Radar

How Do Radars Work?

Radar is a radio frequency (RF) sensor that detects objects within its field of view and determines the object’s distance, velocity, and direction in azimuth and elevation angles. Radars are also used to determine weather conditions, to generate an image of an object, or to identify the object type. Radar Toolbox in MATLAB® helps with the analysis, modeling, and simulation across the entire radar lifecycle, from system engineering through signal processing and data processing.

Figure 1: Design, simulate, and test multifunction radar systems with Radar Toolbox in MATLAB.

Figure 1: Design, simulate, and test multifunction radar systems with Radar Toolbox in MATLAB.

What Is the Radar Equation?

Radar system performance is often defined either by the detection range for a specific probability of detection (Pd) or the Pd for a specific range. The radar equation is a function that predicts radar system performance based on system design parameters, the objects in the radar’s field of view, and the environment.

Link-budget analysis, which determines the received power at the receiver, is also based on the radar equation. This type of analysis is conventionally done using spreadsheets. Alternatively, the Radar Designer app is an interactive tool that allows a user to perform link-budget analysis using visualizations. This app helps users understand the design trade-offs and design a radar system based on a set of performance requirements.

Figure 2: Perform link-budget analysis with visualizations using the Radar Designer app in MATLAB.

Figure 2: Perform link-budget analysis with visualizations using the Radar Designer app in MATLAB.

How Do Radars Work?

Radar captures and analyzes the reflections and RF emissions of objects or the environment within its field of view. Radar either uses its own transmitter or, in the case of passive radars, another available source of RF energy to illuminate the surveillance area. This energy propagates in the environment and scatters in different directions upon encountering an object or a change in the environment. The signal received at the radar includes other signals from the environment, such as noise, interference from other sources of RF energy, and clutter. The detections are generated using signal processing techniques. By applying data processing algorithms on the detections, the radar can generate object tracks, images, and perform object classification.

What Are Different Radar Types?

Radar systems can be implemented in a range of system-level architectures that vary based on the type of application. Some examples include:

Monostatic and Bistatic/Multistatic Radar

Monostatic radar is the most typical radar, characterized by collocated transmitting and receiving antennas. In a bistatic/multistatic radar, the transmitter and receiver antenna arrays are separated by a distance comparable to the estimated range of the object. This type of radar has applications where the energy reflected back to the transmitting antenna direction is very small and not detectable. You can use radarDataGenerator, a statistical radar model in Radar Toolbox, to generate probabilistic detections in both monostatic and bistatic modes.

Figure 3: Detection modes available in radarDataGenerator, a statistical radar model in Radar Toolbox.

Figure 3: Detection modes available in radarDataGenerator, a statistical radar model in Radar Toolbox.

Passive Radar

A special case of bistatic radar, passive radar is characterized by the use of available commercial broadcast or communication signals as a source of energy, instead of an integrated transmitter in the radar system. The signal received by the radar system is processed and analyzed for object detection and tracking. This type of radar is not easily detectable because it produces no emissions. It is also resilient to interference and jamming. With the Radar Toolbox, you can simulate a passive radar to locate objects by leveraging existing transmitters.

Figure 4: Propagation of emissions from the emitter to the passive radar sensor.

Figure 4: Propagation of emissions from the emitter to the passive radar sensor.

Polarimetric Radar

This type of radar system uses both vertical and horizontal polarizations in the transmitting and/or receiving chain for object detection and classification. Polarimetric radar data can be used to improve object detection, or to classify precipitation and atmospheric conditions for weather forecasting. Modeling a polarimetric radar requires simulation and analysis of both polarizations. With Radar Toolbox, you can analyze the performance of a polarimetric radar for detection of targets with orthogonal polarizations.

Imaging Radar

This type of radar system uses a wide bandwidth signal to generate a high-resolution image of an object. The required wide field of view can be achieved by phased array antennas and beamforming techniques or by synthetic aperture radar (SAR), which captures the reflected energy while moving across the survey area. To achieve high resolution in both the range and cross-range dimensions, the range resolution is improved with pulse compression techniques, and the cross-range resolution is enhanced with cross-range compression methods, including range migration and back-projection.

Where Are Radars Found?

