Signal Processing in Noise Waveform Radar

Signal Processing in Noise Waveform Radar

The concept of radar—radio detecting and ranging—emerged in the beginning of the twentieth century. The father of radar was Christian Huelsmeyer, who applied for a patent for his “telemobiloscope” on April 30, 1904. His device worked quite well and detected ships at ranges up to 3 km, but he had no success in selling telemobiloscopes, which is why the early radar concept faded from memory. The reinvention of radar was done almost simultaneously in many countries in the 1920s and 1930s, and great progress in radar technology was made during World War II. In the last 60 years, many different radar technologies have been developed, increasing the radar’s ability to detect targets in the presence of ground and sea clutter, to measure the target range velocity, and to create high-resolution images. In military areas radars are used for detecting, tracking, and imaging enemy airplanes, ships, satellites, and vehicles. They are used also for fire control, missile guidance, proximity detections, target identification, and many other purposes. Radars are also found in many civilian applications. They are used in air traffic control (ATC), ground traffic monitoring (including enforcement of the speed limit by police), collision avoidance and traffic assistance (automotive radars), observation of the Earth (biomass calculation, disaster control, change detection, and ocean currents monitoring) and in industry (measurement of liquid or powder level in tanks, object positioning, movement detection, and so on).

Early radars were noncoherent pulse radars. To obtain a long-range detecting capability, the radar emitted very short electromagnetic pulses with very high peak power (up to several megawatts), and such a large peak caused a lot of difficulties. The transmitter was big and heavy. Microwave waveguides had to conduct megawatt pulses, and their construction was complicated. The large peak power also had an impact on the environment. All life in the surrounding areas near the radar was exposed to this hazard; birds were often killed by electromagnetic radiation, and trees sustained damage.

The maximum detection range of a radar is limited by several factors such as the antenna, the size of the target, the noise level in the receiver, and, one of the most important, the total energy radiated towards the target. It is possible to exchange the transmitted pulse width with the peak power. In a classical pulse radar an increase of the pulse length decreases the radar range resolution, which is generally undesirable. To obtain relatively low peak power and a high-range resolution simultaneously, different types of pulse compression are used. Among them the most popular has been compression based on linear frequency modulation. To improve the radar’s properties, different types of intrapulse modulation have been investigated, such as binary-phase modulation, polyphase modulation, and noise or pseudo-noise modulations.

In military radars there is a continuous contest between the radars and electronic support measurement (ESM) devices. The modern military radar should detect the target before being detected by an enemy ESM receiver. The ESM detection range depends on the peak power of the signal transmitted by the radar, so the concept of low probability of interception (LPI) continuous-wave radars was introduced. Such a radar, usually equipped with separate transmitting and receiving antennas, could simultaneously transmit and receive signals so the pulse length can be extended almost infinitively (ranging in practice from nanoseconds to seconds, several orders of magnitude). As a result, the peak transmitted power could be reduced by even a million times (60 dB). Again, as in the case of pulse compression, different types of signal modulations have been investigated. Many continuous-wave radars used sawtooth frequency modulation of the transmitted signal. The signal processing algorithms for linear frequency-modulated continuous-wave (LMCW) radars are relatively simple (based on the two-dimensional FFT) and require reasonably low computational power, so LMCW radars could be simple and cheap.

Recent progress in ESM technology has made it possible to detect FMCW radars, so radar scientists are looking for new technologies that will improve the LPI properties of the radar and prevent ESM receivers from detecting and classifying the signal transmitted by the radar. As of the time of this writing, there were several candidates for modern LPI radars. The first of them was noise radar technology. The second was passive coherent location technology, and the third was passive technology based on a signal’s time difference of arrival (TDOA) at several different locations.

In most classical active radars the transmit signal is periodic. The periodicity introduces ambiguities in the radar measurements. The long-range pulse radar sends a sounding pulse and waits for an echo to return from the maximal distance. As a result, the pulse repetition frequency is low and such a radar suffers from Doppler ambiguity (Doppler frequency measurement is ambiguous). It is also possible to build high pulse repetition frequency (high PRF) radars, but an increase of the PRF decreases the unambiguous detection range and, as a result, introduces range ambiguities. It was not possible to measure both the distance from the target and the target radial velocity instantaneously without the ambiguity caused by the sampling theorem. To fulfill different requirements, three main types of pulse radar are used: low pulse repetition frequency (LPRF), medium pulse repetition frequency (MPRF), and high pulse repetition frequency (HPRF). To resolve Doppler or range ambiguity, different estimation methods, usually in conjunction with pulse staggering (changes of PRF during target illumination), are used. LPRF radars are free from range ambiguity, but have Doppler ambiguity. They are commonly used as ground-based surveillance radars. HPRF radars have range ambiguity but do not have Doppler ambiguities. They are commonly used as airborne radars and tracking radars. MPRF radars have both range and Doppler ambiguities. This radar operation mode is often used as an additional mode in HPRF airborne radars. Nevertheless, the idea of the pulse radar can lead to a relatively simple design, which can be implemented using analog-only devices.

