The spherical spreading in the wavefront is a geometrical factor the absorption is associated with the amplitude decay due to the energy loss related to heat conversion dispersion refers to the pulse deformation (usually a temporal stretch) related to the frequency dependence of the wave parameters. As the pulse propagates in the medium, it suffers attenuation and distortion as the result of spherical spreading, absorption and dispersion. GPR uses an electromagnetic (EM) pulse within the frequency range 10 MHz–2 GHz. The ground-penetrating radar (GPR) technique has been increasingly used for characterizing and monitoring the surface and subsurface of historical buildings (Colla and Maierhofer 2000, Leckebusch 2000, Leucci 2002, Leucci et al 2003, 2006, 2007, 2011). GPR, attribute analysis, monumental buildings, frequency maps 1. Two real case histories back up the proposed method. In this paper, we propose an approach that analyses the data in the form of ‘frequency maps’ to evidence absorption losses probably linked to higher moisture content. In particular, the spatial variation in GPR signal attenuation can provide important information about the electrical properties of the investigated materials that, in turn, can be used to assess the physical parameters associated with damage. However, the complexity of the geometry and the structural heterogeneity that characterize these old structures often make the GPR results difficult to analyse and interpret. In old monumental buildings, the masonry structures frequently exhibit cracks, voids, detachments and high moisture contrasts that can give rise to reflection events in radar signals. In such cases, a valuable contribution could be provided by geophysical methods (such as electrical resistivity, electromagnetic conductivity, ground-penetrating radar (GPR), etc), which have been proved to be successful tools for sub-surface investigation and characterization of historical buildings. In many other cases, the effects of decay processes are not visible, thus making difficult the diagnosis and the consequent setup of effective rehabilitation and preservation interventions, especially in the presence of a complex geometry and/or a large variability of construction materials. In many cases, efflorescence and moulds are visible on the facades of several monuments of historical importance. Syst Comp Jpn, 36(3): 42–52, 2005 Published online in Wiley InterScience ( DOI 10.1002/scj.The presence of particular microclimatic conditions inside monumental buildings is responsible for bio-deterioration processes. A computer experiment verifies that the proposed model works effectively in an environment which is strongly partially observable. The elements generate a secondary reinforcement signal which results in learning similar to the conventional reinforcement scheme based on the state evaluation function. The proposed network is a four-layered feedforward network in which the pulse neuron elements forming the two hidden layers provide a pseudo-representation of the state in the environment. The proposed model combines pulse neuron elements with different decay rates, which facilitates the processing of the time-series input information and the discrimination of fuzzy states in a partially observable Markov decision process. It is known in particular that a pulse neuron with a high decay rate acts as a coincidence detector. The application of such neural networks has been considered in recent engineering studies. But the pulse neuron is a modeling of the input–output relation of the time-series pulse (spike) and the decay of the internal state (internal potential). The conventional integrator neuron element is modeled in terms of the average firing rate of the biological neuron. This paper considers learning by a pulse neural network and proposes a new reinforcement learning algorithm focusing on the ability of pulse neuron elements to process time series.