TMCnet News
Design Analysis on Optimized Calculation Method of Wireless Sensor Network [Sensors & Transducers (Canada)](Sensors & Transducers (Canada) Via Acquire Media NewsEdge) Abstract: In order to reduce node consumption and energy balance, this paper puts forward strategies for energy optimization based on improved ant colony algorithms, so that it can extend the effective time of network. This paper uses mechanism of positive feedback and negative feedback to effectively balance data traffic. In addition, it also adopts self-automatic probability transition function based on energy optimization, arranges many converged nodes and other strategies, which can effectively realize high-efficiency consumption and balance requirement. It uses emulator of NS2 network to carry out simulation test of the proposed improvement strategies; it makes comparison between ant colony algorithms and basic ant colony algorithm, which demonstrates the effectiveness of optimized strategy. Copyright © 2014 IFSA Publishing, S. L. Keywords: Wireless sensor, Network optimization, Ant colony algorithms, Energy balance. (ProQuest: ... denotes formulae omitted.) 1. Introduction Wireless multi-media sensor network is the selforganization, multi-hop composed by many nodes of wireless communication capacity, the nodes in network have function of collecting multi-media information [1, 2], for example, it can collect multimedia information such as pictures, video, audio etc. While multi-media information generally requires network to provide certain service and qualified guarantee in sensor network communication of wireless multi-media, of which the route choice for multi-media information flow is one key problem for service quality of sensor network, it undertakes the task of communicating flow media data correctly from source node to the targeted node under certain restricted conditions. But wireless sensor network is limited by node energy; the traditional route calculation method can not meet its consumption requirement. Ant colony algorithms is applicable to solve combination and optimization, it has stronger dynamic adaptation, so it is conforms to network characteristics of wireless sensor. Therefore, in order to reduce node consumption and energy balance, this paper put forward the energy optimization strategy based on improved ant colony algorithms. The paper innovatively combines the use of pheromone positive feedback and negative feedback mechanism, effectively balance Data flow. In addition, the adaptive probabilistic energy optimization based on the transfer function, the deployment of multiple sink nodes Strategy, implementation of energy consumption and balance performance requirements [3,4]. The basic ant colony algorithm emphasizes the mechanism of positive feedback of pheromone, guide the evolution of the algorithm to the optimal solution. For wireless sensor networks, if all the data are transmitted along the optimal path, the path nodes with fast energy consumption and failure, the influence of network time. The paper innovatively combines the use of pheromone positive feedback and negative feedback mechanism, effectively balance data flow. In addition, the adaptive probabilistic energy optimization based on the transfer function, the deployment of multiple sink nodes, strategy, implementation of energy consumption and balance performance requirements. In this paper, using the NS2 network simulator experiments on improvement strategies put forward, and the flooding algorithm and basic ant colony algorithm carries on the contrast analysis, results show, energy optimization strategy based on improved ant colony algorithm is proposed in this paper, the reduction of energy consumption, energy balance of nodes and network time, are better than the above two algorithms, and verification the effectiveness of the optimization method. 2. Strategy Analysis Based on Improved ant Colony Algorithms 2.1. Asymmetric Information (Directed Information) As it is indicated in Fig. 1, ant perceives the information of route AB in node A, and perceives information of route BA in node B is the same. This situation will cause one problem: ant may deviate from the converged node and will go further. Symmetric information distribution manner may cause ant deviate from the converged node, or goes through many unnecessary nodes then it can reach the converged node and consume pointless energy. First, the information of route AB and route BA is different, when the first ant returns and comes through DB, BA, AS, but it only can distribute information in the single direction, that means it distributes information in route BD, AB, SA. However, when the subsequent ant reaches node B, it will not be affected by the information of route BA, it can choose the node proximately close to converged node as the next hop node [5]. 