This paper provides a tutorial introduction to the principles of radar for persons with experience in optics but with little or no knowledge of radar. A typical radar system is described and the radar range equation is developed. The detection of signals embedded in a noise environment, the detection of target signals in a clutter environment, and the basic principles of radar antenna theory are discussed.

An important aspect of radar theory is an understanding of the properties of radar waveforms and the demands placed on them by the environment and measurement requirements. The key to this understanding of radar waveforms lies in the ambiguity function which, as a mathematical tool, forms the basis for the results presented in this paper. A tutorial development of the ambiguity function and illustrative examples are presented. The examples include both typical waveforms such as phase codes as well as sophisticated amplitude weighted burst waveforms used for clutter rejection. Optimum waveform weighting techniques are derived and their use illustrated. A discussion of the effects of hardware errors on waveform performance and design is also included.

This tutorial consists of a review of the pertinent fundamentals of optics: concepts of incoherent, coherent, and partially coherent light; propagation of coherent light; diffraction and interference;Fraunhofer diffraction as a Fourier transformation; spatial frequency concepts; properties of image formation, amplitude impulse response, intensity impulse response, and the optical transfer function; simultaneous optical Fourier transformation and image formation; simple spatial filtering operations.

The class of operations achieved by conventional coherent and incoherent optical data processing systems has been somewhat limited. Recent attempts to extend coherent optical processors to space-variant and non-linear operations will be reviewed. In addition, some new thoughts on the possible achievement of complex-valued operations using incoherent light will be presented.

Real time and reusable coherent spatial light modulators for use in optical data processing are summarized. The status of the six leading candidate real time devices that have seen use in coherent light are included. Three electron beam addressed devices are described (the dielectric oil film valve, the thermoplastic optical phase recorder, and electron beam DKDP). Three optically addressed spatial light modulators are also included (the liquid crystal lightvalve, the PROM, and photo DKDP). The operation, specifications, applications, and remarks on directions for future research are included for all of these devices.

Through the supports of the ONR and NSF we have recently developed several schemes which utilize multiple surface acoustic waves for wideband efficient guided-wave acoustooptic (AO) Bragg diffraction. The development of these wideband schemes has made it possible to design and fabricate very wideband guided-wave (thin-film) AO devices, and has thus paved the way for the realization of wideband AO processors in thin-film form. In this paper a review of the more recent developments on wideband (up to 500 MHz bandwidth) guided-wave AO signal processors, emphasizing those being carried out at the authors' institution, is given. The processors to be discussed include spectrum analyzers, convolvers and pulse compressors. Typical performance characteristics that have been obtained are presented. The guided-wave processors, in comparison to their bulk-wave counterparts, possess inherent advantages of requiring less RF drive power, being smaller, lighter weight, less susceptible to environmental effects, and more integratable and thus potentially less costly. Consequently, the guided-wave versions, when fully developed, should complement the existing bulk-wave processors.

Potentially significant improvement in the effectiveness of weapon systems utilizing radar and optical multi-sensor concepts appears realizable. Since similar analytical foundations and techniques are used to describe, model, and analyze optical and radar systems, optical and radar practitioners can gain useful insight into the integration of the two technologies by considering analogies existing between the fields. A number of radar analogies for certain optical processes are presented.

MTI (Moving Target Indication) radar systems have been built for many years, based on sys-tem concepts evolved in the early 1950's. Digital techniques now permit easier implementation, but do not change the basic concepts; staggered repetition periods to eliminate blind speeds; and MTI cancellers with the velocity response shaped by feed forward and feedback techniques. Many pf the existing systems are very successful, even though their performance, measured in terms of MTI improvement factor or subclutter visibility, often falls far short of the performance predicted from linear theory. The basic MTI concepts and definitions are presented, and the real problems of modern surface-based MTI radar systems are discussed. Future signal-processing techniques are postulated.