The rapid progress in digital signal processing (DSP) hardware and algorithms has enabled designers to use more sophisticated ideas and more complicated signal processing algorithms in modern radars. One such idea is to use a continuous wave (CW) instead of high power pulses. This idea is not new; the first Daventry experiment (see Chapter 2 for details) was based on CW radio emissions, but the practical use of CW emissions was limited by the necessity to build a sophisticated matched filter set. Modulated CW radars were built during World War II and discussed in the earliest postwar literature, but most early CW radars were constructed as Doppler-only radars (e.g., police radars) to measure the velocity of the target. More advanced systems used periodical linear frequency modulation (FMCW) to obtain range resolution. Due to a lack of modulation (in single-frequency police radars) or periodicity in modulation (in FMCW radars), the ambiguity problem was not removed, but CW radars have some very strong advantages. Many CW radars use low transmitted power (1 mW up to 100W) and due to low peak power (equal to the mean transmitted power), it is difficult to detect and classify this type of radar, so they are treated as a low probability of intercept (LPI) class of radars. Such radars can detect targets before being detected.

The idea of LPI radars in connection with the requirement of designing an ambiguity-free radar leads directly to the concept of the noise or pseudo-noise radar. In this kind of radar the target is illuminated by continuous noise-like radiation. The reflected power is collected by the radar receiving antenna, and detection is based on the matched filtering of the received signals. Due to the fact that both the target range and the target velocity are unknown, then instead of using a single filter matched to the selected target’s range and velocity, it is necessary to use a two-dimensional set of filters, matched to all possible targets’ range-velocity pairs. The detection of targets in noise radars thus requires very high computational power and cannot be effectively implemented in the analog technique.

The theoretical cross-ambiguity function of the noise signal has a delta shape: infinite value at zero time shift, zero Doppler frequency, and zero elsewhere. The transmit signal is not periodic, so there are no range or Doppler ambiguities in the measurement.

Of course, nothing is free, so we pay a price for the lack of ambiguity. The delta-shaped cross-ambiguity function is for an ideal white noise signal, infinite in time and frequency (bandwidth). Such a signal would have infinite energy and could not be generated by any real radar. In practice we can generate a sounding system limited in time and bandwidth (finite energy). In such a case the cross-ambiguity signal has a clearly visible peak at the zero Doppler frequency and zero time delay, and the width of the peak in time is equal to the inverse of the signal bandwidth and in frequency, the inverse of signal duration time. Such a signal also has nonzero values out of the peak region. Time and range residues of the cross-ambiguity function (which can be treated as the sidelobes in the case where single noise realization is treated as a deterministic signal, for example, stored in memory) have random nature, but the level is BT (the product of the signal bandwidth and duration time) below the peak value. This means that the ambiguity problem was converted to the masking problem. Sidelobes of strong echoes can mask weak ones if the ratio of the strong to weak echoes is higher than the time bandwidth product.

The limitation of weak-signal detection in a noise radar is due to the problem of the strong echo signals, which usually originate from nearby ground clutter and are received simultaneously with the weak target echo. To ensure correct detection of distant targets, the radar has to have a very high dynamic range and very low intermodulation. When the matched filter of the noise radar is tuned to the range and Doppler velocity of the weak echo, the power of all other echoes can be summed with the receiver thermal noise power decreasing the effective signal-to-noise ratio of the weak target. This phenomenon, together with the method of countering it, is described in Chapter 4.

The concept of noise radar may be used for the detection of moving targets, but it can also have many other applications. There are several papers showing the possible implementation of noise technology in an imaging radar working in the synthetic aperture mode or inverse synthetic aperture mode (ISAR). This concept can be extended to the passive detection, imaging, and identification of targets. Noise radar will probably be used in the future in other fields, such as air traffic control, pollution control, and security applications.

This book presents the basic idea of noise radar and details of signal processing. It is assumed that the reader has a basic knowledge of signal processing and general radar technology. However, in Chapter 2 the fundamentals of the classical pulse radar are presented. More detailed analyses of classical radar processing can be found in the radar engineering bible and other radar books. In Chapter 3 the basic noise radar concept and correlation processing are introduced. Chapter 4 presents the effect of masking weak echoes by a strong cross-talk signal and strong echoes. In this chapter the number of algorithms designed to mitigate this problem is discussed. In Chapter 5 the idea of the multistatic noise radar network is presented. This chapter presents the idea of placing the transmitter and receiver in different locations, and discusses the advantages and limitations of such an approach as well as the methods for synchronizing the transmitter and receiver. This chapter also describes a more advanced idea of the use of a multistatic system, where several transmitters illuminate the target using the same carrier frequency. The approach presented leads to an increase of surveillance volume, measurement accuracy, and probability of detection.

In Chapter 6 the principles of noise synthetic aperture radar are presented. In Chapter 7 the new idea of the passive detection and imaging of moving targets is described. The idea presented is based on radiometry and exploits the thermal emission of microwave noise signals originating from moving targets. Chapter 8 shows selected examples of noise radars and draws conclusions on noise radar technology.


扫描下方二维码添加微信号 bookyage 告知所需电子书的书名即可,我们会尽快(一般24小时之内)将相应文件以百度网盘链接的形式发送给您。如果您不方便使用百度网盘,也可以留下电子邮箱地址,我们会以电子邮件附件的形式将pdf发送给您。




QR Code


电子邮件地址不会被公开。 必填项已用*标注