All AESS Chapters and IEEE Sections are encouraged to take advantage of the AESS Distinguished Lecturer and Tutorial Program for their regular or special meetings. We have selected an outstanding list of speakers who are experts in their fields. The AES Society will pay reasonable speaker’s expenses for economy-class travel, lodging and meals. As a general guideline, speaker’s expenses involving travel wholly within North America or within the European Union will be covered up to $1,000. Expenses involving extensive international travel will be covered up to $2,000. The Society encourages arrangements whereby more than one lecture is presented in a single trip, and costs in such situations will be considered on a case by case basis. The inviting organization is expected to cover 50% of the speaker’s expenses.
The procedure for obtaining a speaker is as follows: If a Chapter or Section has an interest in inviting one of the speakers, it should first contact the speaker directly in order to obtain his or her agreement to give the lecture on a particular date. After this is accomplished, the Chapter or Section must notify the AESS VP for Education by sending in a DL Request Form. If financial support from the AESS is required for the speaker’s expenses, he or she must submit an estimate to the AESS VP for Education before actually incurring any expenses. This estimate must be provided at least 45 days before the planned meeting to provide time for feedback from the VP for Education and for changes if needed. The VP for Education must provide written authorization to proceed.
Distinguished Lecturers and Tutorial speakers are ambassadors of the AESS. As such, they should take advantage of the opportunity to stimulate membership in IEEE and AESS in particular. To support this goal, the Society has prepared a short presentation on the benefits of Society membership. Speakers should contact Judy Scharmann well in advance of each lecture to arrange for shipping AESS and IEEE Membership brochures and back copies of Society Publications to hand out.
After giving a lecture, the speaker and/or host should prepare a short report suitable for publication in Systems Magazine and posting on the AESS web site. Pictures taken at the meeting are highly desirable.
Lecture Title: Synthetic Aperture Radar
The first spaceborne SAR system was carried by NASA's SEASAT satellite in 1978. This only lasted for 3 months, when a massive power supply fault cut short its life. Nevertheless, this provided a wealth of data (much of which still remains to be properly analysed), and demonstrated the value of spaceborne SAR for a wide variety of applications in environmental monitoring. Subsequently NASA, the European Space Agency, Japan, Canada and several other Agencies have built and flown satellite SAR systems of increasing sophistication, now often with multiple frequency bands and polarimetric capability. In parallel, data interpretation techniques have progressed - indeed, it has been suggested that the extraction of quantitative information from SAR imagery represents the greatest current problem.
Aircraft-borne SAR is used both for remote sensing, and for high-resolution military surveillance. Resolution of the order of centimetres can be achieved with spotlight-mode operation, and target detection and recognition algorithms are being developed, as well as MTI to separate moving targets from stationary clutter. At such high resolution, characterisation and correction of motion errors becomes more and more important.
Interferometric SAR is currently a very active area of SAR research and development. The technique was first demonstrated with airborne SAR back in the 1970s, but subsequently it has been widely used with spaceborne SAR for high-resolution topographic mapping. With aircraft-borne systems there is the potential to recognise targets from their 3-D signatures.
Differential interferometry has demonstrated remarkable results in detecting changes in topography caused, for example, by earthquakes and volcanoes.
Finally, synthetic aperture techniques have also been applied in the field of sonar, to give high-resolution maps of the seabed, for applications such as the detection of wrecks, in the oil industry, and for the detection of mines. The principles are very similar, but the velocity of sound in water is very much slower (~1500m/s), which introduces certain problems, and the propagation of sound through water is strongly influenced by variations in temperature and salinity.
This lecture gives a subjective and selective overview of some current topics and results in modern synthetic aperture radar.
Lecture Title: Bistatic & Multistatic Radar
Furthermore, bistatic radar systems can utilize VHF and UHF broadcast and communications signals as 'illuminators of opportunity', at which frequencies target stealth treatment is likely to be less effective.
Bistatic systems have some disadvantages. The geometry is more complicated than that of monostatic systems. It is necessary to provide some form of synchronization between transmitter and receiver, in respect of transmitter azimuth angle, instant of pulse transmission, and (for coherent processing) transmit signal phase. Receivers which use transmitters which scan in azimuth will probably have to utilize 'pulse chasing' processing.
