Translator Disclaimer
11 February 2020 Efficiency of using active interference dedicated to medium range surveillance radar
Author Affiliations +
Proceedings Volume 11442, Radioelectronic Systems Conference 2019; 114420W (2020) https://doi.org/10.1117/12.2565241
Event: Radioelectronic Systems Conference 2019, 2019, Jachranka, Poland
Abstract
This paper provides an overview of the characteristics of selected types of active interference dedicated to medium range surveillance radars. That is why at the very beginning examples of platforms used to interfere the operations of both land based surveillance radars and radars mounted on boards of combat aircraft are characterized. The paper focuses mostly on surveillance radars and their resistance to noise jamming. An example calculation of effective jamming range in relation to a medium range radar is considered in the text. Moreover a graphic representation of the influence of jamming on the operation of a radar with reference to the application of jamming and thereof lack are discussed. The basic aim of the considerations is an efficiency assessment of the jamming impact on the operation of the radar. Among others the effectiveness of the radar transmit pulse power changing is assessed with regard to the prevention of intentional jamming. The conclusions provide exemplary methods of mitigating the impact of jamming on the operation of the radar. Additionally the paper provides information on priorities to be taken into consideration in the field of contemporary radiolocation.

1.

INTRODUCTION

The methods and techniques of active interference generation are rapidly developing in a frame of electronic warfare. Active jamming involves, among others, generation of an quasi-echo signal that adjusts to the frequency of the interfered radar. It is also crucial to ensure appropriate power of the jamming signal Pj compared to the power of the real echo signal, Prec. In particular, response jamming is generated with the use of a sensor (radar) probing signal. The jamming device analyses the probing signals of the radar in order to emulate a probing signal that is emitted – with appropriate delay – in the direction of the device whose operation is being distorted. As a result, several targets are displayed within one azimuth, thus imitating targets which are moving in the direction of the radar.

Another example comes from noise jamming whose main purpose is to distort the operation of CFAR systems (CFAR – Constant False Alarm Rate). Noise jamming does not require the knowledge of the radar probing signal. Noise generated with appropriate power in the direction of the radar causes the CFAR level to increase, at the same time masking echo’s from targets with a low RCS value (RCS – Radar Crosse Section). Noise jamming “covers up” tangible structures whose echo disappears in the noise.

Platforms which are used to carry the sources of jamming can be divided into ground-based and air-based ones. Examples of ground-based platforms used by NATO are as follows: Mini NEWVAN (Mini NATO Electronic Warfare Van), NEWVAN (NATO Electronic Warfare Van), Mini TRACSVAN TV (Mini Transportable Radar and Communications Jamming and Simulation Vans TV), TRACSVAN (Transportable Radar and Communications Jamming and Simulation Vans TV), MIJA (Mobile Intercept Jamming Assets), MRV (Mini Radar Van) [1]. They are used within NATO JEWCS (NATO Joint Electronic Warfare Core Staff). Air-based platforms are used to carry such pods as: AN/ALQ-167 (it comes in multiple models which differ in their operational frequencies and functions), AN/ALQ-99, AN/ALQ-131, AN/ALQ-184. Pods are mounted on airplanes in order to be used during air missions. A pod often can be reconfigured by an operator to work with a given type of radar. Jamming is the most effective if the distance from the radar is greater. Pods are often equipped with double transmitter systems, which allows them to generate jamming in both directions (front and back).

Several ground-based and air-based platforms is presented in Figure 1. The technological development of jamming pods describes trends of radiolocation development, and as a result, the need to upgrade today’s radars in order to counteract jamming [2].

Figure 1.

