This report describes several experiments used to characterize and test a network of radiation sensors. The purpose of these tests is to assess the feasibility of using these sensors to detect and track radioactive sources in a large field, as in a battlefield or on a military campus. Simulated radiation measurements are used to compare the result of radiation detection accuracy in tracking the moving target and to find its path as early as possible. This is done via changing the number of sensing nodes deployed (deployment density), as well as the models of the detectors. This thesis describes algorithms for both detecting the presence and tracking the position of radioactive sources. It formulates the detection problem as a nonparametric hypothesis-testing problem that is solved by comparing a statistic computed over some window of observation of the data to a threshold value. If this threshold is exceeded then it is decided that a source is present. The tracking results thus found are compared with the actual chosen path within the implemented experiment. Detection delay has been measured while trading off battery consumption and accuracy.