Over the years a number of bistatic and multistatic radar systems have been built and evaluated. However, rather few have progressed beyond the 'technology demonstrator' phase. Willis, in his book Bistatic Radar, has remarked that interest in bistatic radar tends to vary on a period of approximately fifteen years, and that currently we are at a peak of that cycle. The purpose of this lecture is therefore to present a subjective review of the properties and current developments in the subject, with particular emphasis on 'passive coherent location' and to consider whether or not the present interest is just another peak in the cycle. It draws on material in the book Advances in Bistatic Radar, edited by Willis and Griffiths, and recently published by SciTech.
Lecture Title: Foliage Penetration Radar
• The early history of FOPEN Radar: battlefield surveillance and the early experiments in foliage penetration radar are covered. There were some very interesting developments in radar technology that enabled our ability to detect fixed and moving objects under dense foliage. The most important part of that technology was the widespread awareness of the benefits of coherent radar and the advent of digital processing. Almost as important was the quantification of the radar propagation through foliage, and its scattering and loss effects.
• FOPEN synthetic aperture radar (SAR) with concentration on development results from several systems. These systems were developed for both military and commercial applications, and during a time of rapid awareness of the need and ability to operate in a dense signal environment. A brief description of each radar system will be provided along with illustrations of the SAR image and fixed object detection capability. The next section will quantify the benefits of polarization diversity in detecting and characterizing both man made and natural objects. There is a clear benefit for use of polarization in the target characterization and false alarm mitigation. Finally the techniques developed for ultra wideband and ultra wide angle image formation will be presented.
• New research in Multi-mode Ultra-Wideband Radar, with the design of both SAR and moving target indication (MTI) FOPEN systems. Particular note will be taken on the benefits and difficulties in designing these ultra-wideband (UWB) systems, and operation in real world electromagnetic environments. At common FOPEN frequencies, the systems have generally been either SAR or MTI due to the difficulties of obtaining either bandwidth or aperture characteristics for efficient operation. The last two sections of the tutorial will illustrate new technologies that are appearing in the literature that have promise for future multimode operation: the need to detect low minimum discernable velocity movement; and the operation of bistatic SAR in concert with a stationary GMTI illumination waveform.
Lecture Title: Real World data fusion
be addressed to achieve such improvement: (1) residual bias errors between sensors; (2) dense multiple target environments; (3) unresolved data; (4) errors in data association between sensors; (5) sensor errors that are not fixed in time or space but which are not white noise either. We describe state-of-the-art algorithms that attempt to mitigate such problems. We show simple back-of-theenvelope formulas which quantify the situation, as well as one well known formula that is extremely pessimistic.
Lecture Title: Is there a royal road to robustness?
estimation algorithms that are more robust than Bayesian algorithms. In particular, some engineers think that fuzzy logics or Dempster-Shafer methods are more robust than Bayesian methods.
We discuss a long list of standard methods to improve robustness, as well as a little known fact about the robustness of Bayesian algorithms.
Lecture Title: MIMO radar: snake oil or good idea?
Lecture Title: Never trust a simulation without a simple back-of-the-envelope calculation that explains it
engineering culture that valued simple back-of-the-envelope calculations to provide insight into what was going on. Moreover, I am appalled when I see system engineers blindly believe the
results of simulations. My talk will give five or ten examples of system engineering blunders caused by faulty simulations or erroneous physical experiments, as well as two surprising twists.
Lecture Title: Nonlinear filters with particle flow
many orders of magnitude faster than standard particle filters for problems with dimension higher than four. We do not resample particles, and we do not use any proposal density, which is a
radical departure from other particle filters. We show very interesting movies of particle flow and many numerical results. The key idea is to compute Bayes’ rule using a flow of particles
rather than as a point wise multiplication; this solves the well known problem of “particle degeneracy”. Our derivation is based on freshman calculus and physics. This talk is for normal engineers who do not have log-homotopy for breakfast.
Lecture Title: Navigation Sensors and Systems in GNSS Degraded and Denied Environments
Lecture Title: Inertial System and GPS Technology Trends
Lecture Title: National Missile Defense
The NMD program will continue to be a key technical, political, and legislative issue facing the U.S. and the rest of the world. The Bush Administration focused more on testing and developing new equipment for the NMD system and also investigated a wider variety of sensors (such as space-based and sea-based systems) to detect and track incoming missiles. The upgrade to the existing Early Warning Radars was one of the few features that did not change from the Clinton plan. The Obama Administration is still finalizing its approach to NMD.
