To search, Click below search items.


All Published Papers Search Service


Sonar False Alarm Rate Suppression using Classification Methods Based on Interior Search Algorithm


Sajjad Ravakhah, Mohammad Khishe, Majid Aghababaie, Esmail Hashemzadeh


Vol. 17  No. 7  pp. 58-65


False alarm rate reduction is one of the challenging issues in the sonar systems. This paper uses classification technique to identify real targets from false alarms. For this purpose, Radial Basis Function Networks (RBFNs) are utilized. Taking into account the use of gradient descent and recursive methods in the classic RBFNs, low Classification accuracy, slow convergence rate, and getting stuck in local optima, are the main drawbacks of RBFNs. In order to overcome these shortcomings, this paper suggests the use of the newly proposed Interior Search Algorithm (ISA) for training the RBFN. In order to measure the performance, ISA is compared with five well-known benchmark algorithms named PSO, ACO, GA, DE, and BBO in terms of entrapment in local minima, classification rate, and convergence speed. The results show that ISA is significantly better than the other well-known benchmark meta-heuristic algorithms in identifying real targets from false alarms.


Sonar, Classification, Interior Search Algorithm, Radial Basis Function Networks.