This paper describes a phased array radar emphasizing differences from conventional radars and explores the operational advantages of these differences. Subjects include the general principal of beam formation by the phasing of RF emissions from a large number of radiating elements, the establishment of bandwidth and element spacing constraints and the relationship of the phase shifter and time delay units. Frequency scanned arrays are contrasted with phase steered arrays, and the architectural possibilities associated with corporate fed, space fed, and reflecting arrays are covered in comparative terms. The paper highlights the operational advantages of phased arrays from the standpoint of traffic handling capacity. The operational advantage of bandwidth, inertialess scanning and reliability are given consideration. System situations which justify phased arrays are delineated. Finally, the latest trend in phased arrays, that of total solid state architectures, is singled out as an example.

Laser radar systems are gaining acceptance where unique properties show significant advantage over conventional methods. This paper discusses some illustrative examples of laser radar systems in use today. Simple explanations of the unique properties of these examples are provided.

Sophisticated signal processing is used in synthetic-aperture radar (SAR) systems to yield radar imagery with fine spatial resolution. Coherent optics has played a critical role in SAR signal processing for two decades and continues at present as the standard by which other SAR signal processors are measured. The signal processing operations which are required in SAR systems include display of the signal spectrum, spatial filtering, pulse compression, and post-detection integration. Depending upon the configuration of the radar itself, these may be required as one-dimensional or two-dimensional operations. In an optical processor, they are com-bined with imaging steps to move the signal to various locations in the processor, at which points these operations are implemented. Laser-illuminated spheres, cylinders and combinations thereof perform the signal processing operations. The signal processing requirements of SAR system are reviewed in this paper, and the manner in which elementary coherent optical configurations can be employed to satisfy these requirements is illustrated.

The Acousto-Optic Spectrum Analyzer (AOSA) is probably the simplest optical processor with practical value. It performs a single dimensional real-time Fourier transform using only four active components, a light source, Bragg cell, transform lens, and a photodetector array. A description of the basic operating parameters of the AOSA when used for radar signal processing is presented by comparing a traditional ray trace with a process element diagram and an electrical equivalent circuit. Selected applications and off-the-air results are included.

The processing algorithm and coherent optical processor used for synthetic aperture radar data processing are described and examples of resultant radar imagery are presented.

Law probability of intercept (LPI) surveillance and communication systems can be of vital benefit to a number of military applications. The basic concept in all LPI systems is to spread the transmitted power over as wide a frequency band as possible to make detection as difficult as possible. The critical technology issue involved in implementing this concept is to achieve complex real time matched filter signal processing in the receiver with as large a processing gain as possible. This paper briefly discusses an experimental program, which combined a surface wave tapped delay line and optical coherent integrator to demonstrate processing gains on the order of 70 dB; and the evolving technology in optical signal processing, which needs to be developed to achieve feasible systems application.

The Army has maintained a low continuing profile of interest in optical data processing for radar systems. The field is finally reaching a degree of maturity, as evidenced by this conference. Is the time at hand to start seriously applying optics investigation to radar systems, or are we still in a pie-in-the-sky mode? I believe that most of the straightforward, easy applications have been preempted by the fantastic developments in digital circuitry; but, there are radar systems and possibilities which are hard to imagine without the power of an optical computing system. ODP has made great strides recently, but there are still gaps before we apply our concepts to fielded systems.

The development of laser ranging systems within NASA started in the early 1960's, soon after the invention of the laser. This program has grown substantially in the intervening years, and has produced important results in the areas of precision orbit determination and gravity field determination. In addition, laser ranging is expected in the near future to produce some unique results on earth crustal motions. These results may be very important for understanding the mechanisms which cause earthquakes. In this document we present an overview of the NASA laser ranging program. The dis-cussion is organized as outlined in Figure 1. The principal applications these systems address are outlined, and the characteristics of typical systems built during the 1960's and early 1970's are described. The current state-of-the-art as exemplified by some of the more recent ranging systems are also discussed with respect to performance levels and the error sources which limit this performance. The final section discusses the plans within NASA for using laser ranging systems aboard the Space Shuttle.