2.2. Self-adaptation Transition Chance Based on Energy Optimization This paper puts forward the adjustable self-adaptation chance transition function, it sets energy standard difference threshold, when it chooses the next hop node, and it will firstly compare the energy standard difference and threshold of the current node. If it is less than the threshold, it indicates that the energy consumption using current node as central is in the balance state, we can temporarily neglect the indication function of remaining energy and directly use position information as indication manner, the included angle of neighbor node, self node and converged node will become smaller and smaller, the choice chance of this neighbor node will become bigger. If it is bigger than the threshold, it indicates that the energy value using current node as center has been out of balance; it should use the original chance transition function, consider effect of remaining energy and choose the node with higher energy as the next hop node. 2.3. Arrangement of Much Node Convergence There is only one converged node in most wireless sensor route agreement, and the common node communicates data to this node. This kind of network expansion structure has internal problem, it will easily cause difference in node energy, and energy consumption is much bigger, for this, this paper arranges many converged nodes to make data flow dose not flowed to single direction, possible make each node deal with data of the same quantity so that it can realize energy balance. It is indicated in Fig. 2. The arrow indicates data flowing to the converged node, on the single converged node, data flow only has one direction, it will easily cause node become invalid. If it arranges many converged nodes, data flow has many directions, and the flow reaches balance. And node will easily find the nearby converged node, data communication saves consumption. 2.4. Negative Feedback Idea Update Pheromone Use This is the core of the ant colony optimization algorithm makes the wireless sensor network node energy balance [6], and extend the network effectively between. The above mentioned, to achieve this balance, need to be closely combined positive and negative feedback effect. The following information. Update strategy is proposed based on negative feedback, the node energy balance as the starting point. Ant information life cycle (TTL), every shift to the new node, the TTL decrement, until zero, prevent ants. The survival time of interest is too long. The basic ant colony algorithm on TTL for ant news zero discarded directly, not make corresponding processing. In fact, the news has practical significance. It just shows the path can not reach the sink node, should be used. In this paper. In the algorithm, when TTL is zero ant message did not reach the sink node, also along the original road return, return, it put a dent in the path of pheromone, while adding other neighbor node path pheromone. We call this pheromone more new ways for the pheromone diffusion. The update operation through this, to warn the ants don't go this way not to the sink the path to the node, while encouraging them to open up a new path, it is more likely to arrive at the sink node. The measures taken or around the pheromone update strategy, ants decided to transfer the next hop node, pheromone diffusion mentioned above are. On the next hop path has weakened to the pheromone, the path information to other neighbor nodes were strengthened. The benefit of this is to avoid the information pheromone accumulation by reducing effects of inundation of residual energy, resulting in the transfer probability function were slow, that is to say, the remaining energy reduction can influence on the ants, so as to improve the overall efficiency of the algorithm. The paper innovatively presents periodic cutting path pheromone strategy, while maintaining the pheromone (relative path, the original pheromone many update pheromone is better than other multi) under the premise, reduces the different neighbor node path pheromone gap. It can enhance enlighten the residual energy of nodes, but also take into consideration the front ants prior history. 2.3. The Sink Node Load Overload Processing In this paper, optimization strategy referred to the deployment of multiple sink nodes, solves the single sink node cannot achieve equilibrium problems. On the basis of one step closer to the distributed idea, realizing multiple sink nodes request processing data synchronization. Because multiple sink nodes cannot communicate directly, only know its load. This algorithm sets a threshold of load, when a certain period of time, the number of ants to the sink node is greater than the threshold, the sink node is broadcast news. Receiving node broadcast messages are weakened source path pheromone. In order to avoid the "Implosion" problem, defined the broadcast message hop TTL [10]. Weaken the benefits of pheromone is to encourage the back of ants find load sink node lighter, with load balancing each sink node to achieve energy balance network node. 3. Realization of Calculation Method 3.1 Renewal Rule for Node When after node receiving broadcasted information, it will update route table of neighbor information table [7, 8]. If it