Owing to the unprecedented development in terahertz (THz) sources and detectors in the last decade, technologists and researchers have intensified their efforts to develop advanced THz sensing and imaging systems with superior performance and unique functionalities. One of the key elements to realize such systems is the ability to monolithically integrate high-performance semiconductor devices with THz antennas and other passive structures. In recent years, devices based on interband tunneling in III-V heterostructures have emerged as promising candidates for THz detection that offer extremely high nonlinearity, high sensitivity, low noise, fast response, and room temperature operation. In this paper, we first review the development of heterostructure backward tunnel diodes (HBDs) in the InAs/AlSb/AlGaSb material system that have been demonstrated with detection sensitivity that outperforms the current state of the art (e.g., the fundamental limit of Schottky diodes) and with noise-equivalent power (NEP) below 0.2 pW/Hz½. We then present the monolithic integration of HBDs with planar folded dipole antennas (FDAs) using submicrometer-scale airbridges to achieve optimized impedance matching for high-performance and compact THz detectors and focal-plane array (FPA) imaging systems. In addition, the potential of using HBDs for realizing THz systems with advanced functionalities such as spectroscopic FPAs (using frequency-tunable THz antennas) and polariametric detection/imaging systems will be discussed. Finally, the integration of HBDs into waveguides for more advanced THz sensing and imaging (e.g., a six-port reflectometer for near-field imaging) will be discussed.

In this paper, we introduce a new software development method that emphasizes software testing cycles and makes a cultural change in program development. In our approach, students follow the software lifecycles to develop their programs; specifically, a student who is to write a program begins by writing a test suite and completes other development cycles. Students in programming courses usually develop "toy" programs that are superficially tested, graded, and eventually discarded. Generally, students are not worried about the quality or maintenance of their products. We believe this style of teaching programming courses leaves students unprepared for developing reliable software. Industry leaders also claimed that more than 50% of a software project's budget was spent on activities related to improving software quality. They stated the reason was the inadequate attention paid to software quality in the development phase. Our approach integrates into the existing programming courses without changing the course contents, syllabus, policies, or loads, and our model improves the students' program quality, in terms of black-box testing. Students indicated that our method helped them in understanding the problems and writing the programs, made the code easier to debug, and improved the students' code reliability and quality

The letter studies how the image false alarm rates are related to the target template false alarm rates when target templates are used for target detection. The authors show with both the analytical and simulated examples that when the image size is much larger than the target template size and the image false alarm rate is small, the image false alarm rate is approximately equal to the target template false alarm rate multiplied by the number of pixels in the image.>