SIMULATION MODEL

To evaluate the performance of the proposed hierarchical framework we have conducted simulations by making use of MATLAB. For the simulations, some assumptions are made: (i) the number of layer is equal to or greater than 3, (ii) transmission delay between two adjacent nodes is 2 ms, (iii) data aggregation incurs 100 ms delay, (iv) the maximum number of hops for query or update operations is 10, (v) contention window is 25 ms, (vi) the sensor nodes are deployed in a 120m x 120m square area, and (vii) the radio range of each sensor node is 25 meters.

4.2 Performance evaluation

4.2.1 Hierarchy and anchor nodes

The WSN consists of large number of sensor nodes which generate and deliver lots of data packets to sink node. Thus, we proposed a hierarchical framework with data aggregation to cope with heavy workload at sink. And, the query and update delays can be reduced effectively and the energy consumption is slashed. As shown in Fig. 4.1, the simulation results show that the average update delay of the proposed hierarchical framework is less than the update delay of the flat structure. This validates the proposed hierarchical framework.

Figure 4.2 shows that if the number of sensor nodes in the hierarchy raise, the collision probability of the proposed hierarchical framework raises slowly but the collision probability increases substantially in flat structure. This is because the

Fig. 4.1: Update delay.

Fig. 4.2: Collision probability.

hierarchical framework can decrease redundant transmissions due to data aggregation at cluster heads. Note that data packets are retransmitted if collision occurs. So, the energy is consumed much and transmission efficiency is degraded. Obviously, more sensor nodes result in much collision as shown in Fig. 4.2.

Figure 4.3 shows retransmissions due to collision for the hierarchical and flat structures. The simulation results show that retransmissions in hierarchical structure are less than those in flat structure. This is because the data packets are delivered by CHs in hierarchical structure so that the number of transmissions is reduced and collision probability is decreased too.

Fig. 4.3: Average retransmissions caused by collision.

Fig. 4.4: Average route length.

Figure 4.4 shows the average route length for hierarchical structure, flat structure, and the shortest path (ideal). The average route length from sensing node to sink node for hierarchical structure is shorter than that in flat structure and is closer to the ideal case. The ideal route length is the distance of the straight line from sensing node to sink node.

Fig. 4.5: Number of layer vs update delay.

4.2.2 Framework and QoS Constraints

Figure 4.5 shows the relationships between the number of hierarchical framework and the QoS constraints (i.e., update and query delay constraints).

Similarly, the relationships between the layer for anchor nodes and the QoS constraints are shown in Fig. 4.6. When D is set to some fixed values, i.e., 102, 302, _q and 1002 ms, the relationships between the number of layers in the framework (i.e., h) and the upper bound of update delay (i.e., D ) are depicted in Fig. 4.5. Note that, _u in Fig. 4.5, the value of h must be equal to or greater than 3, so that both constraints of D and _u D can be met. As shown in Fig. 4.6, if _q D is set to 102 ms, the anchor _q layer must be located at the layer that is higher than layer 3 (Fig.4.6). When D _q increases to 302 ms and 1002 ms, the anchor layer must be set at layer 4. On the other hand, if D is set to 1002 ms, the anchor layer can be set at the lowest layer, i.e., _q layer 1. Note that, in Fig.4.5 and Fig.4.6, layer 0 means that no proper layer can satisfy the QoS constraints. In other words, the hierarchical framework can not be constructed to meet the QoS constraints.

Fig. 4.6: Anchor layer vs update delay.

Fig. 4.7: Number of layer vs query delay.

Figure 4.7 and Fig. 4.8 show the number of layers in the hierarchical framework and the anchor layer vary with the QoS constraints, i.e., D and _u D . Similarly, Fig. _q 4.7 shows the similar results to Fig. 4.5. Figure 4.8 shows that if D is set to 202 ms, _u the anchor layer is located at layer 1 due to D . If _q D is set to 402 ms and 702 ms, _u the layer of anchor nodes decreases as D increases. When _q D is much greater _q than D , the anchor layer will be located at the lower layer. _u

Fig. 4.8: Anchor layer vs query delay.

Fig. 4.9: Query/update delay vs anchor layer.

Figure 4.9 illustrates how the query and update delays vary with the anchor layer.

When the anchor layer increases, query delay decreases and update delay increases.

Since the query operation involve a round-trip transmission (i.e., from sink to anchor node and then back to sink), the query delay is larger than the delay of update operation, which is from nodes to the anchor node only.

4.2.3 Node density

If the total number of sensor nodes increases, the node density in the framework raises too. Figure 4.10 shows the relationships between node density and the number of layers in the framework. Note that D represents the number of nodes in a Voronoi

Fig. 4.10: Total number of node vs layers.

region. Obviously, the total number of sensor nodes raises as the layer and D increase.