A two lens optical Fourier transformation is shown to be equivalent to an electronic chirp transform. For discrete time signals this transformation becomes the discrete chirp-z-transform. The chirp-z-transform can be implemented using either charge transfer devices or surface acoustic wave devices. Through the use of appropriate architectures, a long one-dimensional chirp-z-transform can be rewritten as a modular chirp-z transform using both charge transfer devices and surface acoustic wave devices. Thus in a manner similar to the means by which a two-dimensional lens system can process large one-dimensional signals, so a combination of electronic components configured as a modular chirp-z-transform can process the same signal without the need for an optical system.

Current Army tactics are placing greater emphasis on extended engagement ranges and improved adverse visibility performance. Two candidates for providing enhanced capability, in these areas, are (1) the far infrared high density focal plane technology and (2) the millimeter wave technology. The Army is currently exploiting both of these technologies for providing an improved target acquisition capability. Recent technical advances in infrared detector materials and focal plane signal processing have provided the impetus for high performance infrared imaging systems. Focal plane array candidates are available in both hybrid and monolithic structures which can provide single line video outputs. Charge coupled device technology is used extensively in these arrays to provide on-focal-plane time-delay-and-integration (TDI), multiplexing, and signal conditioning functions. Image processing research has evolved to the stage where implementation is feasible both on and off the focal plane. Recent advances in solid state millimeter wave devices have made it practical to consider multispectral tactical systems. Millimeter wave radars can provide an enhanced target acquisition capability and tracking to augment current infrared devices under heavy smoke or fog conditions. A millimeter wave system, using state-of-the-art signal processing can provide adequate resolution for both target detection and tracking using moderate antenna apertures. This paper will discuss current ongoing Army efforts in these technologies, their synergism, and effective utilization in tactical application.

Electra-optic technology appears to offer some rather dramatic payoffs to the ballistic missile defense (BMD) radar systems designer. At the current time, however, we are not able to evaluate the magnitude of the potential payoff or even to determine the functions for which electro-optic technology is most applicable. Uncertainties stem from the still embryonic status of coherent optical processing and from the limited number of BMD applications studies. Radar signal processing, the matched filtering of complex waveforms, has received the lion's share of BMD funding for electro-optic processing. The presumption is that an electro-optic processor would replace either a digital or a surface wave signal processor and would offer advantages in cost, weight, or waveform processing capability. The characteristics of coherent optical processors which make them seem most attractive are as follows: (1) Ability to store about one million words of data on a few square inches of surface area, (2) Ability to transform data cheaply by using lenses, (3) Ability to adaptively manipulate data based upon reference functions stored in electro-optic devices. Doubts about electro-optic processors seem to occur primarily in the following areas: (1) Dynamic range seems to be limited to about 40 dB for straightforward intensity storage, and (2) Available electro-optic storage devices/ modulators are generally limited to cycling rates not greatly in excess of 30 cycles per second.

Real-time convolvers and correlators having time-bandwidth products of about 10,000 have been achieved by utilizing the acousto-optic Bragg interaction. The large time-bandwidth products were achieved by using specially-designed large-bandwidth transducer arrays deposited onto large (~ 15-cm length), carefully polished crystals of lithium niobate (LiNbO₃) and bismuth germanium oxide (BG0). A correlator in which a "live" signal is correlated with a previously stored acoustic signal, by use of the acousto-optic interaction, also has been demonstrated by using a new process for storing acoustic images in LiNbO3. Storage results from the interaction between a high-intensity, short-duration green laser pulse and a surface acoustic wave propagating in the material.

A real-time coherent optical range/Doppler processor has been designed for BMDATC and the breadboard is being built. It consists of a pulse compressor based on an acousto-optic matched filter correlator followed by a two-dimensional range/Doppler processor utilizing a liquid crystal light valve. The design bandwidth is 150 MHz. Some very preliminary results are available.

A coherent optical processor is described which can be used to detect and locate objects in the sky that move against a background of fixed objects. A succession of sky images is recorded by multiple exposure on photographic film with the imaging system stabilized on the stars. Stars appear as points in the composite image, and moving objects create streaks. Space invariant pattern recognition is performed on this composite image by generating its two-dimensional Fourier transform (spatial frequency spectrum) and scanning it with a rotating spatial filter matched to the streak spectrum. The filter angle at maximum energy transmission determines the streak direction and a second Fourier transform provides coordinate information. The design of the optical system is described and some initial experimental results are shown.