This talk will provide background information on the political issues facing NMD. It will also provide technical information on some of the major systems including upgrades to the Early Warning Radars. The talk will also provide system engineering details on the proposed elements of the system that could be installed in Europe.
Lecture Title: Satellite Communication Systems
Lecture Title: Antenna Systems for Aerospace Vehicles
Lecture Title: Global Navigation Satellite System
Global Navigation Satellite System (GNSS) is a vast system of systems, providing global positioning, navigation and timing information to scores of users in oceans, land, air and even in space. The lecture module traces the history of navigation, evolution of navigation satellite systems, the three present constellations (GPS,GLONAS,GALILEO) and the world scenario in this direction including the S-BAS system. The lecture module will also touch upon the basics of position, velocity and time measurements, various GNSS connected aspects, their applications and the technologies associated including the S-BAS system.
Lecture Title: Target Tracking and Data Fusion: How to Get the Most Out of Your Sensors
Lecture Title: Achievement, Breakthroughs and Future Trends in Phased Arrays and Radars – Updated to 2014
Systems: 3, 4, 6 face “Aegis” systems developed by China, Japan, Australia, Netherlands, USA; FAA NexGen ATC system; AMDR, Space Fence, JLENS; S/X-band Dual Band Radar, AN/TPN-2, Airborne AESAs; Low Cost Packaging: Raytheon funding development of low cost flat panel X-band array using COTS type PCB; Lincoln-Lab/ MA-COM developing low cost S-band flat panel array using PCBs, overlapped subarrays and a T/R switch instead of a circulator; Extreme MMIC: 4 T/R modules on single chip possible at X-band costing ~$10 per T/R module; 8-element phased-array on one 6 to 18 GHz receiver SiGe chip having 5-bit phase control and 8:1 combiner RF-beamforming; 16-element 45 to 50 GHz phased array transmitter chips; accurate on-chip built-in-self-test (BIST) at W-Band demonstrated on such extreme MMIC chips that drastically reduces testing and calibration time; wafer scale integration demonstrated at 110 GHz with high efficiency antennas and RF circuitry on-wafer, no dicing or mm-wave packaging; Within the next decade we expect to see such extreme MIMCs in garage door openers, videos players, computers, etc. all communicating with each other wirelessly needing ever higher bandwidth; Also used for communications MIMO at mm-wave with ultra-low cost multi-beam AESAs imbedded in everyday devices; Digital Beam Forming: Israel, Australia and Thales AESAs have an A/D for every element channel, a major breakthrough; Lincoln Lab and AFRL X-band have 600 MHz instantaneous wideband DBF at element development effort; Raytheon developing mixer-less direct RF A/D having >400 MHz instantaneous bandwidth reconfigurable from S to X-band; Low cost DBF at element arrays for on-the-move Ethernet by IMST Germany; MIT Lincoln Lab using 2W chip increases spurious free dynamic range of receiver plus A/D by 20 dB by compensating for receiver plus A/D nonlinearities, a 20 year advance; Radio Astronomy scientists looking at using arrays with DBF; Materials: With GaN can now put 5X to 10X the power of GaAs in same footprint; SiGe for backend, GaN for front end of T/R module. Will be helped by use for PCs, notebooks, cell phones, servers and GaN LED where they are expected to replace incandescent bulbs $100 billion industry; Si replacing GaAs and GaN for low cost from microwaves to mm waves; Metamaterials: Potentially low cost electronically steered antenna not using phase shifters at 20 and 30 GHz being developed; Stealthing by absorption from 2-20GHz, cloaking (where microwave signal goes around the target) demonstrated over 50% band at L-band; Can now focus 6X beyond diffraction limit at 0.38 μm – Moore’s Law can march on; French, SPCI PARISTECH, demonstrated 40X diffraction limit, λ/80, at 375 MHz; Can extend to IR; Can now customize 3-D metamaterials at optical wavelengths; Was used in cell phones to obtain antennas 5X smaller (1/10th λ) and have 700 MHz-2.7 GHz bandwidth simultaneously serving GPS, Blue Tooth, Wi Max and WiFi; Provides isolation between closely spaced antennas and antenna elements, Un. Michigan demonstrated equivalence of 1m separation with only 2.5 cm separation of two antennas on a ground plane using electronic bandgap (EBG) material; n--Doped graphene has negative index of refraction, first such material found in nature; US Army developed very low profile (3.3”) wideband UHF (250 to 505 MHz) magnetic metamaterial antenna which can replace the large very visible whip antenna used on Army vehicles at present; WAIM (Wide Angle Inmpedance Matching) demonstrated for phased arrays using Electromagnetic Band-Gap (EBG) material to reduce mutual coupling; Very Low Cost Systems: Valeo Raytheon (now Valeo Radar) developed low cost, $100s only, car 25 GHz 7 beam phased array radar; about 2 million sold already, more than all the radars ever built up to a very few years ago; Commercial ultra low cost 77 GHz Roach radar on 72mm2 chip with >8 bits 1 GS/s A/D and 16 element array; Un. Frequency Michigan developing low cost 240GHz 4.2x3.2x0.15 cm2 5 gm radar for bird inspired robots and crawler robots, scans 2ox8o beam ±25o; DARPA has goal to build 28,000 element 94 GHz array costing $1/element, 50 W total RF peak power; MIT offers courses for building SARs and AESAs costing only 100s of dollars; SAR/ISAR: Principal Components of matrix formed from prominent scatterers track history used to determine target unknown motion and thus compensate for it to provide focused ISAR image; Army Research Lab demonstrated 12 dB reduction in sidelobes of forward looking SAR back projection images for IED ultra wideband radar by use of Recursive Sidelobe Minimization (RSM) Algorithm; Technology and Algorithms: MEMS: reliability reaches 300 billion cycles without failure, Can reduce the power amplifier (PA) count or T/R module count in an array by a factor of 2 to 4, can also be used as tuneable microwave filters, like 8-12 GHz with ~200 MHz BW; MEMS + Piezoelectric Material = piezoMEMS: Is being looked at for use with flying insect robots; Revolutionary 3-D Micromachining: integrated circuitry for microwave components, like 16 element Ka-band array with Butler beamformer on 13X2 cm2 chip; COSMOS: DARPA revolutionary COSMOS program: Will allow integration of III-IV, CMOS and optics on one chip without bonded wires, Lead to higher performance, lower power, smaller size, components; MIMO (Multiple Input Multiple Output): where it makes sense, point out that contrary to what is believed conventional array radars can provide the same 1, 2 or 3 orders of magnitude resolution and accuracy improvement as claimed for MIMO arrays; Also MIMO does not provide any better noise jammer rejection or hot clutter rejection (noise jammer received by antenna after reflection from ground) than conventional array radars; Graphene: Potential for terahertz clock speeds using graphene transistors; Could be used for non-volatile memory, flexible displays and camouflage clothing, self cooling; Can be used as switch with 100,000 to 1 on/off ratio, IBM producing 200 mm wafers with RF devices; Signal Processing: Potential use of electron spin for memory; Potential for use of 12 iron atoms for 1 bit of memory; provide hard drive with 100X today density; DARPA UHPC Program: 100 GFlops in cell phone using only 2 W instead of the present required 600 W for the same throughput. Goal of DARPA-Intel UHPC program for 100 to 1000 reduction in computer required power by 2018; Intel manufacturing chips using 3D integrated circuits, Moore’s marches on; STAP: Use of STAP equivalent to 6 dB increase in antenna SLL 1-way; Knowledge based STAP using map information provides 15 dB higher S/I when using information on road locations, 10 dB when putting nulls in antenna pattern where strong clutter is; Superconductivity: We may still achieve superconductivity at room temperature. Superconductivity recently obtained for first time with iron compounds. May reveal what leads to superconductivity; Additional Advances: Potential for low cost printing of RF and digital circuits through use of metal-insulator-metal (MIM) diodes; Iridium/GPS (IGPS) Positioning Navigation and Timing (PNT) system demonstrated ability to locate objects to within 1 cm in minute; 3D Display: From 2D image without the need for special eyeglasses. Can be used for displaying 3D SAR and ISAR image on our radar screens. Being used for video games; Butler matrix using CMOS; Biodegradable array of transistors or LEDs for detecting cancer or low glucose; can then dispense chemotherapy or insulin; parallel processing to map our DNA for $1000; Can now grow functioning kidney and heart for rats; New polarizations, OAMs???
