International Journal of Scientific & Engineering Research Volume 2, Issue 10, Oct-2011 1
ISSN 2229-5518
Implementation of a Large Scale Wireless
Sensor Networks
Dr. Anubhuti Khare, Manish Saxena ,Roshan Singh Thakur , Khyati Chourasia
Abstract: This paper explore the ways to deploy the network so that the workload is evenly distributed, thus the network overall behavior degrades in a smooth fashion. The sensors should be evenly deployed within the monitored area, we look at the approach where a set of more powerful nodes (routers) are designated for data relaying. We look at the approach to deploy these relaying nodes that are easy to implement in practice. In particular, we select sub-regions to deploy them at calculated density. We propose a simple method where the density is simply based on the size of the area whose data will be relayed by these nodes. Our experimental results verify the theoretical estimation of the routers’ density.
Keywords:-W ireless sensor network, routers, low router-sensor ratio, power consumption rate, Medium Access Control (MAC) layer, real-time power-aware protocol (RPAR).
—————————— ——————————
In this paper we look for economic ways to deploy a large scale wireless sensor network. The deployed network must deteriorate evenly and be easy to maintain and replenish. Wireless sensor network is deployed to collect environmental data for a certain region. The basic functionalities of a node in the sensor network include sensing and transmission. For a node, in addition to transmitting its own data, relaying data for other nodes is also a significant task. In fact, one major advocated advantage of sensor network is its ability to route data safely to the collecting host.
The drawback that comes with above advantage is the resulting unevenness of sensor usage. Let’s look at how data is routed. The data collecting host is normally connected by wire to a set of points, which we shall call data sinks. A sink functions as if it is a node in the sensor network, but only receives data wirelessly. Any data generated within the sensor network is considered collected if it reaches one of the sinks. We do not worry about the
communications between the host and the sinks. Obviously sensors closer to a sink carry more workload of relaying data. With a fixed amount of energy source, usually in the form of batteries, their lifetime would be shorter than the rest of the sensors.
With uneven sensor lifetimes, the sensor network becomes problematic. First of all, less data will be collected in areas whose sensors die early. This results in a biased picture of the monitored region. Even worse, once all sensors close to the sinks die, data sampled from sensors that are still good could not reach the host anymore. Replenishing these sensors could remedy the problem, but then we have to spend effort to locate the sensors to replenish.
Deploying more sensors near the sinks could relieve some relaying workload, but then extra sensing capabilities are wasted. Besides the network becomes non-uniform and more difficult to deploy than if the sensors are evenly distributed. If we want evenly distributed sensors and also evenly distributed life time, a set of dedicated relaying nodes is required in the network architecture. In this design, all sensors perform the same job, sampling data and transmit the data to the relaying nodes. This is similar to the ZigBee architecture. And we shall use its terminology to call the relaying nodes “router” nodes. Our paper will concentrate on this approach, which will be detailed in Section IV. It should be pointed out that if we have mobile sinks, we could achieve the same goal in which a sensor does not relay data. All it does is to sample data and send it to the mobile sink once the sink comes by. This, however, will not be addressed in this paper. The sensor deployment now becomes straight forward. We calculate the sensor density based on the sensor capacity and required data density. Then we simply disperse the sensors in the monitored region according to the sensor density.
Consequently, what we shall address is how to deploy the routers. Routers differ from sensors in that they do not contain sensing device and have usually more powerful battery and longer transmission range. For the same reason, however, we still want them to be easily deployed and maintained, and carry even workloads. In many cases the routers may be mainline powered, such as is suggested in ZigBee.
IJSER © 2011 http://w w w .ijser.org
International Journal of Scientific & Engineering Research Volume 2, Issue 10, Oct-2011 2
ISSN 2229-5518
There has been a lot of work on the capacity limit of wireless sensor network .Most of these work accordantly draw the conclusion that the network capacity decreases as it scales up and the decrease rate varies for different topologies and measurements. Considering the difficulty to replace or recharge the sensor’s battery, the capacity limit problem encumbers the deployment of large scale sensor networks. To prolong the network lifetime, many aspects of the problem have been extensively studied. Aim to minimize the transmit power expenditure by reducing the power level of the transmitter while maintaining the network connectivity. Formulates the adjustment of the nodes’ transmit powers as a constrained optimization problem and presents two centralized algorithms and also the proof of their optimality. In two distributed algorithms which dynamically adjust transmit power level on a per-node basis are proposed. Presents a novel pair wise transmits power control algorithm in which every node builds a model for each of its neighbors, describing the correlation between transmit power and link quality.
