AUTOMATIC DECISION THRESHOLD LEVEL CONTROL IN DIRECT-SEQUENCE SPREAD-SPECTRUM SYSTEMS

An algorithm for automatic decision threshold-level control in direct-sequence spread-spectrum systems is presented and analyzed. Using this algorithm instantaneous setting of the threshold is possible. An additional loop called threshold loop (Th-loop) is used to improve the system performance. This loop is based on a so-called constant false alarm rate (CFAR) criterion. The purpose of the algorithm is to set the decision threshold in the system which will provide, at the same time, small probability of false alarm (P(a) --> 0) and large probability of signal detection (P(b) --> 1). The analysis has shown that the threshold probability distribution function (pdf) is narrower (better performance) in the case of a CFAR algorithm but comes at the cost of a considerably larger integration time of the system (the system time response is slower). As a compromise the instantaneous algorithm is modified so that k successive samples of the signal are used to determine the value of the threshold. This enables better control of the shape and position of the threshold pdf with respect to the P(a) and P(d) curves. At the same time the time response of the system is good and can be easily controlled.