Bonuccelli, Director of Hair Department:
Paper comments: Paolo Ferragina, Piero Mastrini.
External referee Bassagni, Manisrivastava: Stifano.
Country music e: bugliesi, panzieri, MEO
From February 27th, 2005, 65438,
abstract
This paper solves this problem and reduces the energy consumption of wireless sensor networks. We put forward a set of technology.
The research field of import strategy can be applied to the design of energy-saving protocol sensor networks. They include
Time series prediction, sensor performance of quorums system and interactive protocol design. We can use these technologies to effectively synchronize time issues, collect data from sensor networks, and ensure strong data consistency in mobile networks.
We have proved that), which is used for approximate query and detection of outliers in sensor networks. This idea is based on the combination of "augmented reality" technology and the local model established at each node, which becomes the root of a global model storage network.
(Library), which is used to roughly answer users' queries. Our method uses much less transmission than before, adopts an approximate method model based on augmented reality technology, and organizes network clusters based on similar nodes between data. Where the data definition is based on similarity coefficient model, it is based on "augmented reality" technology, which reduces energy consumption and directly compares data values, so that we can get an effective clustering algorithm, which is probably the best network formed by the total number of groups. Our cluster structure has several interesting characteristics, which make it suitable for mobile networks: first, they can capture geographically similar neighboring nodes; Second, cluster members do not need to consume additional adaptive nodes; Third, there is no need to track and join other nodes in the industrial cluster. In addition, a roughly correct error boundary provides and allows users to dynamically adjust the quality of answers and effectively answer questions in terms of energy and resources.
In addition, we use AR model to solve the time synchronization problem, which is a new perspective of complementary clock synchronization in biology department. More precisely, the sensor nodes in the case we analyzed decided to skip one or more clock adjustments to achieve energy-saving effect, or were temporarily isolated, but still needed an accurate calculation time. This paper presents a model based on approximate correct clock return, which is based on augmented reality technology to estimate the constraints and problems of time at a constant (adjustable error probability). This method is adaptable and allows the sensor to determine how many.
While maintaining the accuracy, the clock adjustment can be skipped, thus saving energy. In addition, we propose a set of determination methods to reduce the time estimation error by at least one factor 2. More precisely, we propose a deterministic clock reading method, called DCR method. By using the sign of clock deviation of relevant information, the frequency adjustment can be reduced by half a clock period while maintaining the same error.
Interested in the practice and theory of this method. In fact, it leads to obvious energy saving, and explains in detail that the strong reality clock model can lead to fine optimal timing synchronization to the maximum deviation clock. In addition, we also propose a generalized version of DCR method to improve its accuracy, which depends on a stable clock, and the monotonicity of the method ensures the time value of production.
We analyzed the technical background of the system for the first time, that is, the statutory sensor network: we reformed it and consumed energy for our own benefit [6]. Quorum systems have the potential to save energy in sensor networks, because they can reduce the obvious communication times, improve the load balance between sensor nodes and improve the scalability of the system. However, the quorum system and quorum measurement standard proposed by wired network are not suitable for sensor networks because they do not solve the performance characteristics and limitations of sensor networks. These observations urge us to redesign the quorum system and the corresponding measurement standards, taking into account the limitations and characteristics of the sensor (for example, transmission cost, limited energy)
Source, physical radio broadcast), network topology. More precisely, we redefine the following legal indicators: load balance, access cost and legal capacity, and design a strategy to reduce the number of people communicating with sensor networks based on some characteristics. We use these strategies to design an energy-saving and highly flexible home number system. In particular, we propose a quorum to reduce the construction cost, and an energy-saving access data diffusion protocol to reduce transmission energy consumption and shorten collisions.
In addition, we analyze the mobility of high nodes in the number system. More precisely, the problem of our learning difficulties ensures that the moving path will follow the unknown node at the intersection of two quotas [7]. We solve this problem, define a novel liquidity model that provides the minimum set of constraints, and derive economic data strong enough to ensure a highly mobile network. In this case, we will tell the famous schools about the previous quorum system and provide a necessary condition to ensure data availability and atomic consistency under high node mobility. We also propose a new class.
The quota system is called mobile communication (brackish water desalination, suitable for high quota mobile networks), and an optimal quota aspect, size (such as communication transmission), is proposed [7]. Then, we use the general manager's quorum system to realize a roughly correct atomic read/write network with mobile and sparse memory.
book list
[1], Captain Toulon D. Madden, National Taiwan Province Province. PAQ: Time series prediction is an approximate query answer.
In a sensor network. In the third process. European seminar, page 1- 1 1. Wireless sensor network 21-37 February 2006.
[2] Captain Toulon D. Madden, National Taiwan Province Province. The similarity of nodes in energy-saving detection sensor networks is constructed by querying.
Submit to the meeting.
[3] Tulong. Global feasibility evaluation of wireless sensor networks under isolated conditions.
Appear in Algorithmica.
[4] Tulong. Energy-saving time estimation of wireless sensor networks. According to the principles of the fourth seminar of Proc. , page 1- 1 1. 52-59 Mobile Computing, June 65438, 2004+10.
Toulon & gt[5]. How can we get the progress between reference times efficiently and accurately? International. Symptoms. Distributed computing, June 5438+00.
In 2003. A brief statement on pages 25-32.
[6] D. Des Moines Toulon & gt& gt. Redesigned arbitration system in wireless sensor networks. Submit to the meeting.
Toulon & gt[7]. In a highly mobile network, is it possible to ensure that strong economic data are relevant? Submit to the meeting.