In the wireless sensor network, the WLAN interference, multi-path fading andattenuation are the main reason for packets’ corruption. Forward errorcorrection is one of the methods for error mitigation. Many coding methods aredesigned to improve the reliability of wireless channel. However, none of thesecoding methods could handle the burst errors that widely exist in wirelesschannel and are mainly responsible for packet corruption. Interleaving isintroduced to robust these coding methods and make those corrupted packetsavailable to be corrected by spreading the burst errors. The Reed-Solomon(15,7) block code is used as the forwarding error correction in theseexperiments. In this paper, seven different symbol interleaving will be firstlysimulated in the Matlab platform under four different channels to evaluate theirpacket error rate performances. Then, two of the seven interleaving with betterperformance and the simplest interleaving algorithm, general block interleavingwill be implemented in the TinyOS platform, and they will be compared witheach other from packet error rate, interleaving time and memory consumptionaspects. Experiments show that, for theoretical channel, interleaving plays animportant role to enhance the capability to correct the corrupted packets exceptfor Rayleigh channel, in which the impact of interleaving is generally slight.Meanwhile for the channel under IEEE 802.11b/g interference, interleavingenhance more than 10% PER in the receiver. However for MFA channel, theenhancement is slight, only 1.85%. Convolutional interleaving has the best PERperformance in AWGN channel and the channel with deterministic SER, whilematrix interleaving has the lowest PER in WLAN affected channel. In thelatency and memory consumption aspect, For the same packet length, generalblock interleaving has the shortest interleaving time while matrix interleavinghas the longest. In the same interleaving algorithm, the execution time becomeslonger with growth of packet size. Within the same interleaving method, thelonger the packet size is, the more memory that would be consumed. Under thesame interleaving length, general block interleaving consumes least memoryand convolutional interleaving has the largest memory consumption.

This work deals with implementation of an agent platform, which is able to run agent code in wireless sensor networks. Implementation has been done for MICAz platform, which uses TinyOS operating system for developing applications. This work contains list of chosen TinyOS parts and illustrates, how such a platform can be used for our purposes. We will describe main features of ALLL language and we will also demonstrate some examples of agents.

The aim of this diploma thesis is to analyze, design and create the architecture of the WSageNt system web interface. The main focus of the system will be traffic monitoring and topology control of the network. The work describes basic technologies, design principles and implementation methods.

Wireless Sensor Networks (WSNs) have gained significant consideration these days, as opposed to wired sensor networks, by introducing multi-functional wireless nodes, which are smaller in size. The main advan-tage is that its wireless, so it costs less to install, maintain and reconfig-ure. These sensor nodes are used in various application areas. For ex-ample: residential, industrial, environmental and military application areas. However WSNs communication is prone to negative influences from the physical environment, such as physical obstacles and interfer-ence. Algorithms must be developed for handling these problems and also to investigate the channel properties. The purpose of this work is to design a testbed, which enables the communication of wireless sensor nodes, to capture the properties of the channel, which will, in the long run, enable better solutions to be designed which are, more appropriate to the errors in the channel. In this testbed, one channel, from the IEEE 802.15.4 channels spectrum, is dedicated as an emergency channel, which is used for handshaking and to handle channel/external interfer-ence or hardware failure between the communication of Transmitter (TX) and Receiver (RX) nodes. The remaining 15 channels are called data channels and are used for actual data transmission and control signals. Peer to Peer transmission of a transmitter-receiver pair is achieved with the introducing of beacons and acknowledgment (ACK) packets. The testbed also has the property of dual reception and data logging on a single PC by maeans of two RX nodes simultaneously from a single TX node. The dependency of the packet on the “Frame Length” byte(in the Frame header) during the reception is eliminated so that if the “Frame Length” byte is compromised, it will replace the programmer defined value with the “Frame Length” byte which helps to investigate the actual packets byte sent from the TX. Also, the Received Signal Strength (RSS) is calculated at the maximum sample rate of the channel. Power consumption is not considered in this testbed. The work is conducted on both the IEEE 802.15.4 physical and the application layers. Linux based TinyOS-2.x is used as an operating system for low power sensor devices. New algorithms are designed for each step in the development of the testbed. MICAz motes are used as nodes and an MIB520 programming board is used for burning the codes and for the purpose of gateways.

This thesis describes the possibilities of wireless sensor networks in monitoring buildings and summarizes the most common use of wireless sensor networks in practice. There are also described communication protocols needed for communication between each node and properties of topology of created network. Importance is lay on implementation of fully dynamic network with automatic network topology establishment. The application is designed for the TinyOS operating system and hardware platforms MICAz and IRIS, and its task is to establish the topology and send the data obtained from sensors to base station. The implemented protocol is based on the Collection Tree Protocol protocol and using best effort delivery is able to establish a dynamic topology. Further there is described  the method of storage and subsequent graphical and textual representation of data collected.

This thesis presents the development of a surveillance testbed using wireless sensors and video cameras for vehicle detection and tracking. The experimental study includes testbed design and discusses some of the implementation issues in using wireless sensors and video cameras for a practical application. A group of sensor devices equipped with light sensors are used to detect and localize the position of moving vehicle. Background subtraction method is used to detect the moving vehicle from the video sequences. Vehicle centroid is calculated in each frame. A non-linear minimization method is used to estimate the perspective transformation which project 3D points to 2D image points. Vehicle location estimates from three cameras are fused to form a single trajectory representing the vehicle motion. Experimental results using both sensors and cameras are presented. Average error between vehicle location estimates from the cameras and the wireless sensors is around 0.5ft.

A wireless sensor network consists of small, limited-resource embedded systems exchanging environment data and activating controls. These networks can be deployed in hostile environments to monitor wildlife habitats, implemented in factories to locate mobile equipment, and installed in home environments to optimize the use of utilities. Each of these scenarios requires network security to protect the network data. The IEEE 802.15.4 standard is designed for WSN communication, yet the standard states that it is not responsible for defining the initialization, distribution, updating, or management of network public keys. Individuals seeking to research security topics will find that there are many 802.15.4-compliant development hardware kits available to purchase. However, these kits are not easily compared to each other without first-hand experience. Further, not all available kits are suitable for research in WSN security. This thesis evaluates a broad spectrum of 802.15.4 development kits for security studies. Three promising kits are examined in detail: Crossbow MICAz, Freescale MC1321x, and the Sun SPOT. These kits are evaluated based on their hardware, software, development environment, additional libraries, additional tools, and cost. Recommendations are made to security researchers advising which kits to use depending on their design needs and priorities. Suggestions are made to each company on how to further improve their kits for security research.