Examples platforms used to generate jamming: Mini NEWVAN, NEWVAN, Mini TRACSVAN (TV), TRACSVAN, MIJA, AN/ALQ-167 [1]

00481_psisdg11442_114420w_page_2_1.jpg

Pods which are equipped with signal analysis and generation systems are used to protect single aircraft or entire formations. The first pods were produced in 1970s, when long-range EF-111A Raven aircraft [3] and electronic warfare and radioelectronic combat EA-6B Prowler aircraft were first used. The latter is of such high quality that it is still used by the US Navy and USAF (United States Air Force) [4]. Aircraft equipped with electronic combat pods are used in every military operation conducted by the United States of America and NATO member states. They serve in the USAF as support used in COMAO (Composite Air Operations), where a total number of 60 different aircraft may be used. The presence of two EA-6B Prowlers allows them to neutralize the enemy’s defense systems, which in turn makes it possible for fighter-bombers to enter. USAF strives to replace numerous planes with a multi-task F/A-18E Super Hornet. They are determined to replace EA-6B Prowlers in the scope of electronic warfare. The version of Super Hornet used in electronic combat is marked as EA-18G Growler. These planes are equipped with ALQ-99 jamming pods [5].

An ALQ-99 pod carried by the EA-18G Growler is shown in Figure 2. In the front part of the pod there is a turbine which generates independent power supply for the pod. This system is not a burden to the aircraft during an operation (there is no need for a power generator for the pod). The pod is covered by a Radome which protects the antennae and the whole equipment against weather conditions such as rain or wind. EW elements used in the EA-18G Growler make it the most expensive aircraft manufactured in the USA. The plane is able to defend itself against other objects which operate in air. Such a highly advanced self-defense technology allows it to disarm other airplanes and ground-based exploratory devices and the pods allow to jam the enemy’s communication or other defense systems, as a result protecting the aircraft or the entire COMAO formation. Pods are formed in various combinations and frequency ranges. Several selected pods are described below.

Figure 2.

The structure of an ALQ-99 pod mounted under the aircraft for self-defense [6]

00481_psisdg11442_114420w_page_2_2.jpg

AN/ALQ-211 pod (SIRFC-Suite of Integrated RF Countermeasures) [7] is a part of the AN/ALQ family, a radio frequency countermeasure set. An integrated combat system provides protection against aircraft radio hazards. The pod is also equipped with a system which notifies the pilot of radiation. Such a system is utilised in F-16 aircraft which form a part of the Polish Air Force.

The ALQ-231 pod utilised by the US Marine Corps is a system which covers the network technology (architecture) and combines all defense systems in an aircraft (EWASA-Electronic Warfare Service Architecture) [8]. The system allows for pod control from the plane, but also by means of a safe tactile radio network, which is why it may also be used in unmanned aircraft. The visual side of the ALQ-231 pod is shown in Figure 3. The lack of a turbine requires external power supplied by the plane.

Figure 3.

An AN/ALQ-231 pod [8]

00481_psisdg11442_114420w_page_3_1.jpg

Figure 4.

The graph shows the correlation between the number of plots in the transmitter and the time of jamming activities for uninterrupted work and a correction of the Neyman-Pearson detection threshold with the application of 20 MHz noise jamming with square modulation amplitude (prepared by the authors)

00481_psisdg11442_114420w_page_7_1.jpg

2.

RADIOLOCATION JAMMING

2.1.

Types of jamming used in radiolocation

Radiolocation jamming may be categorised under various criteria. They are mainly divided into active and passive. The following list presents the categorisation of active radioelectronic jamming [9]:

  • with regard to the jamming bandwidth: selective, barrage, multipoint, jitter;

  • with regard to the structure of the jamming signal: modulated, manipulated, impulse, combined interference;

  • with regard to the impact range: close-range, long-range;

  • with regard to the duration of jamming: short-term, long-term;

  • with regard to the radiation of the jamming signal: directional, omnidirectional;

  • with regard to the type of radiation: weak, strong, very strong.

The above taxonomy is one of several categorisations used in Polish and foreign nomenclature and it was taken from The Air Force Electronic Warfare Manual[10]. The direction of development in NATO in the scope of EW is regulated by the document MC 0064 – NATO Electronic Warfare Policy[11]. It is developed and regularly updated by NEWAC (NATO Electronic Warfare Advisory Committee) and approved by NATO MC (NATO Military Committee).

