Research On Spectrum Sensing Technology In Wireless Cognitive Sensor Networks

Most of current Wireless Sensor Networks (WSN) all work in ISM frequency band without authorization (Industrial Scientific and Medical frequencies, such as commonly used2.4GHz frequency band), and the rapid development of a variety of emerging wireless communication technology makes the public spectrum (such as WIFI, Bluetooth, Wimax, Zigbee, etc.) increasingly crowded, and the interference between the various wireless technology in ISM band has become increasingly serious. In addition, in the traditional wireless sensor network, the spectrum utilization is fixed and static, and the efficiency is very low, which is unable to efficiently harness the use of dynamic spectrum resources. Based on the above reasons, some experts and scholars try to introduce cognitive radio technology to reform the traditional WSN so that the sensor nodes can sense the spectrum environment, detect primary users, use idle spectrum, as Wireless Cognitive Sensor Network(WCSN) equipped with dynamic spectrum utilization ability.Spectrum sensing is one of the main functionality to distinguish WCSNs from traditional WSNs, and is also the main issue researched in this paper. Current spectrum sensing methods include Hybrid sensing techniques, Cooperative sensing and Sensing based on collaborative PU statistics. And current WCSN architectures include Ad-Hoc WCSN, Clustered WCSN, Heterogeneous and Hierarchical WCSN, and Mobile WCSN. The spectrum sensing mechanisms under current WCSN architectures has much room to improve in reducing energy consumption, optimization of the collaborative sensing to reduce network overhead, and reducing the network delay, etc.Based on the analysis of existing network architecture, a new energy efficient hybrid co-operative spectrum sensing technique which associates energy and Eigen value based detectors is proposed to improve the sensing efficiency and sensing reliability to enhance the performance of WCSN. All cognitive sensor nodes are equipped with a simple energy detector, and the detection center is also equipped with Eigen value based detectors. Simulation results show that with the comparison of a cycle-stationary based hybrid scheme, the proposed hybrid scheme can improve the efficiency and reliability of the sensing ability and energy efficiency, and has a stronger adaptability. The lifetime of the entire network has obviously improved.