In this paper, we propose and investigate location fingerprinting with a low complexity generalized ultrawideband (UWB) energy detection receiver. The energy samples at the output of the analog receiver front-end serve as location fingerprints. We formulate the position location problem as hypothesis testing problem and develop a Bayesian framework treating the location fingerprints as random vectors. In order to obtain an accurate stochastic description of the energy samples, which is required by the Bayesian framework, we provide two approaches. First, we derive a numerical algorithm to calculate the exact probability density functions of the energy samples, in case the UWB channel follows a Gaussian process. These results are used for benchmarking and performance prediction. Second, we propose closed form probability density functions based on a model selection criterion and measured energy samples. We show the accuracy and applicability of these closed form probability density functions in terms of performance results of the position location algorithm. The performance of the proposed location fingerprinting algorithm is evaluated based on measured UWB channels. The impact of important system parameters on the performance is investigated as well.