Method for determining jamming characteristics

Abstract

A method (200) for determining one or more characteristics of a potential jamming signal. The method comprises obtaining (201), from a receiver monitoring a frequency band (Br), input regarding a potential jamming signal. The method comprises estimating (202), based on the input, when the potential jamming signal is present within the monitored frequency band and when the potential jamming signal is not present within the monitored frequency band. The method comprises determining (203) a duration of a first time span for when the potential jamming signal is present within the monitored frequency band. The method comprises determining (204) a duration of a second time span for when the potential jamming signal is not present within the monitored frequency band. The method comprises determining (205) one or more characteristics of the potential jamming signal based on the durations of the first and second time spans.

Description

[0001] METHOD FOR DETERMINING JAMMING CHARACTERISTICS

[0002] TECHNICAL FIELD

[0003] The present disclosure relates to techniques for determining one or more characteristics of a potential jamming signal.

[0004] BACKGROUND

[0005] Wireless communication has grown rapidly, both in terms of number of users and carried data volume, over the last few decades and is expected to continue to grow as Fifth Generation (5G) is being deployed. In addition to consumer mobile broadband (MBB) applications, a wide range of business and mission-critical applications (such as, public safety communication, smart manufacturing and industrial automation) are starting to emerge with 5G new radio (NR) access network deployments. Many of these applications demand fast, reliable and secure communications. Disturbances or outage in such type of communication can have catastrophic consequences.

[0006] Of the new services, smart manufacturing is developing rapidly. For instance, Worcester Bosch factory in the UK has launched a 5G private network infrastructure that is provided by Ericsson and manned by British Telecom (Robert Bosch UK, "Leading in 5G Technology," 2020. [Online]. Available: https: / / www.bosch.co.uk / intemet-of-things / leading-in-5g-technology / ). Schneider Electric and Orange have jointly conducted trials of a 5G indoor private network in the Le Vaudreuil factory on experimental frequencies (Orange, "Orange and Schneider Electric Run Industrial 5G trials in French Factory," 28 September 2020. [Online], Available: https: / / www.orange.com / en / newsroom / press-releases / 2020 / orange-and-schneider-electric- run-industrial-5g-trials-french-factory). Such connectivity systems may (as most radio-based connectivity systems) face different kinds of attacks that lead to application service interruption, including communication denial, for example, cloning of SIM cards, tracking of users, DNS spoofing, denial-of-service (DoS) attacks, and injection of manipulated broadcast messages with adequate power to overshadow the legitimate ones. Radio jamming has been identified as a potential threat.

[0007] Although the use of communication jammers is prohibited by law in most countries, wireless systems are still vulnerable to malicious electromagnetic threats from unlawful activity, primarily due to the broadcast nature of the wireless medium. Nowadays, commercial off-the- shelf (COTS) radio jammers may be bought by any individual from a variety of internet sites and may used to target the above-mentioned critical applications. These jammers typically do not require any skill to operate. Measurements show that many COTS jammers are chirp jammers (T. Kraus, R. Bauemfeind, and B. Eissfeller, “Survey of In-Car Jammers - Analysis and Modeling of the RF signals and IF samples (suitable for active signal cancellation),” in ION GNSS, Portland, Oregon, USA, 2011) and (R. H. Mitch, R. C. Dougherty, M. L. Psiaki, S. P. Powell, B. W. O'Hanlon, J. A. Bhatti, and T. E. Humphreys, “Signal Characteristics of Civil GPS Jammers,” in ION GNSS, Portland, Oregon, USA, 2011).

[0008] Moreover, one may also lawfully buy components on the market, e.g., software-defined radio (SDR) kits intended for research and assemble and use them in an unlawful manner to harm wireless communication.

[0009] Reduction of the transmission rate, as a result of a jamming attack, may increase the end-to- end latency of communication links. This may cause an anticipated message to be delayed, so that it is not delivered in the expected timebound in the context of industrial applications or public safety alerts utilizing the wireless connectivity. Jamming attacks may lead to increased number of retransmissions. Retransmissions increase the latency. Increased latency may not be acceptable, especially for applications relying on ultra-reliable low-latency communication, as the electromagnetic threats on the spectrum may lead to significant financial losses via production damage.

