A real-time siren detector to improve safety of guide in traffic environment

A system based on a modified pitch detection method is proposed that can be used for the detection of acoustical signals the frequency components of which vary according to specific periodic patterns. Usually, signals of this category are produced by the siren of an emergency vehicle. The detection of this type of signals can improve the safety of guide for hearing impaired people by alerting the driver with a visual indication for example by putting on a light point on the map of a navigator device. This paper discusses the development of a small, inexpensive Atmel microcontroller based siren detection system that is capable of discriminate the siren signal respect to other sound sources in the traffic. The siren detector is tested on real signals acquired and recorded in city streets with many traffic, different other sound sources (automobile horns, antitheft alarms, motor noisy etc.) and the presence of a siren approaching. The performance of the detector in terms of false alarm rate and missing of the siren signal is analyzed varying different variable parameters.