Compressive Sensing in Wireless Sensor Network
Active In SP
Joined: Nov 2010
16-11-2010, 04:53 PM
The concepts of Wireless Sensor Networks and its issues like, energy efficiency, reduce communication bandwidth for maximize battery life, dynamically changed topology, WSN need a low rate inter sensor communication is required so required compression –estimation sensor. Classified as two Fusion Center(FC) and Adhoc WSNs. Another paper relating to this area :
• Jia Mengl , Husheng Li2 , and Zhu Hanl, " Sparse Event Detection in Wireless Sensor Networks using Compressive Sensing", on information theory , may 2009.
describes a a circuit based on a bandgap reference voltage and a CMOS circuit similar to a beta multiplier. An NMOS transistor in triode region has been used in place of a resistor in conventional beta multiplier. The BGR voltage has a positive temperature coefficient to cancel the negative temperature coefficient of the beta multiplie.r
• Bendali A., Audet Y., “A 1-V CMOS current reference with temperature and process compensation”, IEEE Trans. Circuits Syst. I, vol. 54, no. 2, 2007
Temperature compensation is achieved from a bandgap reference using a transimpedence amplifier an intermediate voltage reference is generated. This voltage applied to the gate of a carefully sized nMOS output transistor provides a reference drain current independent of temperature by mutual compensation of mobility and threshold voltage variations
• M. Filanovsky, A. Allam, “Mutual compensation of mobility and threshold voltage temperature effects with applications in CMOS”, IEEE Trans. Circuits Syst. I, Fundam. Theory Appl., vol. 48, no. 7,Jul. 2001.
Mutual compensation of mobility and threshold voltage temperature effects in MOS transistors to obtain ZTC is analysed. A simple voltage reference circuit realized in 0.35- m CMOS process at this bias point.
• S-Edit User Guide
• B. Razavi, “Design of Analog CMOS Integrated Circuits”, Boston, MA: McGraw-Hill, 2001
• Basic working of MOSFET and its characteristics and short channel effects
• Tanner tool and circuit simulation using S-edit
• Current mirror working and applications and temperature and supply voltage compensation in current reference circuits
• Basic working of FGMOSFET
• Study of programmability of FGMOSFET to obtain various current values using a single circuit
• Layout design using Tanner (L-edit)
COMPRESSIVE SENSING IN WIRELESS NETWORKS.pdf (Size: 85.36 KB / Downloads: 102)
Active In SP
Joined: Jun 2010
24-11-2010, 02:33 PM
Document-ECMPDD.doc (Size: 210.5 KB / Downloads: 45)
Efficient Coverage Maintenance Based
On Probabilistic Distributed Detection
A wireless sensor network (WSN) is a wireless network which was specially discovered to monitor physical or environmental conditions, such as temperature, sound, vibration, pressure, motion. It consists of spatially distributed autonomous sensors to detect the conditions. Now-a-days many wireless sensor networks face the critical challenge of keep maintaining sufficient sensing coverage over long term operations with limited energy. In this project and implimentation we propose a new sensing coverage model based on distributed detection theory which captures two important characteristics of sensing networks such as probabilistic detection and data fusion. And also we propose three coverage maintenance protocols such as centralized, Se-Grid, and Co-Grid protocols to achieve required sensing coverage over long periods of time. Therefore the main aim of this project and implimentation is to increase the lifetime of network (sensing coverage maintenance) which has been effectively used by the users.
Wireless sensor networks are an emerging technology that promises an unprecedented ability to monitor the physical environment. Detection of certain events or targets in the environment is an important application of sensor networks. In this project and implimentation, we address one of the fundamental problems in wireless sensor network, namely coverage. Coverage in general, answers the questions about quality of service that can be provided by a particular sensor network. The main problem faced by many wireless sensor networks is efficient coverage maintenance. Coverage maintenance has been proposed as a promising approach to prolong network lifetime.
For that we propose a new coverage maintenance protocol to provide sufficient sensing coverage over a particular area. This could be achieved by activating a subset of nodes while allowing other nodes in network to sleep. We characterize coverage by the minimum event detection probability and the system false alarm rate from the active nodes. In our model, the event detection probability and false alarm rate are computed based on an established distributed detection approach that correlates sensor data or detection decisions from multiple nodes. Our new coverage model is important in that it bridges the gap between coverage maintenance protocols and distributed detection algorithms.