The directions of development in the scope of EW are formulated on the basis of these documents in the member states of NATO, including the Polish Armed Forces.

Noise jamming is categorised with regard to the structure of the signal (broadband and narrowband), amplitude modulation and repeat time (pulse repetition interval). In this paper prohibitive noise jamming with 20MHz broadband is subject to an analysis. While the operating band of the jamming signal increasing, the loss of power must be taken into account [12]. Designing a broadband jamming device requires a very high level of device power supply. It is not an issue in ground-based devices, however, it becomes a huge drawback in aircraft-mounted devices [2].

2.2.

Effectiveness of the use of jamming

There are numerous ways of assessment of the effectiveness of radiolocation jamming cited in reference publications. The type of jamming signal and the characteristics of the interfered device are of significance here. The effectiveness is represented as the ratio of the interference power to the power of the useful signal of the interfered device [13].

00481_psisdg11442_114420w_page_4_1.jpg

where: Pj – input interference power of the radiolocation receiver, Ps – input useful signal power of the receiver, kd – degradation index of the radiolocation device [13,14,15]. The application of the above index allows to determine the power of the jamming signal at a given distance, RTRAj, required to ensure appropriate loss of radiolocation data with a set value of the probability of aircraft echo detection.

According to EW specialists the degradation index of a given radar ranges from 3 to 8. For impulse radars the value is closer to 8. The value of the factor depends on the parametres of the interfering device, the interfered radar and their respective dislocation. The following considerations present the correlations for an interfering device mounted on an interfering aircraft in order to cover another plane flying in the direction of the enemy’s radar.

The input power of the jamming radar signal can be determined in the following way [14]:

00481_psisdg11442_114420w_page_4_2.jpg

where:

Pj – interfering signal power density measured at the antenna input of the interfered radar 00481_psisdg11442_114420w_page_4_3.jpg;

Ar – effective surface of the antenna aperture of the interfered radar [m2];

F2(βj, εj) – function which describes normalised directional characteristics of the interfered radar;

γj – an index which determines the difference between the radar antenna polarisation and the interfering device

The power of self-noise created in the receiver of the interfered radar, Pn, is calculated with the use of the following correlation [16]:

00481_psisdg11442_114420w_page_4_4.jpg

where:

ρ – Boltzmann constant;

T – absolute temperature;

Nn – receiver noise index;

Δfrec – the width of the operating band of a receiver part in the interfered radar

The power of the jamming signal which reaches the radar is usually much higher than the power of the receiver noise, Pn, which is why it is omitted in the calculation in the majority of studies.

The power of the input useful signal of the radar receiver is described as follows [16]:

00481_psisdg11442_114420w_page_5_1.jpg

By using the above correlations in the equation (1), the quotation for radiolocation prevention looks as follows [14,15]:

00481_psisdg11442_114420w_page_5_2.jpg

where:

Pj – the power of the interfering transmitter, taking into account the losses in the transmission line [W];;

Ps – the power of the interfered transmitter, taking into account the losses in the transmission line [W];

Gj – directivity of the antenna of the interfering transmitter;

Gs – directivity of the antenna of the interfered device;

Δfj —spectrum width of the jamming signal [Hz];

Δfrec – spectrum width of the signal received by the interfering device receiver [Hz];

γj – an index which takes into account the difference between the polarization values of the antennae of the jamming transmitter and the interfered device;

α – an index which takes into account the reduction of the signal in the atmosphere;

Rj – the distance between the aircraft (interfering transmitter) and the interfered device [m];

Rrec – the distance between the covered aircraft and the interfered device [m];

σ0 – radar cross section [m2];

By using the equation (5) it is possible to determine the characteristics of electronic warfare pods and to calculate the jamming influence on a chosen type of radar. By knowing the parameters of the radar and the pod, the value of the Rrecmin distance is established – within a given distance jamming applied with a specified power is ineffective, which makes it possible to detect the aircraft which carries the pod and the assisting aircraft which was masked.