Lecture Title: MIMO Radar – Demystified and Where it Makes Sense to Use
• MIMO typically requires a different orthogonal waveform for each element for an N element MIMO array. We call such an array an element-MIMO array or E-MIMO array. This results in a large computational load. Requires FN2matched filters for N element array vs N for conventional array, where F is the number of matched filters needed per orthogonal waveform due tothe need for different matched filters for different signalDoppler shifts. In contrast a conventional array radar can use a linear FM waveform (chirp waveform) which is not sensitive to the signal doppler so that only F matched filters are needed instead of FN2, a factor of NF less or 30,000 if N=1,000 and F=30.
• It would be thought that because of the larger degrees of freedom that a MIMO array radar has it would provide better barrage noise jammer rejection. Actually its jammer rejection capability is the same. This becomes obvious when one realizes that the jammer rejection can be done first in the receiver without effecting the optimality of signal detection. When doing this the ability to reject the jammer or jammers is not dependent on the waveforms transmitted, and in turn whether it is a MIMO or conventional system. For a receive array of N elements the receiver architecture can consist of the formation of N focused beams for the detection of the targets over the field-of-view. The jammers present in each of the focused beams is rejected using a sidelobe cancellers (SLC) for each focused beam output. The auxiliary signals for the SLCs for a given beam are obtained from the outputs of the focused beams pointed in the directions of the jammers. The location of the beams pointed at the jammers can be easily determined by noting the strength of the outputs of the focused beams. The focused beams are approximations of eigenbeams. Ideally they should have nulls or low sidelobes in the direction of the jammers. This is an application of adaptive-adaptive beam forming for the jammer suppression (Brookner and Howell, IEEE Proc., April, 1986). Next the outputs of each of the jammer suppressed N focused beams is followed by transmit focused beamforming which consists first of a bank of FN matched filters followed the beam formers. This architecture avoids doing the jammer suppression after the jammer signals go through the orthogonal matched filters. Rec
• It has been claimed that MIMO can handle hot clutter (which is barrage noise jammer signals received after reflection from the ground) whereas conventional arrays can not. This is not true, conventional arrays can handle hot clutter just as well as MIMO arrays. Can reject hot clutter coming into the mainlobe of the target beam without rejecting the signal return equally as well for conventional as for MIMO arrays.
• MIMO array radar does provide better clutter rejection because nulls can be adaptively placed in direction of the clutter in the transmit beams as well as the receive beams. But this is at a cost in signal processing. What may be preferred in practice in most situations is to do what can be done for conventional array radars. That is to non-adaptively put nulls or form low sidelobes in the transmitbeams in the direction of the strong clutter whose direction would be known.
• When doing volume search an E-MIMO arrayradar is inefficient re energy usage and occupancy.Should instead use Subarray-MIMO (SA-MIMO) for search. SA-MIMO involves the breaking up the N element array into Nssubarrays on transmit and receive where each subarray behaves like a conventional array.This allows one to tailor the transmitted energy to the element beam shape loss. SA-MIMO will improve the search efficiency for a 120o horizon fence by 3.7 to 5.2 dB for element ideality factors n of respectively 1 and 1.5, where the element one way embedded element gain is given by cosn, where n is the ideality factor. SA-MIMO also reduces the signal processing throughput required by the factor Ns/N.
In near term MIMO radar usefulfor combining radars to get 9 dB higher Power-Aperture-Gain (PAG) and 6 dB higher Power-Aperture (PA). Also potentially useful for over-the-horizon (OTH) high frequency (HF) radars where signal bandwidth and the number of elements are not large. May be useful where size is a premium. Its advantages for airborne systems is not addressed here. MIMO is useful for communications where it takes advantage of channel multipath to increase channel data rate.
Lecture Title: Around The World In 60 Minutes –EXOTIC PLACES WITH A TWIST
This will be a talk of adventure, knowledge, humor and colorful slides. A show not to be missed. All are invited. Family and friends. An opportunity to bring your spouse, children, parents, neighbors and friends. The twist is an explanation that all can understand on how radars and phased arrays work and some of the recent amazing breakthroughs in radars.
Lecture Title: Outstanding Advances in Phased-Arrays and Radar
Many think that radar is a mature field, nothing new to happen, it having been around a long time. Nothing can be further from the truth. When I entered the field in the '50s I thought the same thing. The MIT Radiation Lab. Series 28 book volume set summarizing the highly classified World War II work on radar was just published and provided the definitive coverage and there was to be nothing more to learn. How wrong I was. Since then many amazing new developments have taken place. And astounding developments are still taking place. We live in exciting times.
