The IF signal is a high-frequency alternating current (AC) signal. The first step in detection is usually to extract the "envelope" of this signal, which corresponds to the amplitude modulation of the pulse. This is commonly done using a diode detector. A diode allows current to flow in only one direction, effectively rectifying the AC signal. A capacitor often accompanies the diode to smooth out the rectified signal, producing a varying DC voltage that mirrors the shape of the radar pulse. This output is often called the "video signal."

Mathematical concept: If the IF signal is VIF (t)=A(t)cos(2πfIF t+ϕ), the detector aims to recover A(t), the envelope of the signal, which contains the target information.

The output of the detector (the video signal) is often still quite weak. A video amplifier boosts this signal to a level suitable for display or for input into an Analog-to-Digital Converter (ADC) for further digital processing. The video amplifier's bandwidth is chosen to accommodate the shape of the radar pulse.

The amplified video signal contains not only target echoes but also noise. To decide whether a detected pulse is a genuine target echo or just a random noise fluctuation, the signal is compared against a pre-defined detection threshold.

If the video signal amplitude exceeds the threshold, it is declared a "detection" or "target."
If it falls below the threshold, it is considered noise and is discarded.
The choice of this threshold is critical.
Setting it too low will lead to a high Probability of False Alarm (Pfa), where noise spikes are incorrectly identified as targets (nuisance alarms).
Setting it too high will lead to a low Probability of Detection (Pd), where weak but legitimate target echoes are missed.
The optimal threshold is usually determined based on the desired Pfa and the statistical properties of the noise (often assumed to be Gaussian noise).

Once the presence of an echo is detected, further signal processing is typically applied in modern digital radar systems:

• Analog-to-Digital Conversion (ADC): The analog video signal is converted into a digital stream of data, allowing for complex numerical processing.
• Pulse Integration: Multiple echoes from the same target are summed coherently or non-coherently. This technique significantly improves the Signal-to-Noise Ratio (SNR) and thus increases the detection range or reduces false alarm rates.
• Clutter Rejection Techniques: In environments with stationary objects (like ground, buildings, or weather phenomena), the strong echoes from these "clutter" sources can obscure weaker target echoes. Techniques like Moving Target Indication (MTI) or Pulse-Doppler processing (which analyzes the Doppler shift of each echo) are employed to differentiate moving targets from stationary clutter. These techniques typically involve filtering out signals with zero or near-zero Doppler shift.
• Target Extraction and Tracking: Once detections are made, algorithms are used to group detections into actual targets, estimate their parameters (range, velocity, azimuth, elevation), and track their trajectories over time. This data is then presented to the operator or used by automated systems.