The main areas focused by this project and implimentation are, probabilistic and distributed detection model that characterizes probabilistic detection properties and data fusions. Then the three coverage maintenance protocols that can meet the specified event detection probability and false alarm rate. Finally the simulation result shows that our new sensing model is one of the best sensing coverage models.
However, in existing wireless sensor networks, the coverage maintenance protocols are often designed based on simpler or deterministic detection models such as disc model. In the disc model, a sensor is assumed to have a perfectly circular sensing range within which any event can be sensed or detected. Although this disc model allows a geometric treatment of the coverage problem, but it does not capture the stochastic nature or inherent characteristics of the sensing process. The key challenges that are faced by the existing wireless sensor networks are to reduce both number of nodes activated in network and the network configuration time. Therefore the sensing coverage for the devices in the network for the long time period is not sufficient in many existing wireless sensor networks. These are the major issues that are faced by wireless sensor networks.
In this project and implimentation, we mainly propose a new coverage maintenance protocols based on probabilistic distributed detection and data fusion that can effectively prolong network lifetime by maintaining sufficient sensing coverage over a region using a small number of active nodes while scheduling the others to sleep. We then propose three coverage maintenance protocols that can meet specified event detection probability and false alarm rate. The centralized protocol is used to activate only a small number of sensors but this introduces long coverage configuration delay.
The Se-Grid protocol is used to reduce the configuration time. This could be achieved by dividing the network into separate fusion groups. But this increases the number of active sensors in the network due to lack of collaboration among sensors in different groups. We present a novel distributed coverage maintenance protocol called the Coordinating Grid (Co-Grid) protocol that organizes the network into coordinating fusion groups located on overlapping virtual grids. Through coordination among neighboring fusion groups, Co-Grid can achieve comparable number of active nodes as a centralized algorithm, while reducing the network (re-)configuration time by orders of magnitude. Therefore we can achieve efficient coverage maintenance through the proposed scheme.
Distributed Detection of Nodes
Design of Central Protocol
Design of Virtual Grid Protocols
• Se-Grid Protocol
• Co-Grid Protocol
DISTRIBUTED DETECTION OF NODES:
In this module, we fully concentrate on detection of active nodes in the network which consists of two important characteristics of sensor networks such as probabilistic detection and data fusion. The detection of nodes in network is performed by individual sensors in the region and combining of information from individual sensors are analyzed in data fusion. We have to focus on two basic performance metrics such as probability of detection (PD) and probability of false alarm (PF).
The probability of detection is done by using single sensor detection and data fusion is performed by multi sensor fusion. We use two thresholds namely called as α and β (Performance Bounds) which is used to measure the PD and PF of a node. The event detection probability and false alarm rate are computed based on an established distributed detection approach that correlates sensor data or detection decisions from multiple nodes. To achieve the sensing coverage over a geographic region, first ensures that the PD and PF of a target that appears at any location within the region meet the performance bounds (i.e. α and β).
Single sensor detection:
The probability of detection is nothing but a power of test to find whether nodes target is present or absent in the network over a region. In single sensor detection, the detection performance of a sensor depends on the physical distance between the sensor and the target. We implement two hypotheses namely H0 and H1 which enables us to check whether the nodes target is absent or present respectively. When the signal power decays with distance, the difference between the means of the two hypotheses decreases accordingly, resulting in worse detection performance.
Multi Sensor Fusion:
Multi sensor fusion uses multiple sensors and it is used to combine sensory data from different sources such that the resulting information is in some sense better at the users end. In this project and implimentation, we present a distributed fusion model that can be combined with coverage maintenance protocol. Each node in the network belongs to one or more fusion groups. Fusion group is nothing but essential a cluster in which a node serves as a fusion center and it is responsible for making a final detection decision by fusing data from other nodes in the group. The two existing fusion schemes are value fusion and decision fusion.
In value fusion, each sensor sends its raw energy measurements to the fusion center, which combines the received measurements according to some fusion rule. It then makes the final detection decision by comparing the combined result against a threshold. The decision fusion operates in a distributed manner as follows: Each sensor makes a local decision based on its measurements and sends its decision to the fusion center, which makes a system decision according to the local decisions. Therefore, the target of nodes is detected by using distributed detection module.