Researches On Cooperative Spectrum Sensing Technology Based On Energy Consumption In Cognitive Wireless Sensor Networks

The cognitive wireless sensor network can take advantage of the free frequency to overcome the problems of traditional spectrum resources shortage.Cognitive wireless sensor network is different from classical wireless sensor network.It is a new kind of intelligent wireless communications network.Cognitive wireless sensor networks regularly communicate in the ISM band.It can also use spectrum sensing technology to use the spectrum hole for communication,which improves the performance of the wireless sensor network.In addition,cognitive radio technology not only brings us convenience but also adds additional challenges to us.Because of weak cognitive node ability,we need to consider the energy saving,node cooperation and other issues.We cannot directly apply the traditional technology of cognitive radio network.In order to promote the application of wireless sensor network,it is an important research to increase the speed of data processing for sensor node and lengthen the service life of the network.The thesis firstly introduces the background of wireless sensor network and classification.Due to the scarcity of frequency spectrum resource,it greatly restricts the development of WSN.Cognitive radio technology uses spectrum sensing technology to access the spectrum resources dynamically.In order to improve the utilization of spectrum resources.Our issue "cognitive wireless sensor network spectrum sensing technology" is the combination of wireless sensor network with cognitive radio technology.In this thesis,the second part mainly introduces the classification of data fusion algorithm,as well as three ways of transmitter detection to solve the problem of decision fusion in collaborative spectrum sensing.Matched filter detection,energy detection and cyclostationary characteristic detection are all transmitter detections.Without a priori knowledge about the primary users,energy detection can adapt to a variety of scenarios.This thesis uses the energy detection method.Due to the channel fading and hidden terminal,the spectrum sensing performance of single node is weak.Thus,we use multiple nodes to sense spectrum collaboratively.We also gives detection and false alarm probability expressions of "and" fusion,"or" fusion and "vote" fusion,laying the foundation for later research.In addition to obtain better cognitive performance,this thesis will adopt collaborative spectrum sensing technology.As a result of the energy of sensor nodes are weak and not easy to recharge,energy consumption problem is particularly severe.The third part of the thesis is about minimizing the energy consumption from two ways.The first one is to maximize throughput of the second users and the second way is to minimize the energy consumption.To meet certain detection probability and false alarm probability,select optimal number of participated second users.In addition to channel fading and shadow effect,the information second users send may make mistakes.This thesis also analyzes the influence of channel errors to the optimal number of users.In the final,the simulation of "and" fusion,"or" fusion and "vote" fusion shows that throughput efficiency of voting fusion is highest.The last part of the thesis puts forward the Censoring Scheme and Sleep mechanism to the spectrum sensing.This method can greatly reduce the energy consumption of each node.We ameliorate the strategy with double threshold for the OR fusion algorithm.The second users only send decision 1 and we can derive the primary user is present.For AND fusion algorithm,if all the second users send 1,we can determine primary user is present.In this thesis,the double threshold is improved.The OR fusion algorithm only sends the decision judged to be 1 to the cluster head.AND fusion algorithm only sends the decision judged to be 1 to cluster heads.We can further reduce the energy consumption.Use the sleep mechanism,satisfying the given detection probability and false probability.The simulation shows that this mechanism can greatly reduce the energy consumption.In addition,the sending energy consumption is related to the distance.So this thesis adopts clustering method.The distance of nodes in the same cluster is less than the threshold distance.The thesis is based on clustering cooperative spectrum sensing algorithm and derives the decision threshold parameters and sleep probability.Due to adopting clustering algorithm,not all of the second users directly send decisions to the fusion center.At first,second users send their results to the cluster head,and then cluster heads send calculated results to the fusion center.In this way,energy consumption is reduced largely.