We will cover the following recent outstanding breakthroughs in this talk:
1. Integrated circuits at microwaves (MMIC): Makes it possible to have:
a. Active arrays for applications not feasible before, like simultaneously air-to-air and air-to-ground modes on the F-18.
b. Whole T/R module on a single chip costing $10 at X-band.
c. $19K 35 GHz active phased array costing $30 per element.
d. 8 active array receive channels on one chip -- disruptive technology.
2. SiGe, CMOS: Offers potential for alternative low-cost, low-power per element active phased arrays.
3. Packaging and Assembly of Phased Using Commercial Printed Circuit Boards (PCB): Provides low cost arrays.
4. MEMS (Micro-ElectroMechanical Systems): Reliability has increased 3 orders of magnitude in 3 years. Has potential for providing arrays at 1/10th the cost.
5. Wide bandgap GaN and SiC MMIC chips: Potential of one to two orders increase in transistor power.
6. Digital Beam Forming (DBF): Provides the advantages of:a. Multiple beams
b. Lower search power and occupancy by up to a factor of 2.
c. Fully adaptive array performance without having to do a large matrix inversion (Adaptive-Adaptive Array).
7. MIMO (Multiple-Input Multiple-Outputs): This is the hot topic now. It is not all fantasy. Practical applications are:
a. Coherent combining of radars. With 2 radars we get a 9 dB increase in sensitivity.
b. Maximum signal-to-clutter interference through optimum adaption at the receiver of the transmitter as well as receiver array weights for clutter suppression.
8. Haystack Upgrade: 3 cm Resolution Imaging.
9. SAR (Synthetic Aperture Radar): 4" Resolution achieved.
10. Wideband Antennas:
a. 1.8 to 18 GHz instantaneous bandwidth array built by Raytheon.
b. 33:1 Instantaneous bandwidth antenna built by GTRI; 100:1 possible.
11. Vacuum Tubes:
a. Coherent gyrotron amplifiers: Now available at mm-waves.
b. Bandwidth, power, reliability, and efficiency greatly increased.
12. Metamaterials: Revolutionary negative index of refraction material can:
a. Stealth a radar target.
b. Permit focusing beyond the diffraction limit. Moore's Law marches on.
13. Adaptive Array Processing and Space Time Adaptive Processing (STAP):
13. KASSPER: Applies available environment knowledge to STAP to reduce false alarms by order of magnitude.
The above lecture can be tailored to the interests of the group.
Lecture Title: Overview of High-Level Information Fusion Theory, Models, and Representations
Lecture Title: Business Case for Systems Engineering - Is Systems Engineering Effective?
One of the oft-discussed elements in the field of Systems Engineering is how can one justify the expenditure of program or project monies for systems engineering? In short, what is the payback, or business case, for doing systems engineering? Those who are somewhat knowledgeable in the field of systems engineering know what the value is, but what are the tangible results of doing SE on programs and projects? How do we convince our program and project managers that SE is needed, or essential?
The Systems Engineering Division of the National Defense Industrial Association, in conjunction with the Software Engineering Institute (SEI) of Carnegie Mellon University initiated a comprehensive study in 2008 to try to determine the tangible benefits of performing SE in terms of program/project performance. The study consisted of a series of questions based on SE work products as defined in CMMI® (Capability Maturity Model Integration), which is the currently accepted systems engineering process model in widespread adoption, worldwide. The study concluded that there indeed is a positive correlation of SE performed and program/project performance in terms of budget (cost), schedule and requirements.
The number of responses to this initial study survey was small, in the order of 46 valid responses, from the US defense industry. In order to validate the results with a larger response base to include commercial as well as non-US organizations, in 2011 the NDIA and SEI partnered with the IEEE Aerospace & Electronic Systems Society to reach a broader audience, and the results of this updated survey with over 180 valid responses was completed and released in late 2012.
This lecture will present the results of the updated study of SE performed on programs/projects and program performance in terms of cost, schedule and requirements. It will show that programs with the greater amount of SE performed demonstrate the best performance, while the programs with less SE had a lower rate of success. Since the study correlates program successes in terms of specific SE activities, these results can be used within organizations to assist in establishing systems engineering plans on programs and projects.
