الفهرس | Only 14 pages are availabe for public view |
Abstract Using chaotic codes in communications is relatively a new field of research. These codes have good autocorrelation and cross correlation properties. Also, the linear complexity of chaotic codes is higher than the linear complexity of linear feedback shitt register (LFSR) codes, e.g. m-sequences and Gold codes. These properties attracted the researchers to study the applications of chaotic codes in code division multiple access (CDMA) systems. hi this thesis we aim to investigate and improve the performance of chaotic codes for CDMA systems. Focus is oriented towards three research problems. The first point is the quality assessment of chaotic pseudorandom number generators (PRNGs). The second point is the optimum generation of chaotic codes. The third point is the optimum selection of chaotic codes sets for asynchronous CDMA systems. The standard randomness tests were u.wd to assess the quality of (Jiflcrenl chaotic PRNGs. The results of these tests have shown thai chaotic code» have belief randomness properties than most of classical codes including LFSR codes frequently used in CDMA systems. Therefore, CDMA systems utilizing chaotic codes, or simply chaotic CDMA systems, are more secured than those utilizing LFSR codes. We proposed a new method for the optimum generation of chaotic codes. This method depends on selecting a fixed threshold value to convert chaotic continuous value sequence into binary code. In contrast with the previously proposed method, this threshold value is extracted from the used chaotic map statistics, and not from the statistics of the generated sequences. Computer simulations have shown that our method outperforms other previously proposed methods. Also, we proposed a new method for the optimum selection of chaotic code sets based on Genetic Algorithms (GA). In this method, the GA is used to search the initial conditions space in order to select a code set with iow MAL Computer simulations have shown that the proposed method outperforms the exhaustive search method. Also, computer simulations have shown that chaotic code sets outperform classical code seis. |