A NEW LIDAR-RADAR HYBRID CAN HELP PLATFORMS USE RADAR DETECTION UNDERWATER.
By Dr. Linda Mullen and Dr. Brandon Cochenour
One of the components needed to achieve and maintain assured access to the maritime battlespace is sensor superiority. Undersea threats must be detected, classified, and identified with high accuracy and low false alarm rates so that threats can be targeted. The detection process involves observing a feature that is uniquely relative to the surrounding environment and is consistent with the objects being sought (e.g., a round or large cylindrical object). The next step, classification, happens when operators categorize objects within a group of similar objects (e.g., a mine-like object or a submarine). Identification of the threat requires that object features are resolved accurately and quickly to determine with certainty what the objects are (e.g., a specific type of mine or a particular class of submarine) so the information can be communicated to those who can eradicate the threat.
Radio frequencies, while ubiquitous on land, experience high attenuation in water and therefore cannot be used for wireless communications or detection, classification, or identification undersea. It is for this reason that acoustic-based sensors and modems have historically been and continue to be used for these tasks. Acoustic technologies, however, lack the resolution typically needed for the identification step and have insufficient bandwidth for high-speed wireless communications. Acoustic frequencies also cannot penetrate the air-sea interface.
Lidar Sensors in Water
Lidar, or light detection and ranging, is the laser-based equivalent of radar and sonar. The highly directional properties of the laser output provide lidar systems with the resolution to accomplish the identification task. Lasers also have an inherently high bandwidth that enables them to be used for high-speed wireless communications. Furthermore, the fact that we can see objects in the water from both above and below the air-sea interface provides evidence that light can propagate through the water surface and within the water column. We know from viewing underwater photography, however, that light does not propagate through water the same way as it does in air. As light travels through water, it is absorbed and scattered by water constituents. The higher absorption of certain wavelengths or colors of light leads to the blue-green hue of underwater imagery, while scattering of light in water causes the haze or blurring of details.
Despite the challenges of light propagation in water, lidar sensors can adapt to the underwater environment. Lasers operating in the blue-green portion of the spectrum can be selected to minimize absorption and maximize transmission in water. The scattering problem is more difficult to overcome as light can scatter back to the receiver without ever reaching the object of interest (backscatter) and scatter multiple times at small angles on its path to and from the area of illumination (forward scatter). Backscatter tends to decrease the overall contrast of the collected imagery, while the collection of forward-scattered light causes image blurring and loss of spatial resolution or sharpness of the image. Similar to driving on a foggy night, turning up the laser power (like turning on the high beams) does not improve visibility in murky water since more light will only scatter back from particles in the water. Increasing the separation between the laser and receiver can help suppress backscatter, just as the fog lights that are further away from our line of sight on a car can enhance visibility in fog. The highly directional properties of laser light can be leveraged to reduce scattered light by limiting the receiver aperture and acceptance angle to view only the laser-illuminated spot some distance away. Furthermore, sensors using a pulsed laser source can reduce backscatter by timing the receiver to open at a time corresponding to the round-trip time to the object of interest.
Researchers at the Naval Air Warfare Center Aircraft Division (NAWCAD) in Patuxent River, Maryland, are investigating an alternate approach to enhance optical imaging in water. This hybrid technique (described in patent “Modulator Lidar System,” No. 5,822,047, 13 October 1998) uses a laser to transport a radar signal through the water. By encoding the laser pulse with a radar signal, the receiver can “lock on” to a signal reflection from an object and distinguish it from light scattered randomly from the environment, analogous to how our eye locks onto the strobe light on a school bus on dark, foggy mornings. Using a laser to carry a radar signal through the water provides a way to use the sophisticated radar modulation, demodulation, and signal processing techniques developed for above-water object detection and identification for similar applications in water, an environment where radar signals cannot be used directly because of their high absorption. The encoded waveform also can be altered to include information to be transmitted to another location, which would enable the sensor to be used for both object detection/imaging and wireless optical communications.