00481_psisdg11442_114420w_page_5_3.jpg

The equation (6) [17] is correct if the value α ≈ 0 of the electromagnetic wave suppression in the atmosphere is omitted and if k = kd. This correlation determines the minimum distance which proves that the radar jamming is ineffective.

3.

TEST RESULTS

3.1.

Examples of calculations of the impact of the change of interference power on effective radar range

The calculations were made with reference to medium range radars for the L band and S band and for example parameters of an jamming transmitter mounted on an aircraft with average radar cross section σo = 2 m2 [13, 14, 15].

Moreover, an assumption was made that the pod is mounted on the aircraft for self-defense purposes and the interfered radar is located on the ground. The atmosphere suppression value was disregarded. The following assumptions were made:

00481_psisdg11442_114420w_page_6_1.jpg

The following parameters values of the jamming transmitter were used:

Pj = 150 ÷ 1500 W;

Gj =21 dB;

Δfj = 5 MHz;

γj = 0,7

Radar 1 parameters (examples):

ps =0,3 MW;

Gs =30 dB;

Δfrec = 4 MHz

kd = 6,0

Radar 2 parameters (examples):

ps =1,4 MW;

Gs =30 dB;

Δfrec = 4 MHz

kd = 6,0

The calculations results concerning a minimum distance listed in [km] necessary to detect a jamming aircraft flying towards the radar are presented in the Table 1. Radar 2 has a considerably higher power than the first one, which allows for the possibility of an earlier detection of a jamming device in the radar space. The examples of calculations presented above highlight the importance of the power of emitted signal. A much larger range of the second radar allows the operator more time to react.

Table 1

The influence of the change in the power of the jamming device on effective jamming range

Pj [W]1503004506007509001050120013501500
Radar 1Rrecmin [km]70494035312926252322
Radar 2Rrecmin [km]1511078775676257535048

Additionally, the above considerations did not involve the atmosphere suppression value, which would only slightly influence the results. The change of the transmitter power which is visible here shows how the effective jamming range undergoes changes. A radar which emits a stronger impulse is able to detect a jamming transmitter almost twice as fast, which is confirmed by the results of calculations presented in the table.

3.2.

Checking the impact of jamming on a medium range radar

Numerous scenarios of electronic warfare are performed during different tests. The examples of scenarios performed while using active prohibitive noise jamming are shown in Figure 5 and Figure 7.

Figure 5.

Depiction of the radiolocation situation in normal operation without jamming and with the application of noise jamming (prepared by the authors).

00481_psisdg11442_114420w_page_7_2.jpg

Figure 6.

The graph shows the correlation between the number of plots in the transmitter and the time of jamming activities for a change in the detection threshold with the application of 20 MHz prohibitive jamming with jitter modulation amplitude (prepared by the authors)

00481_psisdg11442_114420w_page_8_1.jpg

Figure 7.

Depiction on the radar display with an incorrectly selected Neyman-Pearson detection threshold (prepared by the authors).

00481_psisdg11442_114420w_page_8_2.jpg

The first stage of the test focused on prohibitive noise jamming with square modulation amplitude. The number of received plots are shown in Figure 4. Three types of operation were analysed: normal operation, work with jamming elements and an attempt to mitigate the results of jamming. A 20MHz band jamming is powerful, which results in a strong impact on the radar. The green colour indicates results of the radar operation without any impact of the jamming activities. A constant number of plots (detections) is visible in the receiver block. Red crosses and points indicate the number of plots during the period of jamming impact. A loss of detections is visible, which proves the fact that the CFAR threshold has increased significantly. Objects with a low RCS value are covered with noise in the receiver. Seeing a loss of plots on the display for aircraft flying on specific routes, the operator adjusts the value of the Neyman-Pearson detection threshold. Lowering the threshold results in the increase of detections, which has been indicated with red points (Figure 4). Lowering the detection threshold too much leads to detection of noise, which in turn results in a very high number of false detections.

