الفهرس Only 14 pages are availabe for public view
 |  | from | 171 |  |  |
| | | from | 171 | | |
Abstract
Wireless Sensor Networks (WSNs) are known to be limited resources networks. This type of networks depends on limited energy components called sensor nodes with limited lifetime. The research is great towards WSNs since they can be used in many applications in many fields. There are applications having limitations and restrictions on the flow rate delivered to the sink (data collector or base station) as found in networks which deal with high data rate e.g. transmission of video signals. There are other ones having limitations in the receiving time and latency for data that the delay in collecting, processing and sending the data is sensitive as in real-time multimedia transmission e.g. video conference. So the trend always is to design an optimized WSN which can achieve good levels of flow rate and lifetime. Two optimization problems for a small application WSN were studied in this thesis (this WSN consists of a small number of sensor nodes with one sink which form a cell inside a gird of many cells which are employed for a certain application and the data collected from every cell is sent by every sink to a central base station). The first was maximization of flow rate in limited energy constrained WSN. The second was related to the maximization of lifetime for the same WSN with other constraints than in the first problem. Additional constraints were added to the two problems which describe the applying of TCP/IP for WSN and restriction for the energies of sensor nodes. The Genetic Algorithm (GA) and the Particle Swarm Optimization (PSO) were the used two optimization algorithms to perform the optimization processes. Good solutions and results were obtained for the two problems using the PSO and some improvements could be obtained using the GA in some cases. A combined problem was formulated from these problems to achieve the targets of maximization the flow rate and lifetime in the WSN and good results were obtained using the GA. Different optimization parameters were studied to show their effects on the different optimization problems. The distances between the sensor nodes and the topology of field where they are located affect greatly the results of the optimization processes. The network simulator software version 2.26 was used to construct and simulate this WSN and show the effect of parameters like velocity and the height of sensor nodes in the sensing area on the performance of the network. Also a grid of cells was simulated as an application for WSN where each cell collects data and sends the data to its sink. Then all data from every sink is sent and collected at a central base station. A suggestion for a control system, which can observe the operation and performance of sensor nodes in the WSN, was presented as a conclusion of the thesis.