The group at NAWCAD has leveraged Office of Naval Research and in-house funding to focus its research in three main areas: environmental characterization (measuring the water optical properties using in-situ instruments and use data collected by the laser system to enable “through the sensor” environmental measurements); performance prediction modeling (using the information collected in the first area as inputs to theoretical models developed both in-house and through collaborations with academia and industry to predict the effectiveness of the approach for different applications and system parameters); and experimental measurements (designing and developing breadboard prototypes to validate model predictions and demonstrate system performance in a controlled laboratory environment). Comparisons between model predictions and experimental measurements are used to provide feedback to the environmental characterization task to identify whether new and/or improved measurements are required to improve the correlation between theory and experiment. Similarly, there is feedback between modeling and experiments to determine the accuracy of the underlying theory and/or to explain the physics involved with new or unanticipated results.
Characterizing the Underwater Environment
Current research has focused on improving measurement through the inherent optical properties of water: scattering and absorption. The NAWCAD group works closely with academia, industry, and other government laboratories to enhance the accuracy of the data provided by state-of-the-art, in-situ instruments. Alternatively, the group has developed custom lidar systems that can extract water optical properties from the detected signal. This environmental characterization uses variations in the system parameters (e.g., receiver acceptance angle, laser/ receiver polarization) to enhance the sensitivity of the sensor to specific water optical properties. For hybrid lidar-radar applications, the group is particularly interested in how the water optical properties influence the propagation of radar-encoded optical signals. Measurements using modulated laser beams have shown that the encoded radar signal is sensitive to small changes in the scattering phase function, which describes the angular distribution of light scattering in water and has traditionally been a very difficult parameter to measure in-situ.
The goal of this research involving theoretical model development is to create a time-dependent model that can predict the effect of water optical properties (absorption, scattering), system parameters (transmitter beam divergence, receiver aperture and acceptance angle, transmitter/receiver separation), and object characteristics (size, shape, reflectivity, depth) on the propagation of an impulse of light through water. Once this optical impulse response is computed, it can be combined with any type of radar or communications waveform and processed accordingly. Monte Carlo methods (i.e., random sampling) fall into the numerical category of underwater models as they trace the path of individual photons through a medium according to the inherent optical properties of absorption and scattering. While computationally intensive, the Monte Carlo method provides an exact solution since it tracks individual photon paths. Numerical models may require significant processing time, however, particularly in turbid environments, to simulate enough photons to be statistically accurate.
The other category of underwater propagation models uses analytical methods, which are based on the solution of the radiative transfer equation, a complicated integro-differential equation of several variables in space and time. Certain approximations are typically made to reduce the problem to provide a manageable solution. Current research efforts are focused on studying how these approximations affect the accuracy of predicting the effect of the water on the radar-encoded signal.
The main challenge in performing experimental measurements with radar-encoded optical signals in water is the hardware required to generate and detect these high-speed signals. On the transmitter side, a high-power, blue-green laser source with high-speed, efficient modulation is required. A wide-bandwidth, high-sensitivity optical detector is needed on the receiver end to recover the radar-encoded signal. Fortunately, the group has leveraged the Small Business Innovation Research (SBIR) program to fund industry collaborators in developing the necessary hardware for breadboard prototypes. Both SA Photonics, Inc. (Los Gatos, California), and Fibertek, Inc. (Herndon, Virginia), delivered blue-green, modulated pulse laser sources through a SBIR Phase II program. These sources produce optical pulses whose radar modulation can be easily controlled via software commands, which provides a way to test the effect of different radar waveforms on system performance.
Through the same SBIR topic, AdvR, Inc. (Bozeman, Montana), is developing a device that can impose the radar modulation on a commercially available pulsed laser. This approach offers an alternative modulated pulse source that does not require the development of a complete custom laser. AdvR also has produced a continuous-wave, modulated blue-green source that has been used for both imaging and communications applications. These hardware developments, combined with the advancements in high-speed digitizers and field programmable gate arrays, has made it possible to generate, detect, and process sophisticated radar modulation waveforms. The water tank at NAWCAD is 25 feet in diameter and 10 feet deep, and is outfitted with windows that provide a convenient way to transmit and receive light through water from a benign, dry environment. Various targets of interest can be easily mounted from an overhead bridge that spans the length of the tank, and the absorption and scattering properties of real-world water types can be reproduced in the lab through the addition of artificial scattering and absorbing agents and monitoring by in-situ optical instruments.
