Radar Systems

Introduction

Radar (Radio Detection and Ranging) is one of the most sophisticated applications of signal processing, combining everything from matched filtering and correlation to Fourier analysis and optimal detection theory. When an air traffic controller tracks aircraft, when your car's adaptive cruise control measures the distance to the vehicle ahead, or when a weather station maps approaching storms — radar signal processing is making it all possible.

For B.Tech students studying signals and systems, radar provides a beautiful real-world example of how matched filters maximize signal-to-noise ratio, how the Fourier transform reveals velocity information, and how correlation functions measure time delays. The mathematics you learn in this course literally enables us to "see" objects hundreds of kilometers away using electromagnetic waves.

Basic Radar Principle

A radar system transmits an electromagnetic pulse and listens for the echo reflected from targets. The fundamental measurement is time delay:

$$R = \frac{c \cdot \tau}{2}$$

where $R$ is the range to the target, $c = 3 \times 10^8$ m/s is the speed of light, $\tau$ is the round-trip time delay, and the factor of 2 accounts for the two-way path.

The received signal is a time-delayed, attenuated, and possibly Doppler-shifted copy of the transmitted signal, plus noise:

$$r(t) = \alpha \cdot s(t - \tau) \cdot e^{j2\pi f_d t} + n(t)$$

where $\alpha$ is the attenuation factor, $\tau$ is the time delay (related to range), $f_d$ is the Doppler shift (related to velocity), and $n(t)$ is noise.

The Radar Range Equation

The received signal power decreases with the fourth power of distance:

$$P_r = \frac{P_t G_t G_r \lambda^2 \sigma}{(4\pi)^3 R^4}$$

where $P_t$ is transmitted power, $G_t$ and $G_r$ are antenna gains, $\lambda$ is wavelength, $\sigma$ is the target's radar cross section, and $R$ is range. The $R^4$ dependence means that doubling the detection range requires 16 times more power — this is why signal processing is crucial for maximizing radar performance without unlimited power.

Matched Filtering: The Optimal Detector

The core signal processing operation in radar is the matched filter. Given a transmitted pulse $s(t)$, the optimal receiver filter (maximizing output SNR) has impulse response:

$$h(t) = s^*(T - t)$$

This is the time-reversed, conjugated version of the transmitted signal. The matched filter output is the cross-correlation between the received signal and a reference copy of the transmitted pulse:

$$y(t) = \int_{-\infty}^{\infty} r(\tau) s^*(\tau - t) d\tau$$

The maximum output SNR achievable is:

$$\text{SNR}_{out} = \frac{2E}{N_0}$$

where $E$ is the pulse energy and $N_0$ is the noise power spectral density. This remarkable result shows that the optimal detection depends only on total energy, not on the particular pulse shape.

Pulse Compression

The Range-Resolution Dilemma

For a simple rectangular pulse of duration $T$, the range resolution is:

$$\Delta R = \frac{cT}{2}$$

Good resolution requires short pulses, but short pulses have low energy. We need high energy (long pulses) for detection AND short effective pulses for resolution. Pulse compression solves this dilemma.

Linear Frequency Modulation (Chirp)

The most common pulse compression waveform is the chirp — a pulse whose frequency sweeps linearly:

$$s(t) = \text{rect}(t/T) \cdot e^{j\pi \mu t^2}$$

where $\mu = B/T$ is the chirp rate, $B$ is the bandwidth swept, and $T$ is the pulse duration. When this waveform passes through a matched filter, the output is a compressed pulse with:

$$\Delta R_{compressed} = \frac{c}{2B}$$

The compression ratio (time-bandwidth product) is:

$$\text{PCR} = BT$$

This means a 10 µs pulse sweeping 10 MHz bandwidth gives compression ratio 100 — achieving the resolution of a 0.1 µs pulse while maintaining the energy of the full 10 µs pulse. The SNR gain is also $BT$ (20 dB for this example).

Doppler Processing

The Doppler Effect

A target moving with radial velocity $v_r$ causes a frequency shift in the reflected signal:

$$f_d = \frac{2v_r}{\lambda} = \frac{2v_r f_c}{c}$$

For a 10 GHz radar and a target moving at 300 m/s (fast aircraft), the Doppler shift is $f_d = 20$ kHz — easily measurable with spectral analysis.

Pulse-Doppler Processing

In a pulsed radar transmitting at pulse repetition frequency (PRF), the Doppler shift is extracted by taking the FFT across multiple received pulses at the same range bin:

$$X[k] = \sum_{n=0}^{N-1} x_n \cdot e^{-j2\pi kn/N}$$

where $x_n$ is the complex sample from the $n$-th pulse at a given range. The result is a Doppler spectrum where target velocity appears as a spectral peak.

Velocity resolution is:

$$\Delta v = \frac{\lambda}{2NT_{PRI}} = \frac{\lambda \cdot \text{PRF}}{2N}$$

Range-Doppler Map

The 2D range-Doppler map is the fundamental radar output for moving targets:

1. Fast-time processing: Matched filtering (pulse compression) along each pulse — resolves range
2. Slow-time processing: FFT across pulses at each range bin — resolves Doppler/velocity

The result is a 2D matrix where each cell represents a range-velocity bin. Targets appear as peaks in this map. Constant False Alarm Rate (CFAR) detection algorithms then set adaptive thresholds to detect targets while controlling the false alarm rate.

Clutter and Interference

Ground clutter, sea clutter, and weather returns are unwanted echoes. Moving Target Indication (MTI) uses the Doppler difference between moving targets and stationary clutter to separate them:

$$y[n] = x[n] - x[n-1]$$

This simple first-difference filter is a high-pass filter in Doppler, rejecting zero-Doppler clutter while passing moving targets.

Applications of Radar Signal Processing

• Air traffic control: Tracking hundreds of aircraft simultaneously with range, velocity, and bearing
• Automotive radar: Adaptive cruise control, collision avoidance (77 GHz FMCW radar)
• Weather radar: Mapping precipitation intensity and wind velocity using Doppler processing
• Synthetic Aperture Radar (SAR): Creating high-resolution ground images from aircraft/satellite platforms
• Ground Penetrating Radar: Subsurface imaging for construction and archaeology

Key Takeaways

• Radar measures range from time delay ($R = c\tau/2$) and velocity from Doppler shift ($f_d = 2v_r/\lambda$)
• The matched filter maximizes detection SNR — its output is the correlation between received and transmitted signals
• Pulse compression (chirp waveforms) achieves fine range resolution without sacrificing detection energy: $\Delta R = c/(2B)$
• The range-Doppler map combines matched filtering (fast-time) and FFT (slow-time) for simultaneous range-velocity estimation
• Radar SNR decreases with $R^4$, making signal processing gain essential for practical detection ranges