Chapter 5 Conclusions

Adopting data aggregation techniques and hierarchical structure in a WSN for tracking moving objects are effective to prolong the lifetime of the WSNs. However, delay constraints on update and query operations for tracking critical objects should be considered in the WSN that adopts data aggregation and hierarchical structure. So, object tracking applications on WSNs, which meet the QoS constraints, can provide accurate object location and real-time response.

In this paper, we proposed a hierarchical framework that is based on the concept of Voronoi diagram. All of the sensor nodes in the WSN are organized into Voronoi regions hierarchically. Each Voronoi region is associated with a cluster head, which collects the sensory data. Then, the cluster head delivers the collected data upward its cluster head at the higher layer. Moreover, we proposed the anchor nodes approach.

The anchor nodes are located at some layer in the framework. Each anchor node is responsible for storing and aggregating the sensory data. Thus, the proposed hierarchical framework with anchor nodes can support critical object tracking and meet QoS constraints, such as query and update delays. Moreover, we conducted simulations to validate and evaluate the proposed framework. Our simulation results show that the hierarchical framework can achieve better performance (i.e., smaller update delay, lower collision probability, and shorter transmission path) than the flat structure. Besides, the simulation study explores the relationships between anchor layer and the query/update delay constraints. Generally speaking, selecting proper layer in the hierarchical framework to deploy anchor nodes can facilitate the

Reference

[1] J. Blumenthal, M. Handy, F. Golatowski, M. Haase and D. Timmermann,

“Wireless sensor networks - new challenges in software engineering,” Proc. IEEE Int. Conf. on Emerging Technologies and Factory Automation (ETFA), Sept. 2003, pp. 551-556.

[2] I. Akyildiz, W. Su, Y. Sankarasubramaniam and E. Cayirci, "A survey on sensor networks," IEEE Communications Magazine, vol. 40, issue 8, pp. 102-114, Aug.2002.

[3] T. He, L. Gu, L. Luo, T. Yan, J. Stankovic, and S. Son, “An overview of data aggregation architecture for real-time tracking with sensor networks,” Proc. Int.

Conf. on Parallel and Distributed Processing Symposium (IPDPS), Apr. 2006.

[4] A.-S. Pathan, H.-W. Lee and C. Hong, "Security in wireless sensor networks:

issues and challenges," Proc. IEEE Int. Conf. on Advanced Communication Technology, Feb.2006, pp. -1048.

[5] Y.-P. Chen, A. Liestman, and J. Liu, “A hierarchical energy-efficient framework for data aggregation in wireless sensor network,” IEEE Trans. on Vehicular Technology, vol. 55, no.3, pp. 789-796, May 2006.

[6] S. Bandyopadhyay and E. Coyle, “An energy efficient hierarchical clustering algorithm for wireless sensor networks,” Proc. IEEE Int. Conf. on Computer and Communications, April 2003, pp1713-1723.

[7] C.-Y. Lin, W.-C. Peng, and Y.-C. Tseng, “Efficient in-network moving object tracking in wireless sensor networks,” IEEE Trans. on Mobile Computing, vol. 5, no. 8, pp. 1044-1056, Aug. 2006.

[8] J. Zhu, S. Papavassilious, S. Kafetzoglou and J. Yang, “An efficient QoS-constrained data aggregation and processing approach in distributed wireless sensor networks,“ Proc. IEEE Int. Conf. on Computers and Communications, June 2006, pp 257-262.

[9] J. Zhu, S. Papavassilious and J. Yang, “Adaptive localized QoS-constrained data aggregation and processing in distributed sensor networks,” IEEE Trans. on Parallel and Distributed Systems, vol. 17, no. 9, pp. 923-933, Sept. 2006.

[10] L. Xiao, J. Wang, K.-S. Lui and K. Nahrstedt, "Advertising interdomain QoS routing information," Proc. IEEE Int. Conf. on Selected Areas in Communications, Dec. 2004. pp. 1949-1964.

[11] A. Beben, "EQ-BGP: an efficient inter-domain QoS routing protocol," Proc.

IEEE Int. Conf. on Advanced Information Networking and Applications, Apr.

2006

[12] J. Luo, H.-Z. Liu, R.-F. Li and L. Bao, "QoS-oriented asynchronous clustering protocol in wireless sensor networks," Proc. IEEE Int. Conf. on Wireless Communications, Networking and Mobile Computing, Oct.2008, pp. 1-4.

[13] S. Diamond and M. Ceruti, "Application of wireless sensor network to military information integration," Proc. IEEE Int. Conf. on Industrial Informatics, June 2007, pp. 23-27.

