Many of the signals encountered in practice behave randomly in part or as a whole in the sense that they cannot be explicitly described by deterministic mathematical functions such as a sinusoid or an exponential function. Randomness arises because of the random nature of the generation mechanism. Sometimes, limited understanding of the signal dynamics also necessitates the randomness assumption. In electrical engineering we encounter many signals that are random in nature. Some examples of random signals are:

1. Radar signal:   Signals are sent out and get reflected by targets. The reflected signals are received and used to locate the target and target distance from the receiver. The received signals are highly noisy and demand statistical techniques for processing.
                         
2. Sonar signal:   Sound signals are sent out and then the echoes generated by some targets are received back. The goal of processing the signal is to estimate the location of the target.
                                    
3. Speech signal:   A time-varying voltage waveform is produced by the speaker speaking over a microphone of a telephone. This signal can be modelled as a random signal.  A  sample of the  speech signal  is shown in Figure 1.
                                  
4. Biomedical signals:   Signals produced by biomedical measuring devices like ECG, EEG, etc., can display specific behaviour of vital organs like heart and brain. Statistical signal processing can predict changes in the waveform patterns of these signals to detect abnormality. A sample of ECG signal is shown in Figure 2.