This work is part of a larger study on modeling and spatial perception in the context of auditory augmented reality. In this context, where the real environment is enhanced by a set of virtual sound events, an important issue is to ensure the coherence of the spatial perception of these virtual sound events and the real sound sources, or visual anchors, present in the room. Ensuring this coherence requires, in particular, the choice of a model to render the sound distance effect and access to information on the acoustic properties of the room. 

In a first step the objective was to evaluate the performance of a model, informed by a single directional impulse response (DRIR) measured in the real environment, to control the apparent distance from a virtual source.
Within this model, the sound distance effect is rendered by the modificationof the energy of different time segments of the reference DRIR. The extrapolated impulse responses are then used in a convolution rendering engine to place the virtual source at different distances from the listener.

The performance of the model was evaluated using a perceptual test protocol in which the participant was placed in an audio only augmented
reality situation in which he or she had to judge the distance of a virtual sound source. Visual anchors (chairs) were also present to help scaling the judgment. Model performance was conducted by comparison with reference binaural impulse responses measured in the test room for each specific distance tested. Statistical analysis of the results, collected from 40 participants over two sessions, demonstrates the perceptually
equivalent performance of this model compared to the reference measurements.
The criteria judged in the analysis are the average perceived distance, inter-subject and intra-subject variability.