A brief history of thermoplastic light modulators is given. The current state-of-the art is summarized and some applications are discussed. A new modulator now under development is described and recommendations for future programs are given.

In June, 1977 a novel electrophotographic film -- known as KC-Film -- was introduced to the Graphic Arts Market. This film consists of a very thin layer (3000 Å thick) of wholly inorganic material sputtered onto a polyester substrate in a proprietary process. The tiny, vertical crystallites (700 Å diameter) function as anisotropic image receptors fea-turing extremely short response times, better than one nanosecond, and they display discharge levels proportional to the amount of absorbed energy. Accordingly, it is possible to digitally record pulses at great bandwidth as either binary or intermediate level signals without reciprocity failure and at theoretical resolution levels of 10,000 cycles/mm. These signals can be developed into visible images with electrostatic toner with great fidelity for storage, reprocessing, viewing, or even printing. With especially formulated toners it is possible to print the recording lithographically, i.e., with multiple colors, generating rather esoteric imagery. The daylight handling, infinite storage, and erasability features are expected to lead this recording material toward many applications in the laser, LED, or electron beam recording fields.

Magnetic materials similar to those utilized in bubble memories can be used as a spatial light modulator in optical processing systems. Material magnetic and optical properties are discussed. Randomally accessed and serially accessed modulators are described and development work in these areas discussed. Performance capabilities of bubble domain spatial light modulators are projected.

Optical signal processors have application to laser and conventional microwave radars. This paper describes optical channelized receivers which offer 100 percent intercept probability, 1 GHz bandwidth and approximately 1 MHz detection cell resolution. The principles of operation of optical channelized receivers are set forth and typical performance specifications derived from actual system measurements are presented. These channelizers provide a detection capability which appears to closely match the coarse frequency determination requirements for coherent laser radar processors. A possible implementation of an optical channelizer in a LADAR doppler receiver is discussed.

High resolution radars receivers and electronic intelligence (ELINT) collection systems demand high performance, large time-bandwidth product (TB) devices for signal processing. The fiber-optic techniques demonstrate some excellent capabilities, such as low attenuation, wide bandwidth and long time delay, which are not achievable with conventional electronics or Surface Acoustic Wave (SAW) methods. Several key fiber-optic microwave signal processing components with TB more than 10,000 are presented in this paper.

The Ruticon is a solid state optical image modulator consisting of a metallized elastomer layer coated on a photoconductor layer. An electrical field is placed between the metal surface and a transparent conductive substrate. An input image, such as fron a CRT or a laser, causes a change in the distribution of electrical fields across the device, and the mechanical forces associated with these electrical fields cause the metallized elastomer surface to deform into an image pattern. Laser light reflected from this surface is phase modulated with the input image information and this modulated light may be used as the input to a coherent optical processing system. The operating characteristics of present and projected Ruticons will be described, along with examples of optical processing done with the Ruticon.

In the face of the increasing need to be able to process two-dimensionally organized data, the present electronic digital processing and coherent optical data processing techniques have proven inadequate. The potential advantage of coherent optical data processing (CODP) is that, as opposed to the serially organized electronic digital computers, it can simultaneously process data on a very large number of parallel channels. The limitation to the usefulness of CODP has been the lack of a high-speed, high-spatial-bandwidth input device. This paper describes the development of a new CODP input device capable of the required speed and spatial bandwidth. The device employs an integral CCD input to accept serial electronic signals and a liquid-crystal light valve output for CODP.

This paper documents internally funded optical signal processing (OSP) work performed at the McDonnell Douglas Astronautics Company. The concepts deal with the current digital radar signal processor configuration and an OSP system that simulates the algorithm sequence of its digital counterpart. A 16-pulse Frank polyphase (FPP) waveform of Order 7 is processed both optically and digitally. The results from both processing systems are compared and it is determined that the accuracy and fidelity of the OSP approach are still not equal to those of the digital processor. However, the use of OSP is justified because parallel processing of multiple wavefornis using OSP techniques can solve the spline-rejection problem.

