The Research On Design And Key Technology Of Cultural Relics Intrusion Detection System Based On WSN

As known to all, cultural relics in the field has great social, historical and economicsignificance. In order to assess the risk of cultural relics in the field, it is necessary toconstruct detection system to collect cultural relics intrusion by acquiring environmentinformation of the cultural relics in the field rapidly and effectively.According to the practical requirement of heritage conservation, this dissertationpresents a method by which the wireless sensor networks technology is introduced intothe cultural relics in the field monitor system. The major study on the applied basicissues aiming at application of the wireless sensor network technology for cultural relicsintrusion detection contains selection of detection object, research of detection method,data compression and reconstruction, signal feature extraction, neural network structureoptimization, work flow of the system, etc. The main innovation points and conclusionsof this study are as follows:1. Combining with the characteristics and process of cultural relics intrusiondetection and based on the existing research methods, this study presents a new systembased on wireless sensor network technology. It presents a corresponding technologicalroute which uses micro seismic as detection objection.2. In order to reduce the energy consumption on communication, theory ofcompressed sensor has been applied to the detection system. Aiming at the datacharacteristic of intrusion detection system and deficiency of compressed sensor datareconstruction and its existing improved algorithm, this paper proposes an improvedalgorithm named Q-CSDR basing on quantum-inspired artificial immune clonalalgorithm. The algorithm frames the data adaptively and applies the quantum-inspiredimmune clonal algorithm to the procedure of data reconstruction to improve theaccuracy of it. The experimental results show that the performance of Q-CSDR in thispaper is much better than existing algorithm when the sparsity is below60, the accuracyof data reconstruction is higher than85percent.3. An algorithm of signal feature extraction under low sampling frequency isproposed, which improves the algorithm called power spectrum reprocessing, and applies it into the feature exaction of micro-seismic signal. The results of simulationexperiment show that the performance of feature extraction of algorithm we proposed isbetter than the existed ones under10sps. Besides, the algorithm we proposed canimprove the classification accuracy to90percent. For the limitations of algorithm, weproposed two methods for signal collection under high frequence at the end of thischapter.4. Optimize the structure of neural network basing on the chaos quantum-inspiredimmune clone algorithm. The main idea of the algorithm is constructing the populationof hidden layer nodes containing certain sparsity, to search the best performance ofneural network by using chaos quantum-inspired artificial immune clonal algorithm.The simulation result shows that by comparison the algorithm we proposed has thebetter convergence speed and accuracy, and has a smaller approximation error inapproximation function tests. The result of pattern recognition experiment shows that,after delete30percent redundant node in hidden layer, the accuracy of classification isstill maintained at more than92percent.5. In order to build a stable and reliable system, the key technologies whichinclude intrusions detection methods, data compression and reconstruction, featureextraction and classification structure optimization are applied into the cultural relics inthe field intrusion detection system. The experimental results show that the system canadapt to various geological environment，and work stably over2400hours under theaccuracy of classification above93percent.In short, the research results in this paper can provide basis for practical applicationof detection system for culture relics intrusion on the strength of wireless sensornetwork. Meanwhile, it provides an effective new idea to promote the informatization,automation and the intelligentization of culture relics protection for our country.