exists, it will update node energy and location according to broadcasted information. If the information doses not exist, it will add to new table item, indicating that there is new neighbor joins in. At the same time, according to the establishment time of table, it can calculate how long this table has no updating, if the time is overdue, indicate that corresponding node of table may be invalid, it can send out broadcasted information, then it will delete this table. When the ant goes back to source node, it makes updating for the most optimized path in this period. The renewal rule is as follows: of which, ant Path and energy Path are the two word section included by the returned ant; bestMeasure is the calculation measure of the most optimized path in this period. Bestmeaure is calculated by measure. When the measure of returned ant (ant path) is bigger than the most optimized index of the former ant, then it renews the best route, otherwise it is difficult to maintain the original value. ...(1) 3.2. Renewal Rule for Information Information renewal is the core element of adjusting feedback [9]; it has key function for energy balance. There are many places define information in this paper, using balance node as load: 1) Toward the node ant passes by, consumes corresponding route information. The renewal rule is as follows: of which, indicates information on node indicating route ij, p indicates the information factor. ...(2) 2) Going back to the node that ant goes by, the rule of releasing directed information is as follows: ...(3) ...(4) ÄTij indicates the node that ant goes by, it is the information left by route ij. Q is the constant of information, C is one constant related to original energy, sumPathEnergy is the total remaining energy of all the route path of ant from source node to converged node, consumeEnergy indicates the consumed energy total of ant going this path. From the formula we can see that sumPathEnergy of route remaining energy of becomes much more, consumeEnergy of route will become less, the information released by ant will become much more. 3) It is triggered by the timer, the periodic reduction information, the renewal rule is as follows: Tq is the original information, ...(5) 4) In the process of communicating monitoring data to the converged node, and distribute information. The information reaches node i, make the following renewal: spread is the weakened information, N is the neighbor node of node i. ...(6) ... (7) ... (8) 5) Poly node load overload, notify the neighboring node updates pheromone. The sink node broadcasts neighbor node pheromone update message, message and set TTL values, limit Multicast hops. Node receives the type of message, all the updates pheromone table (neighbor information table): ... (9) Chance transition function the detailed chance transition function is as follows, of which zij is the information of route ij, rjj is illumination function. ... (10) ... (11) where sd is the energy standard difference of all the neighbor node of node i, threshold is set in advance. Energy is the remaining energy of node j, Qij indicates node j, node I and converged node s makes the angle Zjis . When sd is bigger than threshold, it indicates that neighbor node energy has much difference and is out of balance, the illumination function uses remaining energy of node. Otherwise it indicates that group energy of neighbor node is in the balanced state, it can neglect energy illumination, Zjis becomes smaller, the value of illumination function will become much bigger. 3.3. The Algorithm Flow Step 1A: monitoring node to send broadcast information of neighbor nodes, (start timer. Timer timeout, turn step 1A). Step IB: monitoring node monitoring network input, if the type of message is broadcast message, to step 2 A; if it is Ant message step 3. Step 2A: according to the broadcast message updating neighbor information table, if the neighbor information list, update, or the new neighbor information list. To step IB. Step 2B: structural ant message, according to the transfer function of forwarding probability, (start timer, turn step 2B). Step 3: message is forward ants ants or return, if forward ants, to step 4, such as Fruit is a return to the ant, to step 7. Step 4: ants to the sink node, if not the sink node, to step 5; if is convergence Node to step 6. Step 5: according to the transition probability function chooses the next hop forwarding messages, ants. To step IB. Step 6: construct returned ant message, message returning along the original path. To step IB. Step 7: determine whether the message to the source node of ants. If not, turn to step 8; if it is, to step 9. Step 8: using residual energy path node as parameters to update the pheromone, the ant in return. Path forwarding the message. To step 1. Step 9: determine whether the returned ant routing period optimal routing, if is, update the optimal routing path. To step IB. 4. Test Analysis In the simulation test of this paper, the node scale is from 50 to 200 nodes, they are respectively 50, 100, 150, 200, and generate 4 simulation scenes to implement test. In order to avoid arranging some node, this paper makes node arrangement standardization, makes emphasis on the validity of demonstration calculation method. The test parameter of simulation is indicated in Table 1. It uses the minimum remaining node, node energy standard difference, effective consumption and effective time of network of node as the judgment basis. The chosen operation system is Ubuntu 10.04, the simulation software of network is NS2 2.34, the development language is C++, and then it gets test result picture of Fig. 3 to Fig 7. 