[14] J. Du and W. Shi, "App-MAC: An application-aware event-oriented MAC protocol for multimodality wireless sensor networks," IEEE Vehicular Technology, vol.57, issue 6, pp.3723-3731, Nov.2008.

[15] M. Hsieh and U. Ozguner, "A parking algorithm for an autonomous vehicle,"

Proc. IEEE Int. Conf. on Intelligent Vehicles, June 2009, pp1155-1160.

[16] T. Inoue, M. Dao and K.-Z. Liu, "Development of an auto-parking system with physical limitations" IEEE SICE 2004 Annual Conference, vol. 2, Aug.2004, pp.

1015-1020.

[17] R. MacRuairi, M. Keane and G. Coleman, "A wireless sensor network application requirements taxonomy," Proc. IEEE Int. Conf. on Sensor Technologies and Applications, Aug.2008, pp. 209-216.

[18] Y. Guo, P. Corke, G. Poulton, T. Wark, G. Bishop-Hurley and D. Swain,

"Animal behaviour understanding using wireless sensor networks," Proc. IEEE Int.

Conf. on Local Computer Networks, Nov.2006, pp. 607-614.

[19] W.-Y. Chung, Y.-D. Lee and S.-J. Jung, "A wireless sensor network compatible wearable u-healthcare monitoring system using integrated ECG, accelerometer and SpO2," Proc. IEEE Int. Conf. on Engineering in Medicine and Biology Society, Aug.2008, pp. 1529-1532.

[20] D. Yupho and J. Kabara, "Continuous vs. Event Driven Routing Protocols for WSNs in Healthcare Environments," Proc. IEEE Int. Conf. on Pervasive Health, Dec.2006, pp. 1-4.

[21] J. Zao, S.-C. Fan, B.-S. Yang, S.-H. Hsu, H.-C. Cheng, M.-L. Liu, H. Huang and F.-Y. Teng, "Custos remote on-demand healthcare aided with wireless sensors and mobile phones," Proc. IEEE Int. Conf. on Systems, Man and Cybernetics, Oct.2008, pp. 2264-2269.

[22] L. Evers and P. Havinga, "Supply chain management automation using wireless sensor networks," Proc. IEEE Int. Conf. on Mobile Ad hoc and Sensor Systems, Oct.2007, pp. 1-3.

[23] M. Faschinger, C. Sastry, A. Patel, N. Tas, "An RFID and wireless sensor network-based implementation of workflow optimization," Proc. IEEE Int. Conf.

on World of Wireless, Mobile and Multimedia Networks, June 2007, pp. 1-8.

[24] V. Tseng, L. Eric and K-W Lin, “An energy-efficient approach for real-time tracking of moving object in multi-level sensor networks,” Proc. IEEE Int. Conf.

on Embedded and Real-Time Computing Systems and Applications, Aug 2005, pp 305-310.

[25] S. Cai, L. Gao, W. Gong and W.-Q. Xu, "On generating internet hierarchical topology," Proc. IEEE Int. Conf. on Decision and Control, Dec. 2004, pp.

4655-4660.

[26] M. Sekine, and K. Sezaki, “Hierarchical aggregation of distributed data for sensor networks” Proc. IEEE TENCON, Nov. 2004, pp 545-548.

[27] S. Duan and X. Yuan, "Exploring hierarchy architecture for wireless sensor networks management," Proc. IEEE Int. Conf. on Wireless and Optical Communications Networks, Apr. 2006.

[28] S. Brown and C. Sreenan, "A study on data aggregation and reliability in managing wireless Sensor Networks," Proc. IEEE Int. Conf. on Mobile Adhoc and Sensor Systems, Oct.2007, pp. 1-7.

[29] L. Krishnamachari, D. Estrin and S. Wicker, “The impact of data aggregation in wireless sensor networks,” Proc. Int. Conf. on Distributed Computing System, July 2002, pp. 575-578.

[30] R. Misra and C. Mandal, "Ant-aggregation: ant colony algorithm for optimal data aggregation in wireless sensor networks," Proc. IEEE Int. Conf. on Wireless and Optical Communications Networks, 2006.

[31] O. Islam, S. Hussain and H. Zhang, "Genetic algorithm for data aggregation trees in wireless sensor networks," Proc. IEEE Int. Conf. on Intelligent Environments, Sept.2007, pp. 312-316.

[32] J. Chen, L. Gong, P. Zeng and Y. Yang, "Several repacking algorithms for data aggregation in wireless sensor network," Proc. IEEE Int. Conf. on Information Acquisition, Aug.2006, pp. 632-637.

[33] Y. Wang, X. Liu and J. Yin, "Requirements of quality of service in wireless sensor network," Proc. IEEE Int. Conf. on Networking, Apr.2006, pp. 116.