• Size: 768.96 KB
  • Uploaded: 2019-02-13 23:04:56
  • Status: Successfully converted

Some snippets from your converted document:

Middle East Technical University Department of Computer Engineering Fall 2011 PROJECT PROPOSAL FOR INTRUDER DETECTION SYSTEM BEE - TECH Çağlar Seylan _1631126 Fatih Semiz _1752476 Güner Orhan _1631050 Tuğba Demir _1630722 SPONSOR COMPANY 1) Motivation & Purpose Intruder Detection Systems are important for homeland security and defense applications. In recent years, there has became many advances in Seismic Footstep Detection Systems for perimeter and zone protection. Seismic security systems have many advantages: 1. They can be hidden underground or within buildings. 2. They do not send any electromagnetic field and because of that, it is not easy to detect their presence and functions. 3. They can be put into areas where it is difficult to observe directly. However, existing seismic systems have also some disadvantages. When there is strong changes in characteristics of seismic signals and associated noise , systems are not operate reliably. Footstep signal levels are quite low and with increasing distance from the sensor, signal levels decreases rapidly. These levels are also affected by weather and climatic factors such as rain, snow, hot weather, etc. Seismic systems for military applications must be convenient for use with heavy environment conditions. Military systems have very strict requirements such as low power consumption ( sensors use battery so it must consume low power for a long duration. ), low false alarm rate. Our purpose is to develop a new method for analyzing seismic signals. We aim to increase sensor detection range for 2 to 3 times and reduce the false alarm rate to an acceptable level. While making development , we will also consider the power consumption ( it must be low ) . If we achieve to eliminate (or at least reduce) the interferences caused by swaying trees, animals, vehicles, and trains,etc., we can reach the false alarm rate we desired. 2) Detailed Description of The Project In this project, an intruder detection system based seismic sensors will be provided. The system will be put into the ground and send an alarm signal to the control center when it detects a seismic movement due to an intruder in a certain radius. Geophone [1] will be used as the seismic sensor. A geophone is a transducer which converts seismic energy to electrical energy. Its response is proportional to the ground velocity. Thus, the more it is closer to the center of seismic activation, the higher amplitude the output will have. The geophone will be integrated with a processor card and a wireless communication unit. When the geophone detects seismic movements, the processor will process the signal to decide whether the seismic movements is due to an intruder or not. If the cause of the seismic movements is an intruder, the wireless communication unit will send an alarm signal to the control center indicating an intruder is detected. At the very basic, the project will consist of four parts. First part is designing an algorithm for signal processing which will be embedded into the processor card. The algorithm will determine whether the signal is due to an intruder or not. The second part is designing a software for wireless communication unit which will be the communication backbone between the detection system and the control center. The third part of the project is designing the software of the control center. The parameters for the system (e.g. threshold values for noise reduction) will be determined through this software. It will have a user-friendly GUI. We will also design a simulation environment (It will have a GUI similar to the GUI of control software) so that we will be able to test the system in this simulator before trying the system in real environments. Since finding an appropriate place to control our system is hard, creating a simulation is an inevitable part of our project which is the fourth part. While designing the project, we will have the following considerations:  The system should reduce the noise as much as possible while preserving the noiseless part of signals undistorted.  The detection range should be as large as possible. Our intruder detection system will have more detection range than ASELSAN’s current detection system algorithm.  False alarm rate should be low as much as possible. We can achieve this enhancement by creating powerful noise reduction algorithm, so that false alarm rate will decrease.  The system should consume low power as much as possible. This enhancement is very crucial due to the difficulties on changing the batteries of the system.The processor card will have also a sleep mode for this purpose.  The system should classify the signals according to whether they are caused by an intruder or not. Classification of the detected objects is the heart of the project. Because no one wants to give an alarm when

Recently converted files (publicly available):