Lecture Title: Compression Based Analysis of Image Artifacts: Application to Satellite Images
Artifact detection in Earth observation images becomes increasingly difficult when the resolution of the image improves. For images of low, medium or high resolution, the artifact signatures are sufficiently different from the useful signal, thus allowing their characterization as distortions; however, when the resolution improves, the artifacts have, in terms of signal theory, a similar signature to the interesting objects in an image. Although it is more difficult to detect artifacts in very high resolution images, we need analysis tools that work properly, without impeding the extraction of objects in an image. Furthermore, the detection should be as automatic as possible, given the quantity and ever-increasing volumes of images that make any manual detection illusory. Finally, experience shows that artifacts are not all predictable nor can they be modeled as expected. Thus, any artifact detection shall be as generic as possible, without requiring the modeling of their origin or their impact on an image.
Outside the field of Earth observation, similar detection problems have arisen in multimedia image processing. This includes the evaluation of image quality, compression, watermarking, detecting attacks, image tampering, the montage of photographs, steganalysis, etc. In general, the techniques used to address these problems are based on direct or indirect measurement of intrinsic information and mutual information. Therefore, this thesis has the objective to translate these approaches to artifact detection in Earth observation images, based particularly on the theories of Shannon and Kolmogorov, including approaches for measuring rate-distortion and pattern-recognition based compression. The results from these theories are then used to detect too low or too high complexities, or redundant patterns. The test images being used are from the satellite instruments SPOT, MERIS, etc.
We propose several methods for artifact detection. The first method is using the Rate-Distortion (RD) function obtained by compressing an image with different compression factors and examines how an artifact can result in a high degree of regularity or irregularity affecting the attainable compression rate. The second method is using the Normalized Compression Distance (NCD) and examines whether artifacts have similar patterns. The third method is using different approaches for RD such as the Kolmogorov Structure Function and the Complexity-to-Error Migration (CEM) for examining how artifacts can be observed in compression-decompression error maps. Finally, we compare our proposed methods with an existing method based on image quality metrics. The results show that the artifact detection depends on the artifact intensity and the type of surface cover contained in the satellite image.
Lecture Title: Over-The-Horizon Radar: Fundamental Principles, Adaptive Processing and Emerging Applications.
The first part of the tutorial introduces the fundamental principles of OTH radar design and operation in the challenging HF environment to motivate and explain the architecture and capabilities of modern OTH radar systems. The second describes mathematical models characterizing the HF propagation channel and adaptive processing techniques for clutter and interference mitigation. The third delves into emerging applications, including HF passive radar, blind signal separation and multipath-driven geolocation. A highlight of the tutorial is the prolific inclusion of experimental results illustrating the application of robust signal processing techniques to real-world OTH radar systems. This is expected to benefit students, researchers and practitioners with limited prior knowledge of HF radar and with an interest in the application of advanced processing techniques to practical systems.
Lecture Title: Robust Adaptive Array Processing for Radar
Lecture Title: Tracking and Sensor Data Fusion – Methodological Framework and Selected Applications.
Lecture Title: Multistatic Exploration – Introduction to Modern Passive Radar and Multistatic Tracking & Data Fusion
Lecture Title: Sensor selection for multistatic radar networks
The concept of bistatic ambiguity function (BAF), often used to measure the possible global resolution and large error properties of the target parameters estimates, will be introduced and its relation with the Fisher Information Matrix (FIM) and Cramér-Rao Lower Bounds (CRLBs) highlighted. Some example will be provided concerning active LFM radar and passive radar using an UMTS or FM signal as source of opportunity.
The information gained through the calculation of the bistatic CRLBs can be used in a multistatic radar system for the dynamic choice of the optimum Tx-Rx pair or set of bistatic channels for radar target tracking in a multistatic scenario. Taking advantage of the knowledge of the CRLBs is a kind of “radar cognition”, that, applied in multistatic realistic scenarios with both active and passive sensors, can improve the performance of the target tracker and reduce the computational load of surveillance operations. Some results will be shown in both certain and uncertain radar measurements.
Lecture Title: Advanced Techniques of Radar Detection in Non-Gaussian Background
In this talk, after a brief description of modern statistical and spectral models for high-resolution clutter, coherent optimum and sub-optimum detectors, designed for such a background, will be presented and their performance analyzed against a non-Gaussian disturbance. Different interpretations of the various detectors are provided that highlight the relationships and the differences among them.
After this first part, some discussion will be dedicated to how to make adaptive the detectors, by incorporating a proper estimate of the disturbance covariance matrix. Recent works on Maximum Likelihood and robust covariance matrix estimation have proposed different approaches such as the Approximate ML (or Fixed-Point) Estimator or the M-estimators. These techniques allow to improve the detection performance in terms of false alarm regulation and detection gain in SNR.