Radar system designs can be tailored for different functionalities and applications across industries. Typical radar applications include:

Air Traffic Control

Airport Surveillance Radar (ASR) is used as part of the air traffic control system in airports for detection and localization of aircraft in the vicinity of the airport. Assisting airplanes for final approach in bad weather is another application of radar systems in air traffic control.

Figure 5: Automatically detect deviations and anomalies in aircraft making final approaches to an airport runway using Sensor Fusion and Tracking Toolbox (see example).

Automotive

Radars mounted on vehicles are used to detect obstacles and other vehicles, their locations and speed. There are two classes of radar used in automotive applications: short-range radar (SRR) used for blind-spot monitoring and parking assistance and long-range radar (LRR) with applications including adaptive cruise control, collision avoidance, and blind spot detection. With Radar Toolbox you can generate probabilistic radar detections, clusters and tracks that include multipath effects.

Figure 6: Highway vehicle tracking with multipath radar reflections in MATLAB (see example).

Weather

Weather radars capture and analyze reflections from the atmosphere to detect, measure, and locate precipitation. Doppler radars and polarimetric radars are the two common radar technologies used for weather forecasting. Doppler radar is used to determine the direction and speed of precipitation movement. Polarimetric radar is used to detect the precipitation type or to measure the turbulence. With Radar Toolbox you can generate in-phase and quadrature (IQ) signals for a weather radar and measure parameters such as shear or turbulence.

Figure 7: Comparison of measured and simulated spectrum width to assess velocity dispersion (shear or turbulence) with Radar Toolbox (see example).

Figure 7: Comparison of measured and simulated spectrum width to assess velocity dispersion (shear or turbulence) with Radar Toolbox (see example).

Remote Sensing

Imaging radars are utilized in remote sensing applications to provide environmental and geospatial information. For example, satellite synthetic aperture radars generate images of Earth’s surface to provide information including water levels, forest heights, and habitat changes. Ground penetrating radar (GPR) is another imaging radar, based on synthetic aperture radar, that is used to generate subsurface images for applications including geology, archaeology, utility mapping, and concrete inspection. With radarTransceiver in the Radar Toolbox, you can generate IQ signals that are useful in developing signal processing and imaging techniques for remote sensing applications.

Aerospace and Defense

The main application of radar systems in aerospace and defense is searching for and tracking objects at long ranges and within a wide bearing coverage. Based on the search environment, these radars are often further categorized as air surveillance, maritime surveillance, or ground surveillance radars. Multifunction radars that can perform several tasks, such as search, confirm, and track, are used for these applications. With Radar Toolbox you can model multifunction radar systems with flexibility in design parameters such as frequency and pulse repetition frequency (PRF).

Radar with MATLAB and Simulink

With MATLAB and Simulink, you can design, analyze, and model different radar types at different abstraction levels, from statistical to physics-based modeling. Radar Toolbox provides a library of apps, functions, and algorithms that enable you to:

  • Improve the fidelity of your link-budget analysis by including the system component and environment loss in Radar Designer app
  • Model realistic scenes and scenarios including environment clutter and multiple-platform motion
Figure 8: Radar performance analysis over terrain with Radar Toolbox and Mapping Toolbox (see example).

Figure 8: Radar performance analysis over terrain with Radar Toolbox and Mapping Toolbox (see example).

  • Simulate ghost target detections and tracks due to multipath reflections with both statistical models and physics-based models

Figure 9: Simulate radar ghosts due to multipath return with statistical and waveform-level models in MATLAB (see example).

  • Perform closed-loop radar simulation for multifunction radar systems and improve the resource management efficiency by modeling adaptive tracking techniques

Figure 10: Adaptive tracking of maneuvering targets with managed radar (see example).

  • Generate synthetic aperture radar images to train deep learning algorithms for target recognition
Figure 11: Automatic target recognition (ATR) in SAR images (see example).

Figure 11: Automatic target recognition (ATR) in SAR images (see example).

  • Synthesize radar data for deep learning models to classify objects and received signals

Figure 12: Pedestrian and bicyclist classification using deep learning (see example).

See also: Phased Array System Toolbox, radar solutions for aerospace and defense, electronically steered array (ESA) radar, Antenna Toolbox, RF Toolbox, RF Blockset, Signal Processing Toolbox, Sensor Fusion and Tracking Toolbox, beamforming, Mapping Toolbox, MATLAB Coder, Embedded Coder, HDL Coder, Fixed-Point Designer