1) The simulation time is 500 seconds, the sensor nodes are initialized with 200J, 5-1 and 5-2 is different. Two forms of nodes under the average residual energy of node. X axis nodes, the nodes are 50, 100, 150, 200, 4 times to do the experiment. The end of the Y axis simulation, the average residual energy of wireless sensor node values. The average node residual energy curve and column are shown in Fig. 3 and Fig. 4 respectively. In this paper, an improved ant colony algorithm node average residual energy is higher than the basic ant colony algorithm and Hong Flooding algorithm. When the node size is small, as shown in Fig.4 node number is equal to 50, the node flooding algorithm of average residual energy is higher than the basic ant colony algorithm. With the increase of nodes, the nodes flooding algorithm of each generated or received a message, to broadcast the neighbor node, the node size will lead to the increase of news "broadcast storm problem, cause Festival "Energy consumption faster. The path the basic ant colony algorithm consumes less energy, and thus the average node residual energy meter the performance is stable. Optimization strategy and improved ant colony algorithm to benefit from this node, in different scale are shown Color performance, less energy consumption. 2) The smallest node residual energy is the minimum value of all the remaining energy of nodes, it is a reflection of the energy consumption of nodes and is Scale index. The comparison curve of minimum remaining energy is shown in Fig.5. The energy consumption fees of improved ant colony algorithms is low, improved ant colony algorithms calculation method dose not choose certain route and consume certain node energy in group, but the minimum remaining energy of node will not become too small, therefore it reflects energy balance of node. While the node quantity is 50 and 100, the minimum remaining energy of basic ant colony algorithms is lower than fooding, because the former is restrained into the route with small consumption. 3) Energy standard node difference is reflected in the node residual energy is directly index equilibrium. Standard deviation is greater, that node. The higher the energy dispersion, deviate from the average energy is far. The comparison curve of node energy standard difference is shown in Fig.6. The energy standard difference of fooding is less than the basic ant colony algorithms calculation method, it indicates that the node energy of fooding is much more balanced than calculation method of basic ant colony algorithms, but there is one precondition: sensor node consumes much more energy in calculation method, so this kind of balance has no meaning. The cause that node balance in basic ant colony algorithms is that calculation method is restrained into the route with small consumption, and it has not considered energy balance. The effective energy consumption is the ratio of the total energy consumption and arrives at the message sink node number of messages. It is seen from Table 2, the basic ant colony. The algorithm is higher than the total energy consumption of the improved ant colony algorithm, reaching number nodes should be small, so the improved ant colony algorithm has. The effectiveness of consumption is better than the basic ant colony algorithm. The comparison curve of effective consumption is shown in Fig. 7. The basic ant colony algorithm and improved ant colony algorithm effectively energy consumption comparison is shown in Table 2. In different nodes, the improved ant colony algorithm to the number of the sink node is higher than the basic ant colony algorithm, but with the increase in size, the difference is more and more, the number of nodes is 50 1.19 times, the number of nodes is 100 Is 2.22 times the number of nodes, 150:3.18 times, where 200: is the number of nodes 4.07 times, showed some ants news because TTL Not be restricted to the sink node, is actually the routing message invalid, wasted energy cost. The improved ant colony optimization Method from optimization strategy, ant messages to the sink node success rate will remain basically unchanged, maintain high performance. Calculation method of improved ant colony algorithms benefits from the optimized strategy, the success chance of ant reaching converged node basically has no change and keeps high-efficiency performance. Description: the flooding does not participate in the comparison, the reason is based on the