For realizing the above object, the technical route that the present invention adopts is:

The first step is that the Sink node periodically floods and broadcasts Interest;

In the second step, the network node, such as A, establishes the gradient value between the node sending the Interest according to the order of the arrival of the Interest. Ignore the Interest sent by the node that has established the gradient. For the Interest sent by a node that has not established a gradient, the following formula is used to establish the gradient,

G _r =r

Among them, r represents the serial number of the Interest flood received by A, and G _r represents that A returns the gradient value of the rth (1≤r≤n) arrival node. In this way, the node that arrives first has a minimum gradient of 1, and it increases sequentially.

In the third step, when the network node transmits data, the node with the smallest gradient value between the node and the node is selected as the next-hop node each time. If there are multiple next-hop nodes with the same minimum value, one of them is randomly selected as the next-hop node. By analogy, the data is transmitted from the source node to the sink.

In the fourth step, the node adjusts the gradient with the adjacent nodes according to its own energy consumption. A node queries its own remaining energy every time T ₀ , and when it finds that the energy is lower than the set threshold, it notifies all adjacent nodes to increase the gradient value between them, and then updates the threshold. The present invention adopts the "1/2" method, that is, the new threshold is equal to half of the original threshold. The node's initial initial threshold is equal to half of the node's initial energy. The method of adjusting the gradient between nodes is assuming that the node with the gradient value G _r sends a message to increase the gradient because the energy is too low, and node A receives the message. Node A adjusts the gradient between itself and its neighbor nodes according to the following formula:

$\left\{\begin{array}{cc}{{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}<span\; class="notranslate">\; ;</span>& <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\\ \left.\begin{array}{cc}{{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ;</span>& \\ \left.\begin{array}{cc}{{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}& <span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\\ {{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}& <span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&NotEqual;</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\end{array}\right\}& <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; \&NotEqual;</span>\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; )</span>\end{array}\right\}& <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\end{array}\right.$

Among them, n is the number of neighbors of node A, and G _i ' means that A returns the new gradient value of the ith (1≤i≤n) Interest information to the node.

Beneficial effects of the present invention: the present invention uses time delay as a gradient, and the node selects the next node deterministically according to the gradient of the minimum time delay, and establishes the current shortest route between the source and the sink. At the same time, according to the energy consumption of nodes, the gradient between nodes is adjusted. This avoids selecting paths with redundant nodes and reduces network energy consumption. On the other hand, through the adjustment of the gradient between nodes, the energy consumption of network nodes is balanced. The present invention adopts the "long-term balance" method to balance the energy consumption of network nodes, instead of the "short-term balance" method of the DD protocol. The comparison results show that the present invention has smaller node energy consumption and better balance of node energy consumption than the DD protocol, especially the calculation is simpler.

Description of drawings

Figure 1 is the process of Interest flooding;

Figure 2 is the gradient between nodes;

Figure 3 shows that redundant nodes participate in data transmission according to probability;

Figure 4 is the energy comparison of the nodes with the least energy;

Figure 5 is a comparison of the average remaining energy of nodes.

Detailed ways

Step 1 The Sink node periodically floods and broadcasts Interest. Interest contains the description of information events, such as the characteristics of the object to be detected, etc., for each sensor node to judge whether it matches the data it senses. Figure 1 provides an example of Sink flooding to all network nodes including the data source Source. Sink first broadcasts Interest to all its neighbor nodes, and the node receiving the Interest then broadcasts the received Interest to its neighbor nodes. However, if it receives the Interest broadcasted by the node that its own flooding has reached, it discards the received packet and does not repeat the broadcast. And so on until all nodes in the network receive the Interest.

Step 2: Network node A establishes and sends gradients between the Interest nodes according to the arrival sequence of the Interest. Ignore the Interest sent by the node that has established the gradient. For the Interest sent by a node that has not established a gradient, the following formula is used to establish the gradient:

G _r =r

Among them, r represents the serial number of the Interest flood received by A, and G _r represents that A returns the gradient value of the rth (1≤r≤n) arrival node.

In this way, the node that arrives first has the smallest gradient and increases sequentially. Moreover, the gradient value established between a node and its neighbor nodes is unique. The following is a specific description in conjunction with Figure 2:

1. When Sink floods Interest, it first transmits to node F, and node F establishes a gradient of 1 with Sink after receiving it. Then node F floods Interest to Sink and nodes E, B and D.