A number of requirements exist for processing signals of very long duration and moderate bandwidth (~10 MHz), such as in high resolution radar and secure communications systems. Acousto-optic methods have been known for many years, but have not been adopted in practice largely due to the simplicity of acoustic surface wave devices. Acousto-optic processors are, however, competitive, where very long storage times, up to one millisecond, are re-required. We have built delay lines with paths folded in two dimensions for storage up to 350 microseconds, and delay lines folded in three dimensions with storage near one milli-second. The acousto-optic processor was applied to the demonstration of synchronization and detection of data of a secure communications system, and also to the cross-correlation of a Barker coded signal with a set of Doppler shifted references.

Space variant optical processors using coordinate transformations are considered as advanced optical pattern recognition systems. A general formulation of the processing method used is provided together with seven examples of the use of this processing technique in both image and signal processing.

The basic physical theory of the photorefractive effect will be presented. Methods will be discussed for utilizing the effect for the storage of information as well as optical display of information. The fundamental limits on writing time, access time and storage density will be related to the physical properties of the materials involved. Particularly promising materials, such as LiNb0₃ will be discussed in detail.

Recent work has shown that optical processors still hold promise of continuing to provide the highest data rate capacities. Estimates are that optical processors will continue to be three to four magnitudes greater than comparable digital systems well into the 1990's. The problem with optical processors still remains in the selection and design of appropriate interface (input/output) devices. More fundamentally is the provision of an adequate theoretical method for translating system design parameters from the operator need base to optical device design characteristics or parameters for the optical element design. The problem seems to be one in identification of appropriate trade-offs and parameter identification that have a common meaning for each possible device consideration. Even more important is the implementation of the so-called real-time requirement. This meaning of real-time and its impact on device sensitivity, resolution, and cycle time (and life) are not clearly defined. In this paper attention will be given to defining a basis of comparison of device design and translation of processor system requirements to optical system design requirements. A comparison of various device characteristics will be given.

A brief summary of activities in optical signal processing at the University of Arizona is presented.

In this paper a particular type of numerical optical data processor is presented. It uses the prime residue representation and a controlled birefringence spatial light modulator to perform the arithmetic operations. In this paper we will first outline the numerical representation and then describe the proposed physical implementation.

1-D to 2-D mappings have been used in optical signal processing to increase the information throughput of optical systems for 1-D signals and to allow shift-variant processing of 1-D signals. We review two examples of such mappings: the spectral analysis of raster recorded signals, and frequency-variant spectral analysis. We then present two new optical signal processing operations where 2-D signals are mapping into 1-D formats for processing: a method for general space-variant processing of 2-D signals, and a novel nonframing video imaging technique.

The optical analog computer offers the user a very high-speed one-dimensional or two-dimensional system for conducting spectrum analysis, matched filtering, or correlation detection. This type of computer offers a significant advantage in hardware performance (equivalent bits per second per dollar) over the digital computer in certain important signal-processing areas.

This paper presents a tutorial overview of some recently developed coherent optical pro-cessing techniques for performing certain types of computations often encountered in radar signal processing applications. The processors described utilize two one-dimensional transparencies which represent the spatial analogs of temporal waveforms. These transparencies are commonly rotated 45° in opposite directions, and serve as the inputs to a coherent optical processor which performs a one-dimensional Fourier transform. For various orientations of the input transparencies, possible operations include a simultaneous range/Doppler display of the input's ambiguity function (for identical input transparencies) and a cross-ambiguity function display (for non-identical transparencies). One dimensional coherent correlations and convolutions can similarly be performed without the motion or Fourier transform encoding commonly required in conventional coherent processors. Sample experimental results are presented along with the basic theory.

The technique of acousto-optic spectrum analysis can be used to improve the signal to noise of r.f. signals by the noncoherent integration of photoelectrons in a properly configured optical sensor. The requirements placed upon the sensor and acousto-optic device to obtain a given processing gain, resolution, analysis bandwidth, and dynamic range are discussed for bulk optical devices.