Radar Technology (hb 2016)

In this book “Radar Technology”, the chapters are divided into four main topic areas: Topic area 1: “Radar Systems” consists of chapters which treat whole radar systems, environment and target functional chain. Topic area 2: “Radar Applications” shows various applications of radar systems, including meteorological radars, ground penetrating radars and glaciology. Topic area 3: “Radar Functional Chain and Signal Processing” describes several aspects of the radar signal processing. From parameter extraction, target detection over tracking and classification technologies. Topic area 4: “Radar Subsystems and Components” consists of design technology of radar subsystem components like antenna design or waveform design. In this chapter basics of radar target classification technologies were introduced. A classification technology was presented, that decomposes a pattern recognition module of any modern radar system in the following components: Data acquisition part, signal preprocessing and feature extraction part, classification and subclassification part, data and information fusion part and finally object recognition or identification or typing part. For the data acquisition part an active or passive radar frontend can be used, that uses selfor friendly generated waveforms to reconstruct information from the environment with different objects or targets. The data acquisition part usually provides such backscattered radar echo signal in I- and Q-form in the baseband. For the signal preprocessing part some basic techniques were described in order to filter and normalise the sampled signal. It was mentioned that some measures must be taken into consideration in order to respect the basics of information theory. For the feature extraction part several basic techniques can be used. It was also mentioned that one of the most successful philosophies in designing modern radar systems for classification purpose is the best handling of the feature extraction. This philosophy consists of best understanding of the physical behaviour of a radar system in its environment. Based on this understanding characteristical feature must then been mathematically described depending on the given requirements. For this purpose the following basic methods were presented as central components of the feature extraction process: Short-Time-Fourier transform, cepstral analysis, wavelet transform and Fuzzy-logic. For the classification and subclassification part two main philosophies were presented. The first philosophy consists of learning processes. The second philosophy consists of knowledge based evidence. The different kind of classification and subclassification methods in the most modern radar systems can be divided into deterministical methods, stochastical methods and neural methods. The deterministical methods introduced in this section were essentially based on the handling of logical operators and knowledge based intelligence. The stochastical methods described in this section were based on finite stochastical automats. The finite stochastical automats presented in this section were based on different variants of learning Hidden Markov Models. Furthermore the neural methods presented in this section illustrate the capability of solving pattern recognition problems in modern radar systems by using different kinds of artificial neural networks. It was also shown that for specific classification or subclassification challenges in modern radar applications hybrid classifiers can also be recommended. This classifier uses depending on the situation learnable or non-learnable algorithms. The learnable algorithms can be designed using supervised or unsupervised learn concepts. For the data and information fusion part it was pointed out that different techniques and strategies can be used in order to fuse information from different sensor systems. It was also shown that the introduced data fusion techniques can also be integrated in a stand-alone sensor system in order to produce a robust classification and recognition result. For this purpose three technologies were presented in order to solve the given problems: Bayesian networks based method, Dempster-Shafer rules based fusion methods and finally classical rule based methods. For the object recognition, identification or typing part it was mentioned that in modern radar systems, recorded data as well as recorded intelligence information can additionally In this chapter, the basics of the antenna and phased array are reviewed and different wideband antennas for modern radar systems are presented. The concepts of the radome and frequency selective surface are also reviewed. The main contents include important parameters of the antenna, and theory and design consideration of the array antenna. Various wideband antennas are introduced and their performances are demonstrated, including: (1) for the phased array radar, the slotted waveguide array antenna has been widely used in airborne and marine radar systems; (2) for the ground penetrating radar, TEM horn antenna, broadband monopole antennas, and adaptive bow-tie antenna are presented; (3) for the noise radar, ridged horns and log-periodic Yagi antenna are introduced; (4) for the UWB systems, radiation characteristics of two newly developed UWB antennas (a coupled sectorial loop antenna and a spiral antenna with Dyson-style balun) are demonstrated. Furthermore, the radome and frequency selective surface used to protect the antenna/radar systems are brought in, whose functions, requirements, structures, and performance evaluation are presented, including state-of-the-art designs. The design considerations and future trend of the radar antennas and associated devices have been discussed and suggeted. More detailed can be referred to given references. The development of the wideband antenna technology triggers the advancement of the antenna in the radar system. The progress of the antenna design also reveals applications of many innovative materials and structures. Designing low-profile, ultra-wideband, and electrically small antennas have become one of the primary goals in the antenna society. Future radar systems will require ultra-wide/multi-band antennas with capabilities of highspeed, high resolution, and high reliability. Wideband antennas covering ELF-VLF and HFUHF will be challenging but eagerly desired. Smart impedance matching network should integrate with the miniaturized wideband antennas to form high efficiency antenna system. Also, fast and low-cost conductive printing technology can assist antennas conformal to various mobile platforms and hence become invisible. These innovative technologies will benefit future radar systems that may initiate a new research field for the wideband antennas In this chapter, we have first presented how to extract natural poles from the simulated transfer function of a target and how to use them for identification purposes. We have introduced a new representation of poles with quality factor Q and natural pulsation of resonance 0 in order to better separate information: the Q parameter permits to bring out clearly the resonance behaviour of targets (Q is a discriminatory of the aspect ratio of targets), and the natural pulsation of resonance 0 depends on dimensions of targets. Next, we have extended our study to resonances of targets with apertures. The simple example of a PC rectangular open cavity has permitted to show how resonance parameters depend on object dimensions and how internal and external resonances can be distinguished, by comparison of poles of the open cavity and the closed box. Internal resonances having a lower damping coefficient m than external resonances, they have a higher quality of resonance Qm, and can therefore be more easily extracted. That is why, we have used this interesting property of internal resonances in order to identify an aircraft with air-intakes. Thus, we have shown that the use of selective narrow frequency bands permits a better extraction of poles in the case of complex objects, as aircrafts with air-intakes