ant message to calculate the sink node number of messages arrive, but is not calculated according to the number of data messages, flooding in more meaningful. The proposed optimization strategy to solve energy and equilibrium problems, the ultimate goal is to improve the networkTime. The simulation time is 800 seconds, the initial energy of nodes is 50 cokes. Fig. 8 shows the improved ant colony algorithm. Network time method was higher than the basic ant colony algorithm and flooding in the central node scale. Visible, the optimization method of reductionEnergy consumption, energy balance, and extend the network time, practical significance. The effective time of improved ant colony algorithms is higher than calculation method and fooding under the scale of concentrated node. So we can see that optimized strategy has practical meanings to reduce energy consumption, solve energy balance and extend effective time of network. 5. Conclusion The line sensor network consists of a large number of sensor nodes, the wireless communication way through multi hop self-organizing Route the network structure monitoring regional sensing data to the sink node. Its application prospect is broad, has become more than study on hot cross discipline. Research focus in wireless sensor networks is energy optimization problems. The traditional routing algorithm. With many limitations, in a wireless sensor network for example, easy to cause the failure of some nodes, dynamic adaptation is not strong .Points, can not meet the requirements. Ant colony algorithm is suitable for solving combinatorial optimization problems, dynamic adaptability, with wireless sensor Characteristics of network. The paper put forward the innovation of energy optimization strategy based on improved ant colony algorithm, and reduces the energy consumption of nodes, Achieve energy balance, thus prolonging the network time. Wireless sensor network is composed of many sensor nodes, by using wireless communication passing by multi-hop route, transmits the perceived data in monitoring area to the network structure of converged node. Its application is very broad; it has become the research hot topic of multidisciplinary. The research key point of wireless sensor lies in energy optimization. For network of wireless sensor, data concentration chooses the most optimized path, which will enable energy consumption of part node accelerate and will become invalid, shorten the effective time of network. Acknowledgements The authors would express their appreciation for the financial support of Shandong Natural Science Foundation, grant NO. ZR2011EL016. The authors also would express their thanks for Shandong Science and Technology Development Project on Safety in Production, grant NO. LAJK2013-183. References [1]. Sun Limin, Li Jianzhong, Dong Yu, Wireless Sensor Network, Tinghua University Press, 2005. [2]. Yaghmaee M. H, Adjeroh D. A., Priority-based rate control for service differentiation and congestion control in wireless multimedia sensor networks, Computer Networks, Vol. 53, Issue 11, 2009, pp. 1798-1811. [3]. Luo Wusheng, Study on Wireless Multi-media Sensor Network, Electronic and Information Journal, Vol. 30, Issue 6,2008, pp. 1511-1516. [4]. George Hloupisa, Ilias Stavrakas, Konstantinos Moutzouris, Alex Alexandridis, Dimos Triantis. WSN Open Source Development Platform: Application to Green Learning, Procedía Engineering, Vol. 25, Issue 1,2011, pp. 1049-1052. [5]. Ahmad El Kouche, Louai Al-Awami, Hossam Hassanein, Dynamically Reconfigurable Energy Aware Modular Software (DREAMS) Architecture for WSNs in Industrial Environments, Procedía Computer Science, Vol. 5,2011, pp. 264-271. [6]. Wen Hao, QoS System Structure of Wireless Sensor Network, Computer Journal, Vol. 32, Issue 3, 2009, pp. 432-440. [7]. Toh C. K., Maximum battery life routing to support ubiquitous mobile computing in wireless Ad hoc networks, IEEE Communication Magazine, Vol. 06, 2001, pp. 138-147. [8]. Jin Rencheng, One Kind of Node Applicable to Wireless Multi-media Sensor Network Non-intersect Route Agreement, Sensor Technology Journal, Vol. 23, Issue 7,2010, pp. 1000-1005. [9]. Li Shanshan, One of Congestion Perceived Multiple Route Flow Distribution Calculation Method in Wireless Sensor Network, Computer and Engineering, Vol. 30, Issue 3,2008, pp. 86-88. [10]. Md. Nafees Rahman, M. A. Matin, Efficient Algorithm for Prolonging Network Lifetime of Wireless Sensor Networks, Tsinghua Science and Technology, Vol. 16, Issue 6,2011, pp. 561-568. [11]. Evren Güney, Necati Aras, Í. Kuban Altinel, Cem Ersoy, Efficient solution techniques for the integrated coverage, sink location and routing problem in wireless sensor networks, Computers and Operations Research, Vol. 39, Issue 7,2012, pp. 1530-1539. 1,2 Cai Tianfang,2 Yang Zhongguo College of Mechanical and Electrical Engineering, Zaozhuang Institute, 277160, China 1 Tel.: 13589610269, fax: 06323986368 1 E-mail: [email protected] Received: 7 June 2014 /Accepted: 30 June 2014 /Published: 31 July 2014 (c) 2014 IFSA Publishing, S.L. |