[0010] Consider a scenario with a radio access network, e.g., 2G, 3G, 4G, 5G, or Wi-Fi or Low-Power Wide Area Network (LPWAN), in operation. For performance reasons and / or safety reasons and / or concern for the equipments that are using the network, early detection and classification of electro-magnetic threats (EMTs) may be important to mitigate the impact of jamming attacks on the communications performance.

[0011] Existing solutions for detection of electromagnetic attacks are often highly specialized, expensive stand-alone systems that do not lend themselves to easy integration with the communication system. Furthermore, the detection methods described in the literature often employ spectrum analysis techniques that are computationally burdensome.

[0012] SUMMARY According to a first aspect, there is provided a method for determining one or more characteristics of a potential jamming signal. The method comprises obtaining, from a receiver monitoring a frequency band, input regarding a potential jamming signal. The method comprises estimating, based on the input, when the potential jamming signal is present within the monitored frequency band and when the potential jamming signal is not present within the monitored frequency band. The method comprises determining a duration of a first time span for when the potential jamming signal is present within the monitored frequency band. The method comprises determining a duration of a second time span for when the potential jamming signal is not present within the monitored frequency band. The method comprises determining one or more characteristics of the potential jamming signal based on the durations of the first and second time spans.

[0013] In an embodiment according to the first aspect, the method further comprises determining a duration of a third time span for when the potential jamming signal is present within the monitored frequency band, and wherein the determination of the one or more characteristics of the potential jamming signal is further based on the duration of the third time span.

[0014] In an embodiment according to the first aspect, the first time span and the third time span correspond to an odd-numbered time span and an even-numbered time span respectively, in a consecutive enumeration of time spans during which the potential jamming signal is present in the monitored frequency band.

[0015] In an embodiment according to the first aspect, the method further comprises determining an average of the durations of odd-numbered time spans during which the potential jamming signal is present in the monitored frequency band; and determining an average of the durations of even-numbered time spans during which the potential jamming signal is present in the monitored frequency band, wherein the determination of the one or more characteristics of the potential jamming signal is based on the determined averages.

[0016] In an embodiment according to the first aspect, the method further comprises determining a duration of a fourth time span for when the potential jamming signal is not present within the monitored frequency band, and wherein the determination of the one or more characteristics of the potential jamming signal is further based on the duration of the fourth time span.

[0017] In an embodiment according to the first aspect, the second time span and the fourth time span correspond to an odd-numbered time span and an even-numbered time span respectively, in a consecutive enumeration of time spans during which the potential jamming signal is not present in the monitored frequency band.

[0018] In an embodiment according to the first aspect, the method further comprises determining an average of the durations of odd-numbered time spans during which the potential jamming signal is not present in the monitored frequency band; and determining an average of the durations of even-numbered time spans during which the potential jamming signal is not present in the monitored frequency band, wherein the determination of the one or more characteristics of the potential jamming signal is based on the determined averages.

[0019] In an embodiment according to the first aspect, the method further comprises determining a signal pattern of the potential jamming signal based on at least the durations of the following: the first time span, the second time span, the third time span and the fourth time span.

[0020] In an embodiment according to the first aspect, the signal pattern is at least one of: a triangular chirp, a saw-tooth chirp, a sinusoidal chirp and an exponential chirp.

[0021] In an embodiment according to the first aspect, the method further comprises determining, based on the estimation, that the potential jamming signal is not a jamming signal.

[0022] In an embodiment according to the first aspect, the estimating is based on an intra-set variance minimizing algorithm.

[0023] In an embodiment according to the first aspect, the estimating is based on a peak detection or threshold-based algorithm.

[0024] In an embodiment according to the first aspect, the estimating includes performing filtering of the input or of an intermediate result obtained based on the input.

[0025] In an embodiment according to the first aspect, the input comprises squared magnitude of samples of the potential jamming signal.

[0026] In an embodiment according to the first aspect, the one or more characteristics of the potential jamming signal includes at least one of: center frequency of the potential jamming signal, bandwidth of the potential jamming signal, period of the potential jamming signal.