If the Neyman-Pearson detection threshold is set in a correct way, the difference in echo detection is noticeable. Too low threshold value results in detection of lost plots and, additionally, in detection of a high number of random plots which are not real objects.

The display of a medium range radar is presented in Figure 5. The green colour shows current detections, while the colour white indicates pulses (histories) from the previous rotation of the antenna. The blue colour shows a tracked route. The right side of Figure no. 3 shows the result of the active jamming on the radar. Yellow arrows indicate losses of plots of schedule objects (with a high ERS value). Also visible is the loss of other plots generated around the radar. The red colour indicates repeated information from objects, collected with the help of the IFF system (Identification Friend or Foe). If the IFF system fails to identify an element, the operator loses their ability to track the objects.

Lowering the value of the threshold to 30 (while the value for normal operation conditions is 100) results in a fourfold increase of the number of detected plots, which is shown in Figure 6. The majority of detections are false plots, not connected to real objects. Detections can be randomly observed in various azimuths around the radar as they are generated in all directions in large groups of plots and mask real objects.

Correctly selected detection threshold depends on numerous factors in the radar operation. Weather conditions and the type of jamming applied disruption all influence the correct setting of the threshold. Operation in normal conditions requires the range of the threshold of 100-150 range. Lowering the threshold results in an increased sensitivity of the receiver. The appropriate selection of the detection threshold is crucial in jamming combat. Additionally, the operator is able to control the range threshold (RTC – Range Threshold Control). In the cases presented in the paper the value of RTC has not been adjusted in order to show the impact of the change in the value of the Neyman-Pearson detection threshold. The characteristics of RTC are strictly determined by the manufacturer of a given device, which allows for its adjustment through changing the index of a characteristic feature corresponding to its shape. Adjustment of the value of the index removes the excessive number of plots generated in close proximity of the device, usually within a 70 km radius from the site of the radar.

Appropriate selection of the index is crucial in combat operation of the device, as it allows to avoid sending false plots to superior officers working on the basis of radiolocation information.

The self-disguise effect of an aircraft which uses a pod mounted on the board is shown in Figure 8. Purple shading informs us on the direction of the emitted jamming. Identification is possible only with the application of the IFF system (Identification Friend or Foe). The object was marked with a yellow arrow. Only detection in the secondary system (red squares) is possible. If the radar is not equipped with the IFF system, an aircraft remains invisible until it crosses the Romin distance, where jamming is ineffective. An experienced operator with the knowledge about a jamming emitter within a given azimuth can undertake actions in order to ensure that the jamming is deliberate. The easiest way of establishing deliberate jamming is to observe neighbouring aircraft in order to determine the loss of plots in a given azimuth. The loss of real objects occurs in the scope of activity of a jamming emitter.

Figure 8.

Depiction of a case of failed detection of a plane which uses a pod for self-defense purposes.

00481_psisdg11442_114420w_page_9_1.jpg

3.3.

Analysis of the results

If a jamming device generates a noise signal, the average level of noises increases in the radar receiver. Therefore, in order to “produce” a relatively constant number of false plots (do not overload the receiver with a very high number of false detections), the CFAR system must increase the decision threshold [14]. At this time lower echoes, which in normal conditions would pass the CFAR threshold, will be filtered out and not admitted for further signal processing (they will be masked). Echoes with a high RCS value which pass the threshold will be processed further [18, 19, 20].

The effectiveness of changes in the frequency of the device operation was also verified during the tests. Changes in the frequency caused an escape from the interfered frequency, which resulted in operation similar to normal conditions. The effect of jamming performed with the use of a device located approx. 1 km from the interfered radar is presented in Figure 5. If a device (POD’s) is located within a short distance, it is possible to determine the direction of the emitted jamming signal. It is a very significant piece of information needed in order to locate the jamming emitter and to remove it accordingly.

4.