In the Medium Access Control (MAC) layer, multiple power saving approaches are proposed focusing on conserving battery power by switching off their radio when they do not have to send or receive packets. presents a new power saving MAC protocol, NPSM, which removes the ATIM window overhead from PSM in IEEE 802.11 in order to increase channel capacity for data transmission and reduce the energy consumption. In sensor- MAC (S-MAC) protocol is explicitly designed for wireless sensor networks. It reduces the waste of energy from collision, overhearing, control packet overhead and idle listening by accepting some reduction in both pre-hop fairness and latency. Power-aware routing is also a hot research topic and has been explored in several previous works. These protocols are proposed in different scenarios to maximize sensor network lifetime. Studies three power-aware routing algorithms: centralized version, distributed version and zone-based routing and demonstrates their good empirical competitive ratios with the off-line optimal algorithm. To support real-time communication, proposes the real-time power-aware protocol (RPAR) which achieves application-specified communication delays at low energy cost by dynamically adapting transmission power and routing decisions.
Current research works on sensor network deployment mainly focus on maximizing network coverage and maintaining the connectivity of the network. In the idea of random sampling in geometric sets is applied .It proposes a sampling based approach to decide how many samples (sensor units) must be drawn to make every point in a possibly unknown scene covered by at least one sensor. Studies the problem in the mobile sensor network scenario and deploys nodes one-at-a-time into an unknown environment with each node making use of information gathered by previously deployed nodes to determine its target location and ensuring that nodes retain line-of-sight with one another. With the constraint that the sensor mobility is restricted to a distance- bounded flip, proposes a minimum-cost maximum-flow based solution to determine a movement plan for the sensors in order
to maximize the sensor network coverage and minimize the number off lips. In a cluster-based distributed sensor network scenario, presents a virtual force algorithm (VFA) to use a judicious combination of attractive and repulsive forces to determine virtual motion paths and the rate of movement for the randomly placed sensors. All these approaches target at achieving the network connectivity with a small number of sensors but they ignore the capacity limit problem and are not scalable. our proposed sensor network deployment approach differs from these methods in three ways. First, we deploy the sensors in a random manner. Secondly, we maintain the network connectivity with introduce of routers which are also used to relay the packets from sensors to the sink. Last and most importantly, we achieve the even energy consumption rates for both the sensors and the routers to prolong the lifetime of the network and make the recharge or re-deployment process easy.
A router node does not measure data. It receives data
collected by sensor nodes and routes the data to the sinks via other router nodes. A sensor belongs to a router, to who all its measured data is sent. This requires that for every sensor there must be at least one router within its transmission range. This is not new. In fact it is one of the many possible topologies defined in ZigBee standard, shown in Fig. 1. ZigBee defines
Fig:1 ZigBee topology models
“Reduced Function Device” that does not relay data. There is a “Full Function Device” connecting to each reduced function device to relay its data. Routers and Coordinator are full function devices. Coordinator coordinates routers and devices.
In small scale, routers are mainline powered and wired to the host. Fig.1 shows three basic topologies in ZigBee. In a star topology, sensors are connected to a central router/coordinator. In a cluster tree topology, routers form a tree. Sensors connect directly with tree nodes. In a mesh topology, stars and cluster trees are connected via their routers. A large scale ZigBee network is a mesh. The Reduced Function Device in Fig. 1 is what we call sensor in this paper.
