Abstract: A method and apparatus for reducing peak-to-average power ratio in a paging transmitter broadcasting an independent sideband amplitude modulation protocol is disclosed. A digital signal, for example a digital voice signal, is provided to a control gain generator which generates a digital gain sequence. The digital gain sequence is such that during peak values in the digital signal the gain is low and during low values in the digital signal the gain is high. The digital gain sequence is multiplied with the digital signal to compress the dynamic range of the digital signal.

Abstract: A linear transmitter (101) using predistortion includes a modulator (103), a predistorter (107), a digital quadrature modulator (111), an upconverter (113), a power amplifier (115), and an antenna (117). In addition, the transmitter (101) has a feedback loop including a coupler (119), a downconverter (123), a digital quadrature demodulator (125), and a trainer (131). The digital data to be transmitted is provided into the modulator (103), which converts the digital data into in-phase and quadrature component signals. The in-phase and quadrature component signals are then provided to the predistorter (107), which "predistorts" the component signals prior to amplification. The digital quadrature modulator (111) converts the component signals into a single analog signal. The upconverter (113) upconverts this signal from the predistorter (107) into the desired frequency of transmission, which is provided to the power amplifier (115) and the antenna (117) for amplification and broadcast.

Abstract: A improved siren detector for detecting siren sounds which precess at known warble rates, such as yelp, wail, and high-low, within a selected frequency band. A transducer detects the siren sounds and produces a corresponding electrical output signal. This electrical output signal is filtered to reject signals outside of the selected siren frequency band. The signal is processed to determine the amplitude of the electrical signal, and hence the sound level of the siren sound at the transducer input. This signal is also processed by an amplitude limiter and frequency discriminator to determine the instantaneous frequency of the siren sound. This discriminator is followed by a non-linear filter to remove the FM clicks characteristic of siren sounds having a low signal to noise ratio. Selection filters are used to analyze the precession rates, maximum frequency, minimum frequency, and shape of the precession characteristic to classify the siren as to its type, such as yelp, wail, and high-low.