2. Since the gradient with F has already been established, the sink ignores the flooding of F.

3. Since nodes E, B, and D receive Interest for the first time, they all establish a gradient of 1 with F.

4. Using the same method,

1) Node E floods to node B, this is the second time B receives Interest, so the gradient between B-E is 2;

2) Node E floods to node A, and the gradient between A and E is established as 1; if E floods later than B, since E is the first

Interest is received twice, so the gradient between E-B is 2;

3) Node D is flooded to node C, and the gradient between C and D is established as 1;

5. Node C floods to node B, this is the third time B receives Interest, so the gradient between B-C is 3;

Step 3 After establishing the gradient field among the nodes of the whole network, start from the Source node, select the node with the smallest gradient as the next hop node, and continue until the Sink.

Different from the DD protocol, the present invention no longer selects the next-hop node from nodes with different gradient values based on the probability of the gradient value. The following will be described in conjunction with FIG. 3 .

In Figure 3, B→F has three optional paths, namely B→F; B→E→F; B→C→D→.

In the DD protocol, one of the three paths is selected with a certain probability, and the present invention only selects the node F with the smallest gradient, and jumps from B to F directly.

If two other paths are selected, the hop numbers between B and F are 2 and 3, respectively. That is to say, it is necessary to add 1 to 2 intermediate nodes to forward data, which correspondingly increases the energy consumption of the network, which is unnecessary in terms of data transmission.

Step 4. The node queries its own remaining energy at intervals such as T ₀ . If the remaining energy is lower than a predetermined threshold, it notifies all its neighbor nodes to increase the gradient value with the node, and then the node updates the threshold.

The threshold update method is to update the new threshold equal to half of the original threshold, and the initial threshold of the node is half of the initial energy of the node.

The method of gradient adjustment is that after node A receives a signal to increase the gradient from node N _r with gradient value G _r , then A adjusts the gradient between itself and its neighbor nodes as follows:

1. For node N _r :

1) If A has only one neighbor N _r , that is, n=1, then G _r '=G _r ;

2) Otherwise G _r '=G _r +1;

2. For a node N _i , where 1≤i≤n and i≠r,

1) If Gi=1, then G _i '=G _i ;

2) Otherwise, G _i '=G _i -1.

Where n is the number of neighbors of node A, and G _i ' means that A returns the new gradient value of the ith (1≤i≤n) Interest information reaching the node.

Fig. 4 illustrates the time from the start of network operation to the energy exhaustion of the first node of the protocol of the present invention and the DD protocol under the same network conditions. Fig. 5 illustrates the variation of the average remaining energy of nodes in the protocol of the present invention and the DD protocol over time under the same network conditions.

Claims (1)

1. a kind of directed diffusion routing protocol based on the improved wireless sensor network of return time delay, it is characterized in that: (1) The Sink node periodically floods and broadcasts Interest; (2) Network node A establishes and broadcasts the gradient between the Interest nodes according to the arrival sequence of the Interest; (2.1) Ignore the Interest broadcast by the node that has established the gradient; (2.2) For the Interest sent by a node that has not established a gradient, use the following formula to establish a gradient, G _r =r Among them, r represents the serial number of the Interest flood received by A, and G _r represents that A returns the gradient value of the rth (1≤r≤n) arrival node. The node that arrives first has the smallest gradient and increases sequentially; (3) When the network node transmits data, (3.1) Select the node with the smallest gradient value between the node and the node as the next hop node; (3.2) If there are multiple nodes with the same minimum gradient, randomly select one of them as the next hop node; (4) The node queries its own remaining energy every interval T ₀ , if the remaining energy is lower than the threshold, (4.1) Update the new threshold equal to half of the original threshold, and the initial threshold of the node is half of the initial energy of the node; (4.2) Notify all neighbor nodes to increase the gradient value with this node. After neighbor node A receives the gradient increase signal from node N _r with gradient value G _r , A updates the gradient between its neighbor nodes as follows: $\left\{\begin{array}{cc}{{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}<span\; class="notranslate">\; ;</span>& <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\\ \left.\begin{array}{cc}{{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ;</span>& \\ \left.\begin{array}{cc}{{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}& <span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\\ {{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}& <span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&NotEqual;</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\end{array}\right\}& <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; \&NotEqual;</span>\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; )</span>\end{array}\right\}& <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\end{array}\right.$ Among them, n is the number of neighbors of node A, and G _i ' is the new gradient value of the ith (1≤i≤n) Interest information that A returns to the node.

Images