The deployment of the network now becomes the deployment of the routers. Just like with sensors, we want routers to be easily deployed and maintained. If routers are main line powered, we could afford uneven energy usage among routers. And the simple approach would be to deploy routers evenly within the region. However, we shall look at how to deploy the routers that they consume power at close rate; hence they deteriorate at close rate and could be replenished at the same time. This is especially important if routers are also battery
IJSER © 2011 http://w w w .ijser.org
International Journal of Scientific & Engineering Research Volume 2, Issue 10, Oct-2011 3
ISSN 2229-5518
powered. In the following we shall assume battery powered routers when we talk about their lifetime.
Since routers do not measure data, they do not need to be deployed throughout the region. We could afford uneven router deployment because the number of routers is far less than the number of sensors. Routers are more powerful, transmit more data, and have longer transmission range than the sensors. They also last longer. We could, on the other hand, think of a router as a sensor in which the data it senses is the data it collects from its sensors. In this thought the router as a sensor can measure different size of area. They do not need to be evenly deployed.
E E =a1+ a2_d n
router deployment sub-regions and calculate the densities within each sub-region. In practice, however, the monitored region does not have perfect geometric shape. For example, in the region in Fig. 2 need to deploy a sensor network in the rectangular area. It contains a building A, a lake B, and a tree C. The host resides in the building A. The sinks are installed around A’s walls. Sensors are deployed in the rest of the areas. Applying methods in above section is not straight forward.
Fig. 2. Irregular region
The energy consumption rate of the sensors is the
same except the energy to send data. The further a sensor is from a router, the more energy it consumes. Now let’s consider the energy consumption by the router. For each bit of data received and forwarded to the next router at distance d, the energy used E E is
E=a1+a2 * dn
In eq a1, a2, and n nare constants. If the data rate passing a
router is r, then its energy consumption rate is
P= (a1+a2*dn) *r
The time to replenish the routers could be calculated by
dividing the total battery power by P. Or given required router life time, we could calculate P, which in turn dictates how many routers should be deployed.
Our task is now to design a deployment method so that P is the same among routers. To achieve this, we propose to locate a set of sub-regions and define the density within each region, then deploy routers uniformly within each sub-region of the calculated density.
There are several ways to find the region and density:
1) Pre-select the sub-region, and then calculate densities.
Symbol Definition
R the radius of a circular area or the length around an area .
Ds the sensor deployment density. r the router deployment density . rs the sensor data generation rate.
d the distance between two sub-regions.
w the width of the router deployment line .
N the total number of routers deployed. A,B,C the names of areas in consideration.
2) Pre-define the same density among all sub-regions, then locate regions accordingly.
3) Pre-select sub-region and same density, then calculate the size of each sub-region.
We shall give some examples in the next section and once we have the sub-regions and densities, the deployment replenishment become easy.
We have illustrated the ways to locate
Based on our analysis in above section, we propose a simple method that does not produce exact workload among routers,
is simple and produces even enough workloads. It contains the following steps:
1) Draw lines in the area
for router deployment. The lines are selected so that sensor data could be relayed from routers in one line to the routers in the next line, until it reaches the sinks. The distance between two adjacent lines d dshould be the same and less than the transmission range.
2) For each line, calculate the router density factor f. f fis the value of the length of the line dividing the size of the sensor area it covers. If a sensor’s data passes through this line, the area monitored by the sensor is covered by the line. For the region in Fig. 2(a), we draw lines as in Fig. 2(b). In it, the covered area of line 1 is all sensor areas to the left ofit; the covered area of line 4 is the left upper corner.
3) Deploy routers on the lines with the density Dr=f*D0, proportional to its density factor f.D0 is the basic density value. The average power rate of the routers could be estimated as P=Ds*rs*(a1+a2*dn)/D0. From P we can estimate the average lifetime.
The actual application of the method is depicted in Fig. 2(c). Seven lines are drawn with equal distances, which is close to the maximum transmission range. Line 1 covers all the area to the left of and above it; line 2 covers the area to the left of it and above the second dotted line; line 3 covers the area to the left of it and below the second dotted line; etc. We should point out that although we may not derived exact energy usage rate because of the complexity of the energy rate equation, we could at least achieve close usage rate among routers. Some explanations are necessary to defend our approach. The method we proposed assumes uniform sensor distribution and is hence based only on geometric analysis. In a large scale sensor network, uniform sensor distribution makes practical sense. It makes deployment simple. In cases where sensors could not be deployed individually, even sensor density is most likely we can hope for.