CONCLUSIONS

An important aspect of contemporary radiolocation is the development of defense techniques against radioelectronic jamming. The use of ever newer methods of disrupting surveillance radars causes difficulties in fighting them. Although the new radars have many possibilities of interference in the transmitter parameters such as: Neyman-Pearson detection threshold, RTC, transmitter power control, change of duration of the probing impulse, change of refresh time of the radiolocation information etc. are considered in the paper. Appropriate application of the radar’s capabilities offers a high chance of success in the process of jamming defense. An important element in EW is the level of training received by the appropriately trained staff. A well-trained operator is able to reconfigure the radar in an appropriate way to counteract intentional jamming. The training should focus on practical usage of a radar in the presence of intentional jamming. The knowledge of methods and techniques of counteracting intentional jamming guarantees approximately a 70% success rate of defense. The characteristics of interference and their influence on the operation of the radar should be sent to the remaining users of the air defense SYSTEM, as the forwarded information ensures that the jamming emitter is detected much quicker. If the direction of interference changes the remaining devices which operate in the air defense SYSTEM can easily hold off the attack. That is why the main priority in the contemporary radiolocation is practical training of staff members during allied forces training sessions. The number of practical trainings results in the increase of staff experience and further success in dealing with intentional jamming.

Publication financed by the Military University of Technology in 2019 under the PBS 659 project.

REFERENCES

[2] 

Jędrusik P., Pietrasiński J., “Properties of aircraft pods in the aspect of the electronic warfare,” Electronic nr 6, SIGMA-NOT, (2019). Google Scholar

[3] 

F-111 Aardvark - USAF’s Ultimate Strike Aircraft, Tony Thornborough, Osprey Aerospace, (1993). Google Scholar

[4] 

Gordon Swanborough and Peter M. Bowers, Smithsonian, United States Military Aircraft Since, (1989). Google Scholar

[5] 

“Electronic Warfare and Radar System Engineering Handbook,” (2013). Google Scholar

[9] 

EW in Polish Air Force, “Manual Edition Air Force Command, Recognition and EW Management,” Warsaw,2010). Google Scholar

[12] 

David L. Adamy, Artech house, “EW 101 A First Course in Electronic Warfare,” Boston, London (2000). Google Scholar

[13] 

Leroy B., “Applied ECM. Vol. I, II, III. EW Engineering, Inc P.O. Box 28, Duun Lorgin, VA22027, USA,” (1991). Google Scholar

[14] 

Matuszewski J., “Evaluation of jamming efficiency for the protection of a single ground object,” in 2017 Radioelectronic Systems Conference, Jachranka, Poland, Nov 14-16, 2017, Proc. of SPIE, 107150B (2018). https://doi.org/10.1117/12.2316629 Google Scholar

[15] 

Matuszewski J., “Jamming Efficiency of Land-Based Radars by the Airborne Jammers,” in MIKON 2018 - 22nd International Microwave and Radar Conference, Poznań, 324 –327 (2018). https://doi.org/10.23919/MIKON.2018.8405214 Google Scholar

[16] 

Merrill Skolnik, “Radar Handbook,” ThirdMcGraw-Hill,2008). Google Scholar

[17] 

David L. Adamy, Artech house, “EW 102 A Second Course in Electronic Warfare,” Boston, London (2004). Google Scholar

[18] 

U.S. Department of Defense, “Electronic Warfare Fundamentals,” (2014). Google Scholar

[19] 

Naval Air Warfare Center Weapons Division, “Electronic Warfare and Radar Systems Engineering Handbook,” FourthPoint Mugu, California (2013). Google Scholar

[20] 

Welch M., Pywell M., RTO AGARograph 300 Flight Test Technique Series, “Electronic Warfare Test and Evaluation,” 28 (2012). Google Scholar
© (2020) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Piotr Jędrusik and Jerzy Pietrasiński "Efficiency of using active interference dedicated to medium range surveillance radar", Proc. SPIE 11442, Radioelectronic Systems Conference 2019, 114420W (11 February 2020); https://doi.org/10.1117/12.2565241
PROCEEDINGS
11 PAGES


SHARE
Advertisement
Advertisement
RELATED CONTENT


Back to Top