Title :
Ambiguity in radar and sonar
Author :
Rendas, M. João D ; Moura, José M F
Author_Institution :
CNRS Lab., Univ. de Nice-Sophia Antipolis, Valbonne, France
fDate :
2/1/1998 12:00:00 AM
Abstract :
We introduce a new ambiguity function for general parameter estimation problems in curved exponential families. We focus the presentation on passive and active radar and sonar location mechanisms. The new definition is based on the Kullback (1978) directed divergence and reflects intrinsic properties of the model. It is independent of any specific algorithms used in the processing of the signals. For the active single target problem, we show that our definition is equivalent to Woodward´s (1953) radar narrow-band ambiguity function. However, the new ambiguity is much broader, handling radar/sonar problems when there are unknown parameters (e.g., unknown power level in active systems), when the signals are random (e.g., passive systems), when the signals are wideband, or when there are model mismatches. We illustrate the new ambiguity in localization problems in multipath channels
Keywords :
maximum likelihood estimation; multipath channels; radar detection; radar signal processing; random processes; sonar signal processing; Kullback directed divergence; active radar; active single target problem; ambiguity function; curved exponential families; localization problems; maximum likelihood estimation; model mismatches; multipath channels; parameter estimation problems; passive radar; radar location; radar narrow-band ambiguity function; random signals; signal processing; sonar location; unknown parameters; wideband signals; Closed-form solution; Computer errors; Multipath channels; Narrowband; Parameter estimation; Passive radar; Power system modeling; Signal processing; Sonar measurements; Stochastic processes;
Journal_Title :
Signal Processing, IEEE Transactions on