IJSER © 2011 http://w w w .ijser.org
International Journal of Scientific & Engineering Research Volume 2, Issue 10, Oct-2011 4
ISSN 2229-5518
As data generated turns to be evenly distributed as well in a large scale area. At some point in time some small area may have more dynamics and generate more data, but over the long run the total amount of data generated from a unit-sized area equals. This also justifies equal sensor distribution. After all, we have equal interest in every place of the whole area covered by the large scale network. If the covered area is small and we could set up sensors individually, the approach proposed in this paper does not apply. There are many active researches on this case .Combining large scale and practical, we need a deployment method that is quick and easy, which is exactly what we set off achieve. The method is not perfect. For example, we do not consider the energy cost difference between sensors of different distances to the routers; we assumed that all sensors could reach a router in one hop. In reality, sensors may not die out as evenly as we hope, but the proposed approach could be a starting point to practically deploy large scale sensor networks. We finish this section with a few words on data routing. In general the sensors send data to routers in the immediate sub-region towards the sink. It is desirable that all routers in the sub-region share sensors equally. Likewise, a router in one sub-region receives data from equal number of routers in another sub-region. However, we have not discussed routing establishment in this paper. Methods studied in other research could be applied
This section presents important results from our simulation studies of the proposed sensor network deployment approach.
A presents our simulation model and parameters. The goal of the experimental studies is to demonstrate that our approach can dramatically extend the lifetime of the sensor network while maintain a low router-sensor ratio and an even router power consumption rate.
The sink is deployed in the center of the round simulation region where sensors are evenly distributed. The routers are
deployed in circles with equal distance from each others. The energy dissipation model follows the setting and for each experiment,100 trials with different sensor distributions are conducted to get the average simulation results.
In this subsection, we first evaluate the accuracy of our
calculations for the required router number in the deployment, the router- sensor ratio and the router power consumption rate respectively comparing with the theoretical analysis. Then, we demonstrate that compared with centralized collection approach and CREM, our approach can dramatically extend the lifetime of the wireless sensor networks.
1)Approach Properties: In this subsection, we evaluate three important properties of the proposed deployment approach: the required router number, the router-sensor ratio and the power consumption rate of the router. For each property, we compared the practical requirement with the theoretical
analysis and the comparisons demonstrate the accuracy of our calculations.
Fig. 3. Router number:
theoretical vs. practical
Fig.3 presents the comparison between the practical required router number and the theoretical estimation. In Fig. 3, we have the observation that the
practical routers required are consistently larger than the theoretical estimation. This is because some remedy routers are needed to maintain the network coverage. But this difference is shrinking along with the increase of the network density and their curves converge when the density is equal or larger than
0.95 in our experiments.
Fig. 4. Sensor numbers vs. router numbers
Fig. 4 compares the practical router-sensor ratio with the theoretical results. According to the analysis, the router- sensor ratio should be not sensitive to the
network density which is clearly shown in Fig. 6. Considering
the effect from remedy routers in practice, the practical router- sensor ratio is larger than the theoretical one but they will quickly converge when the density is increased to 0.95. Another important observation is even when the network density is as low as 0.3, the router-sensor ratio is just around
0.04 and only 39 routers are needed to help relay the packets from more than 850 sensors.