Putting It All Together
Recent imaging experiments have focused on the use of wideband “chirp” radar modulation schemes and subsequent pulse compression processing at the receiver. The technique uses a modulation waveform whose frequency is swept—or chirped—as a function of time. By transmitting a unique modulation signature on the optical signal, a receiver that knows the transmitted waveform can use “pattern-matching” techniques to look for its own unique signature being echoed back from targets. This process of pattern matching a chirp waveform is referred to as “pulse compression,” where a longer pulse encoded with a wide-bandwidth waveform is compressed into a short pulse at the receiver. Thus, the chirp modulation and subsequent matched filter processing provides a way to obtain a high time (or range) resolution measurement by using a wider transmitted pulse. Furthermore, when applied to the underwater sensing, the frequency content can be tuned to optimize the rejection of unwanted scattered light, which enables the system to adapt to different water environments.
Recent experiments were conducted in the water tank at Patuxent River to test the chirp modulation/pulse compression technique against realistic targets and in different underwater environments. A plastic manta mine-like target was suspended in the water column and illuminated with the system. Two- and three-dimensional images were created in both clean water (no scattering agents added) and in murky, harbor-like conditions. The results show that the technique has the potential to provide the high-resolution imagery needed for object identification in challenging underwater environments.
The Navy is trending toward using compact, unmanned, autonomous platforms to improve access to strategic areas of interest without the risk involved with manned platforms. The size, weight, and power of current laser-based sensors, however, are not compatible with small, unmanned, and autonomous underwater vehicles because these existing systems incorporate transmitter and receiver hardware on the same platform. To improve the compatibility of laser-based sensors with unmanned aerial and subsea vehicles, the NAWCAD team developed a technique (described in patent “Extended Range Optical Imaging System for use in Turbid Media,” No. 8,373,862, 12 February 2013) where the transmitter and receiver are located on separate platforms. While unique to laser-based sensors, this bistatic geometry has been used extensively in both sonar and radar sensors. For a laser-based sensor operating in degraded visual environments such as murky water, the bistatic configuration enables the transmitter to optimize its distance from the object of interest so that the amount of light scattered on the path to the scene is minimized. The laser is encoded with information concerning the scan, such as scan rate or scan angles, and the receiver decodes and uses this information to reconstruct the underwater image in real time, expediting decision making by eliminating the need to wait for the illuminator to return to the operator before data can be downloaded and analyzed.
The strength of this approach is that the transmitter and receiver are entirely autonomous and are linked only by a wireless communication signal that is carried by the light scattered from the object and from the environment. Furthermore, this approach supports distributed sensing since a swarm of laser illuminators can be deployed to survey an area of interest. Mission time is reduced as a single receiver can immediately collect and process information from many illuminators. The multistatic architecture also offers multifunctionality since both high-resolution imaging and high-speed laser communications are available from the same sensor suite.
The hybrid lidar-radar approach enables the use of well-established radar modulation, demodulation, and signal processing techniques for optical sensing and communicating in a wide range of underwater environments. This hybrid approach provides a solution for generating high-quality imagery so that underwater threats can be identified, and the same hardware can be used to communicate the threats to those in danger. By using the same hardware for sensing and communicating, unique system configurations are possible that make laser-based sensors more compatible with small underwater platforms. Ongoing research in environmental characterization, theoretical modeling, and experimental validation will help close the loop between experiment and theory so that the performance of these hybrid systems can be accurately predicted for scenarios not easily represented in a controlled laboratory environment. This will be an important step in developing the next generation of sensors to achieve and maintain assured access to the maritime battlespace.
About the authors:
Drs. Mullen and Cochenourare researchers at the Naval Air Warfare Center Aircraft Division.