Some of results with simulated and real recorded data will be shown.
Lecture Title: Sea and land clutter statistical analysis and modeling
The lecture first introduces radar sea and ground clutter phenomena, measurements and measurement limitations, at high and low resolution, high and low grazing angles with particular attention to classical model for RCS prediction. Most part of the lecture will be dedicated to modern statistical and spectral models for high resolution sea and ground clutter and to the methods of experimental validation using recorded data sets. Some comparison between monostatic and bistatic sea clutter data will be provided together with some results on non-stationarity analysis of the high resolution sea clutter.
Lecture Title: Cognitive Dynamic Systems (CDS)
a) The perception part of the CDS is based on sparse coding, well known in neuroscience.
b) The control part of the CDS is built around the cognitive reinforcement learning algorithm.
c) The perception and control are reciprocally coupled by means of a Probabilistic Reasoning Machine (PRM) that builds on four primary functions of the preferential cortex:
. working memory
. attentional set (i.e., perceptional attention and control attention)
. error monitoring
. decision making.
Putting all these functions under its umbrella, the PRM proves itself to be a system stabilizer.
d) Risk control: For the first ever, I will describe a new pre-adaptation control mechanism that addresses this most difficult problem of them all.
The lecture will expand on a joint paper, involving myself and Professor Joaquin M. Fuster , UCLA, well known around the world for his contributions to Cognitive Neuroscience. The paper will be published in the April Issue of the Proc. IEEE, which is devoted to Cognitive Dynamic Systems.
Lecture Title: Cognitive Control
I will demonstrate important properties of this new reinforcement leaning, namely
a) Linear law of computational complexity, the best it could be.
b) Unlike traditional reinforcement learning algorithms, there are no approximations whatsoever
3) It is optimal in the Bellman sense
4) It is convergent
Simply put, the new reinforcement learning algorithm is a game changer.
The related utilities include:
. Explore-exploit strategy
Lecture Title: Cognitive Radar
1. The perception action cycle that embodies the transmitter and receiver inside a global feedback loop
to gain information about the environment that increases from one cycle to the next.
2. Memory, that consist of perception memory in the receiver and control memory in the transmitter, with working memory
coupling them reciprocally together. Memory is dynamic with the ability to learn . It is responsive for
predicting consequences of action taken by the transmitter.
3. Attention is algorithmic in nature and builds itself through local perception action cycles
4. Intelligence is distributed throughout the system, with decision making and optimality for action on the environment.
In this lecture I will demonstrate the superior performance of Cognitive radar when it comes to:
a) Management of resources,
b) Accelerated rate of convergence by orders of magnitude
3) Mitigation of the effects of unexpected disturbance
Lecture Title: Radar Adaptivity: Antenna Based Signal Processing Techniques
• Introducing Radar: from its conception to recent industrial achievements,
• Operational needs requiring adaptivity,
• Side lobe blanking and cancellation techniques,
• Adaptive arrays of antennas,
• Some practical application examples of adaptivity,
• Conclusions and way ahead.
Each part is structured with some mathematical background, presentation of key processing algorithms, performance evaluation of the algorithms either in closed form or via Monte Carlo simulation, practical engineering implications related to the implementation of processing algorithms and, finally, examples of application potentials. A comprehensive set of technical references is also provided for further study and investigation.
Lecture Title: Effective Maritime Domain Awareness - A Systems of Systems approach to Generating Actionable Intelligence
Maritime Domain Awareness (MDA) is about generating actionable information for confidence-based decision support. This requires collecting information pertaining to the whereabouts of all maritime targets in the surveillance area, including classification of vessel type and activity, positive identification, and threat assessment. No single sensor can achieve this and effective MDA requires a combination of passive and active surveillance and reconnaissance systems.
Assembling this picture, however, is only part of the solution. The key to effective MDA is the use of decision support tools that analyze vessel track information and identify anomalous vessel behavior.
1) Requirement for Maritime Domain Awareness
2) Maritime Regions and the Legal Framework
3) Layers of Maritime Domain Awareness
4) Comparison of Surveillance/Reconnaissance Options – Passive and Active
5) Requirement For Persistent Surveillance
6) Data Association, Fusion and Data Mining
7) Finding “the needle in the haystack” - identifying anomalistic behavior
8) Architectures for Maritime Domain Operations Centers