Fig. 5. Router lifetime distribution
In Fig. 5, our proposed approach is demonstrated to achieve an even power consumption rate theoretically regardless oft he distance from the router to the sink. These properties make it easy to estimate the lifetime of the routers and recharge them periodically. The practical simulation results are presented in Fig.6 and Fig 7 in which we increase the network density to 1to get rid of the infection from remedy routers. In Fig. 6, we sort the routers according to their distance to the sink and have the observation that the routers in the distant circle away from the sink will
IJSER © 2011 http://w w w .ijser.org
International Journal of Scientific & Engineering Research Volume 2, Issue 10, Oct-2011 5
ISSN 2229-5518
relay more sensors’ packets in their sub-region. The reason for this phenomenon is the routers in the inner circle also need to help relay the packets transmitted from outer routers. Fig. 7 presents the total number of packets needed to be relayed by each router from the total simulation region. We observe from
Fig. 6. Packets relayed from
Sub Region
Fig 7. Packets relayed from Total Region
Fig 8. Fidelity vs. monitoring cycles
Fig. 7 that statistically the routers will relay similar number of packets to the sink regardless of their distance to the sink but there exists a fluctuation especially when the router is close to the sink. This is because in our deployment approach, the outer circle will have fewer routers compared with the inner circle and the routers in the outer circle will only relay their packets to the nearest router in the inner circle which causes this fluctuation. A simple improvement is to let the router in the outer circle relay their packets to several near routers but not only the nearest one.
2) Lifetime Comparison: In this subsection, we demonstrate the lifetime comparison among our deployment approach, the centralized collection approach and a cluster-based monitoring approach with data aggregation, named CREM. The centralized collection approach flushes the sensor network and constructs a topology tree before the monitoring stage. It collects the packets along the topology tree from leaves to the
root. CREM partitions the sensor network into clusters and organizes the sensor information in an aggregated way to reduce the transmission cost. If we define the fidelity as the percentage of the sensors whose packets can be received by the sink, and assume fidelity of about 90% is acceptable, then from Fig. 8, we have the observation that our approach can dramatically extend the lifetime of the sensor network. It results in a 25 fold increase compared to centralized collection approach and a 4 fold increase compared to CREM. At the same time, the fidelity of our approach is much higher than that of CREM as the fidelity will be reduced as a result of the data aggregation. The underlying principle of this improvement is: in our approach, the sensors only need to relay their data to the associated router in one-hop which eliminate the data redundancy in centralized collection approach and the inaccuracy raised from the data aggregation in CREM.
In summary, in this section, extensive experiments are conducted to present that the proposed sensor network deployment approach achieves a set of important features including the low router-sensor ratio and even router power energy consumption ratio. Based on these properties, our proposed approach can easily deploy and recharge a small set of routers and greatly extend the lifetime of the sensor network compared with other approaches.
The deployment of large scale wireless sensor network has
practical concerns. Many times we could not configure and deploy sensors individually because of either cost or time. Another concern is that we expect even workload so that the network could be easily maintained and replenished. We concluded that evenly distributed sensors contradict evenly distributed workload if the sensors have to relay data. In order to resolve this problem, we look at two layered approach. While sensors measured data, routers relay data. The deployment of sensors is simplified, as well as maintenance and replenishment. Routers do not need to be deployed evenly throughout the region because they are bigger, more powerful, have longer transmission range and bandwidth, and are less in quantity. Still, by pre-selecting sub-regions for routers, we simplify the deployment, maintenance, and replenishment of routers. Our approach is aimed at keeping workload equal among routers. The method could be applied to different energy usage equations. It is demonstrated through extensive experiments that the proposed deployment approach can extremely extend the lifetime of the sensor network and maintain even power consumption rates for both the sensors and routers.
Mrs.Anubhuti Khare one of the authors is indebted to Director
UIT RGPV Bhopal for giving permission for sending the paper to the journal. Manish Saxena is also thankful to the chairman, Bansal Institute of Science &Technology Bhopal for giving permission to send the paper for publication. Last but not least, I would also like to thank our HOD and colleagues for supporting us and also Mr..Roshan who help me a lot.
IJSER © 2011 http://w w w .ijser.org
International Journal of Scientific & Engineering Research Volume 2, Issue 10, Oct-2011 6
ISSN 2229-5518
References
[1] T. F. Abdelzaher, S. Prabh, and R. Kiran, ”On Real-time Capacity Limits of Multihop Wireless Sensor Networks,” Real-Time Systems Symposium, December 2004.
[2] G. Barrenechea, B. Beferull-Lozano, and M. Vetterli, ”Lattice Sensor
Networks: Capacity Limits, Optimal Routing and Robustness to Failures,” the
third
international symposium on Information processing in sensor networks, April
2004.
