Tuesday, March 15, 2011

Upcoming AP-S & AES-S Talk: Efficient Beam Scanning, Energy Allocation, and Time Allocation for Search and Detection

The IEEE AP-S is thrilled to co-host an upcoming seminar with AESS entitled, "Efficient Beam Scanning, Energy Allocation, and Time Allocation for Search and Detection." This seminar will be presented by Duane J. Matthiesen from Technia Consulting, located in Lexington, Massachusetts.

This seminar will take place on March 22nd at 6pm at the MIT Lincoln Laboratory A-Cafe, 244 Wood Street, Lexington, MA. For directions please see:

For more details, please visit:

Be sure to bring your friends and colleagues to this event. I look forward to seeing you there.

Christy F. Cull, AP-S Boston

Duane J. Matthiesen abstract

Search Is an important function of every radar. It is quite interesting – even startling – that after at least 70 years of extensive radar system development there are no papers, books, or chapters (e.g., in Skolnik’s Radar Handbook) on radar search. And although there are many papers, books, and chapters on single-look signal detection, none of these pose or solve the fundamental and practical problem of making sensor detection efficient in terms of radar energy expenditure and radar time expenditure. This talk fills those voids.

This talk reviews recently-developed unique and innovative concepts for both search and detection that were derived for an active radar sensor. However, the formulation is completely general, so these results also apply to other both active (transmitting and receiving) electronic sensors and passive (receive only) electronic sensors such as infrared, optical, sonar, seismic, astronomy, or passive RF sensors, and astronomy (optical, infrared, radio, etc.).

These results were developed by the author over the past 25 years when tasked with designing practical and efficient search and detection modes for various radars, both electronically-scanned and mechanically-scanned.

The talk shows how to select for the next search action (look) the optimal (most efficient for detection):

  • beam position to point the center of the beam coverage during the look
  • signal energy to transmit for the look, for an active sensor such as a radar
  • radar time to allocate for the look, for either an active or a passive sensor.

These optimal search and optimal detection principles provide long-missing search and detection design philosophy/insights/goals and a quantitative measure of the efficiency of actual search and detection designs.

An electronically-scanned antenna and an energy-variant search waveform suite with transmitted energy levels of nominally 3 dB (± 1.5 dB maximum from optimal) allow implementation of these optimal search and optimal; detection techniques to provide essentially ideal (100 % efficiency) search and detection.

The most efficient detection performance is provided by optimal detection. Optimal detection theory is an extension of classical (1940’s to 1970’s) signal detection theory which considers only optimal noise/interference filtering and single-look detection. Optimal detection theory extends classical detection theory by additionally considering multiple-look detection and derives the signal energy allocation per look which maximizes the cumulative probability of detection for a total expenditure of energy (several looks).

There are at least a dozen books on radar signal detection theory and D. O. North derived the conjugate (matched) filter for the fundamental white noise/interference case in 1943. So the general perception of the electrical engineering community is that "everything is known about signal detection and there is nothing new to learn". This talk changes that perception appreciably. The problem is that all those signal detection papers and books only consider single-look detection, whereas any practical detection scheme utilizes multiple-look detection – an entirely new concept for optimization. Every real-world radar or other sensor system employs multiple-look detection, because single-look detection is highly unreliable and inefficient (requires a large expenditure of time and/or sensor energy to obtain a high probability of detection such as 0.999).

Further, optimal detection theory was soon determined to extend the classical (1940’s to 1970’s) optimal search theory significantly and complete it for radar and other electronic systems. The optimal detection optimization turns out to be a suboptimization of the optimal search optimization.

Optimal detection optimizes detection in one search beam when the probability of a target being located in all beams (potential detection cells) are equally likely. Optimal search optimizes detection over all search beams (potential detection cells) when the probabilities of a target being located in each beam are, in general, different.

Electronically-scanned beam agility is essential to implementing optimal search with essentially zero overhead time lost due to scanning (moving) the beam to its next required search position for the next look (target dwell). However an agile mechanically-scanned beam can often scan short-range targets with only slightly less than ideal (90 % - 95 %) scan efficiency.

The concepts in this paper are applicable to both passive and active detection sensors. For radar and other active sensors the resource expenditure (“effort”) consists of both transmit energy and radar (antenna) time (where the time consists of transmit time, round-trip propagation time, receive time, and scanning movement overhead time, some of which can be overlapping). For a passive sensor such as an infrared, optical, or passive RF sensor, the sensor resource expenditure (“effort”) consists of only the dwell time to receive the target's radiated energy.

Duane J. Matthiesen is an electrical engineer with 36 years of diverse expertise and experience in radar systems engineering for many different types of radars: ground-based, airborne, ship-based, and space-based. He has worked on various phased array radars during about 30 of these years.

He received a BS in electrical engineering from Oklahoma State University in 1967. He was a graduate student at MIT during 1967 – 1969 working on a MS in electrical engineering. He completed 11 graduate courses in electrical engineering and applied mathematics. His primary areas of concentration were in electromagnetics, probabilistic applications, random processes, and detection and estimation.

He has presented 10 professional papers on radar systems at IEEE and other professional conferences.

He is currently an independent consultant in radar systems engineering. He is writing a book on “Optimal Search and Optimal Detection” for radar and other electronic sensor systems. These concepts were conceived, developed, and completed during the past 25 years while he was tasked with developing various radar detection mode designs for both phased array radars and mechanically-scanned radars.

He has a long history of service to the IEEE at the Section, Region, and Institute levels. He was Chair of his IEEE Student Branch during his senior year of undergraduate. He has served with the IEEE Boston Section Aerospace and Electronic Systems Chapter for many years. He served as Boston Section Membership Development Chair and Region 1 Membership Development Chair for many years. He was Chair of the IEEE Boston Section during 1987 - 1988. He served as a Director of the IEEE Boston -New York Electro conference during 1990 – 1994. During the past 20 years he has served on the IEEE Boston Section conference committees of six very successful international radar and phased array systems/technology conferences.

He is an IEEE Senior Member. He received an IEEE Distinguished Service Award from the IEEE Boston Section in 2006. He received an IEEE Millennium Medal for outstanding achievements and contributions to the IEEE from the IEEE Boston Section in 2000.





No comments:

Post a Comment