Acquisition of pseudonoise sequences in direct-sequence spread-spectrum systems

Various schemes for the acquisition of a pseudonoise (PN) sequence in a baseband direct sequence spread-spectrum (DS/SS) system are investigated. First, we consider acquisition schemes for receivers with chip synchronization, that is, the chips of the transmitter and receiver PN sequence generator are synchronized, and only the starting point (or phase of the PN sequence) is unknown. Acquisition schemes for such receivers can be divided into active acquisition schemes, in which the receiver uses the observations to directly deduce the phase of the received sequence, and serial search schemes, in which the receiver guesses the possible phases in some predetermined order and uses the observations only to check if a particular guess is correct.Several active acquisition schemes are analyzed, and tight bounds on the probability of error of parallel search schemes are obtained. A method to reduce the complexity of a parallel search is outlined. We also consider parallel schemes with hard limited (binary) observations and show that, as might be expected, the probability of error for such schemes is very high compared with that for parallel schemes with soft decisions.Different approximations for a PN sequence are considered in order to analyze the worst-case and the average-case performance of serial search schemes. A worst-case sequence is obtained that may be used to derive the worst-case performance of a certain class of serial search schemes. The average performance of serial search schemes is studied using sequential probability ratio tests (SPRTs), which are designed and analyzed using a zero sequence model and a random sequence model of a PN sequence. Simulation of the two tests on an actual PN sequence shows that the random sequence is an excellent approximation for a PN sequence and may be used to predict the performance of each of the SPRTs when applied to a PN sequence. In addition, significant performance degradation results if the test is based on a zero sequence model rather than on a random sequence model.Finally, parallel acquisition schemes for receivers without chip synchronization are considered. Four schemes for such receivers are obtained, and approximations to the probability of error of two of the schemes are derived. The exact error probabilities of each of the schemes are obtained via Monte Carlo simulations. The results show that the approximations to the error probabilities are reasonably close to the values obtained by simulation.U of I OnlyETDs are only available to UIUC Users without author permissio