[3] P. Gupta and P. R. Kumar, ”The Capacity of Wireless Networks,” IEEE Transactions on Information Theory, 46(2), March 2000.
[4] M. Bhardwaj, T. Garnett, and A. P. Chandrakasan, ”Upper Bounds on the
Lifetime of Sensor Networks,” IEEE International Conference on
Communications, Pages 785 - 790, 2001.
[5] D. Chen, A. K. Mok, J. Yi, M. Nixon, T. Aneweer, and R. Shepard, ”Data
Collection with Battery and Buffer Consideration in a Large Scale Sensor Network,” The 11th IEEE International Conference on Embedded and Real- Time Computing Systems and Applications, August 2005.
[6] Z. Hu and B. Li, ”On the Fundamental Capacity and Lifetime Limits of Energy-Constrained Wireless Sensor Networks,” Real-Time Technology and Applications Symposium, May 2004.
[7] R. Ramanathan and R. Rosales-Hain, ”Topology Control of Multihop Wireless Networks using Transmit Power Adjustment,” in Proceedings of the IEEE
Conference on Computer Communications (INFOCOM), pages 404-413, March, 2000.
[8] R. Wattenhofer, L. Li, P. Bahl, and Y. M. Wang, ”Distributed topology
control for power efficient operation in multihop wireless ad hoc networks,” in Proceedings of the IEEE Conference on Computer Communications (INFOCOM), 2001.
[9] S. Narayanaswamy, V. Kawadia, R. S. Sreenivas, and P. R. Kumar, ”Power control in ad-hoc networks: Theory, architecture, algorithm and
implementation
of the COMPOW protocol,” in European Wireless Conference, 2002. [10] M. Kubisch, H. Karl, A. Wolisz, L. Zhong, J. Rabaey, ”Distributed algorithms for transmission power control in wireless sensor networks,” Wireless
Communications and Networking (WCNC), March, 2003.
[11] S. Lin, J. Zhang, G. Zhou, L. Gu, J. A. Stankovic, T. He, ”ATPC: adaptive
transmission power control for wireless sensor networks,” in Proceedings of the 4th international conference on Embedded networked sensor systems,
2006.
[12] G. Ferrari, O. K. Tonguz, S. Panichpapiboon, ”Optimal Transmit Power in
Wireless Sensor Networks,” IEEE Transactions on Mobile Computing, 5(10),
2006.
[13] W. Ye, J. Heidemann, and D. Estrin, ”An energy-efficient MAC protocol for wireless sensor networks,” in Proceedings of the IEEE Conference on Computer Communications (INFOCOM), 2002.
[14] S. Singh, C. S. Raghavendra, ”Power efficient MAC protocol for multihop radio networks,” The Ninth IEEE International Symposium onPersonal, Indoor and Mobile Radio Communications, 1998.
[15] Y. Tseng, C. Hsu and T. Hsieh, ”Power-Saving Protocols for IEEE
802.11-Based Multi-Hop Ad Hoc Networks,” in Proceedings of the IEEE Conference
on Computer Communications (INFOCOM), 2002.
[16] E.-S. Jung and N. Vaidya, ”A power saving mac protocol for wireless networks,” Technical Report, UIUC, July 2002.
[17] S. TAKEUCHI, K. SEZAKI and Y. YASUDA, ”An Improved Power
Saving Mechanism for MAC Protocol in Ad Hoc Networks,” IEICE Transactions
on Communications, E88-B(7):2985-2993, 2005.
[18] J. Aslam, Q. Li, and D. Rus, ”Three power-aware routing algorithms for sensor networks,” Wireless Communications and Mobile Computing, vol. 2, no.3, pp. 187–208, Mar. 2003.
[19] O. Chipara, Z. He, G. Xing, Q. Chen, X. Wang, C. Lu, J.A. Stankovic and
T.F. Abdelzaher, ”Real-time Power-Aware Routing in Sensor Networks,”
IEEE International Workshop on Quality of Service (IWQoS’06), June 2006.
[20] V. Isler, S. Kannan, K. Daniilidis, ”Sampling based sensor-network deployment,” IEEE/RSJ International Conference on Intelligent Robots and Systems,
2004.
[21] S. Kuo, Y. Tseng, F. Wu, C. Lin, ”A Probabilistic Signal-Strength-Based
Evaluation Methodology for Sensor Network Deployment,” Proceedings of the
19th International Conference on Advanced Information Networking and
Applications, 2005.
[22] S. Chellappan, X. Bai, B. Ma, D. Xuan, ”Sensor networks deployment using flip-based sensors,” IEEE International Conference on Mobile Adhoc
and
Sensor Systems Conference, 2005.
[23] A. Howard, M. J. Matadd, and G. S. Sukhatme, ”An incremental self-
deployment algorithm for mobile sensor networks,” Autonomous Robots, Special
Issue on Intelligent Embedded Systems, 2002.
[24] S. Poduri and G. S. Sukhatme, ”Constrained Coverage for Mobile Sensor Networks,” In IEEE International Conference on Robotics and Automation, April, 2004.
[25] Y. Zou and K. Chakrabarty, ”Sensor deployment and target localization based on virtual forces,” in INFOCOM, 2003.
[26] M. Sheldon, D. Chen, M. Nixon, and A. K. Mok, ”A Practical Approach
to Deploy Large Scale Wireless Sensor Networks,” Workshop on Resource
Provisioning and Management in Sensor Networks, 2005.
[27] S. Han and E. Chan, ”Continuous residual energy monitoring in wireless
sensor networks,” The 2nd International Symposium on Parallel and
Distributed
Processing and Applications, December 2004.
[28] ZigBee Alliance, http://www.zigbee.org.
[29] S. Roundy, P. Wright, and J. Rabaey, “A Study of Low Level Vibrations as a Power Source for Wireless Sensor Nodes,”
Computer Comm., vol. 26, no. 11, pp. 1131-1144, July 2003
[30] H. Kulah and K. Najafi, “An Electromagnetic Micro Power Generator For
Low-Frequency Environmental Vibrations,” Proc
17th IEEE Int’l Conf. Micro Electro Mechanical Systems (MEMS) pp. 237-240, Jan. 2004
[31] S. Meninger et al., “Vibration-to-Electric Energy Conversion,” IEEE Trans.
Very Large Scale Integration (VLSI) Systems, vol. 9, no. 1 pp. 64-76, Feb. 2001 [32] H. Bottner et al., “New Thermoelectric Components Using Microsystem Technologies,” J. Micro Electro Mechanical Systems vol. 13, no. 3, pp. 414-
420, June 2004
[33] “Panasonic Solar Cell Technical Handbook ’98/99,” Aug. 1998
[34] J.M. Kahn, R.H. Katz, and K.S. J. Pister, “Next Century Challenges:
Mobile Networking for ’Smart Dust’,” Proc. Mobicom 1999, pp. 271278, 1999 [35] J. Rabaey et al., “PicoRadio Supports Ad Hoc Ultra-Low Power Wireless Networking,” Computer, pp. 42-48, July 2000..
[36] B. Otis, Y. Chee, and J. Rabaey, “A 400µW-RX 1.6mW-TX Super- Regenerative Transceiver for Wireless Sensor Networks,” in IEEE Int. Solid-State Circuits Conf. Dig. Tech. Papers, Feb. 2005, pp. 396–297,
606.
[37] M. Gehring, R. Moen, L. Ragan, and D. Ash, “A Sequential-Gain IC
for Miniature Short-Range Battery-Powered Wireless Data Receivers,”
in IEEE Int. Solid-State Circuits Conf. Dig. Tech. Papers, Feb. 1995, pp. 142–143.
[38] T. S. Rappaport. Wireless Communications: Principles and
Practice. Prentice Hall, Upper Saddle River, NJ, 1996.
[39] N. D. Tripathi, J. H. Reed, and H. F. Vanlandingham. Handoff in cellular systems. IEEE Personal Communications,
5(6):26–37, Dec. 1998.
IJSER © 2011 http://w w w .ijser.org
International Journal of Scientific & Engineering Research Volume 2, Issue 10, Oct-2011 7
ISSN 2229-5518
IJSER ©2011
http //www IJSer org