[0027] In an embodiment according to the first aspect, the method is performed by Radio Access Network, RAN, equipment or a RAN node. According to a second aspect there is provided a device for determining one or more characteristics of a potential jamming signal. The device comprises a memory and a processor, the memory containing instructions which when executed on the processor, cause the device to: obtain, from a receiver monitoring a frequency band, input regarding a potential jamming signal; estimate, based on the input, when the potential jamming signal is present within the monitored frequency band and when the potential jamming signal is not present within the monitored frequency band; determine a duration of a first time span for when the potential jamming signal is present within the monitored frequency band; determine a duration of a second time span for when the potential jamming signal is not present within the monitored frequency band; and determine one or more characteristics of the potential jamming signal based on the durations of the first and second time spans.

[0028] In an embodiment according to the second aspect, the memory containing instructions which when executed on the processor, cause the device to perform the method according to one or more embodiments of the first aspect.

[0029] According to a third aspect there is provided a computer program comprising instructions which, when executed on the device, cause the device to carry out the method according to one or more embodiments of the first aspect.

[0030] According to a fourth aspect there is provided a computer program product comprising a computer readable storage means on which the computer program according to the third aspect is stored.

[0031] BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 illustrates a jammer interfering with wireless communication according to one or more embodiments.

[0033] Figure 2 illustrates an example sawtooth signal pattern.

[0034] Figure 3a illustrates a method according to one or more embodiments.

[0035] Figure 3b illustrates optional method steps for the method in Figure 3a, according to one or more embodiments.

[0036] Figure 4 illustrates an example triangular signal pattern. Figure 5 illustrates an example sinusoidal signal pattern.

[0037] Figure 6 illustrates an example exponential signal pattern.

[0038] Figure 7 illustrates an RC circuit with a common resistor for charge and recharge, used for generating the exponential signal pattern of Figure 6 according to one or more embodiments.

[0039] Figure 8 illustrates an RC circuit with different resistors for charge and discharge, used for generating the exponential signal pattern of Figure 6 according to one or more embodiments.

[0040] Figure 9 illustrates a method for determining one or more characteristics of a potential jamming signal, according to one or more embodiments.

[0041] Figure 10 illustrates an implementation of a device according to one or more embodiments.

[0042] Figure 11 illustrates another implementation of a device according to one or more embodiments.

[0043] Figure 12 illustrates a device for determining one or more characteristics of a potential jamming signal, according to one or more embodiments.

[0044] DETAILED DESCRIPTON

[0045] The COTS jammers, mentioned in the background section above, may be referred to as barrage jammers meaning that the jammer transmits a jamming signal in a certain frequency band without any deeper knowledge about the structure of the signals to be jammed. Barrage jamming may be realized by transmitting noise, a single tone, a comb-like pattern of multiple tones or a chirp. A chirp is a signal in which the frequency increases (up-chirp) or decreases (down-chirp) with time. The increase and decrease of the instantaneous frequency are repeated according to some pattern. The chirp jammer is advantageous for an attacker in the sense that it may generate sufficient power over a large enough bandwidth to adversely impact the targeted communication system.

[0046] Cellular networks feature adaptive modulation and coding, whereby the transmission rate is dynamically changed to match the quality of the propagation channel between a transmitter and a receiver. Such a feature may handle the inherent propagation fading effects in a wireless medium and, to some extent, the interference that exists in the channels (for example, cochannel interference from other cells), by means of adaptive modulation and coding. However, reducing the transmission rate may increase the end-to-end latency of communication links, which causes the anticipated message to not be delivered in the expected timebound in the context of industrial applications or public safety alerts utilizing the wireless connectivity. Increased number of retransmissions could be a result of a jamming attack. Retransmissions increase the latency which may not be acceptable, especially for applications relying on ultrareliable low-latency communication, as the electromagnetic threats on the spectrum may lead to significant financial losses via production damage.

[0047] Figure 1 illustrates an example scenario of a jamming attack carried out in a communication network 100. Consider a UE 110 transmitting a signal 111 to a device 120 (for example a radio access network node) in the communication network 100. The UE 110 may be transmitting the signal 111 at a particular frequency or in a particular frequency band. Now consider a jamming device such as the jammer 130 that intends to maliciously interfere with the ongoing communication between the UE 110 and the device 120 and thus transmits a signal 131 to cause the interference. Such interfering signal 131 may herein be referred to as a jamming signal. The jammer 130 typically transmits a jamming signal 131 in a certain frequency band without any deeper knowledge about the structure of the signals to be jammed. Some examples of such jamming signals 131 include a noise signal, a single tone, a comb-like pattern of multiple tones or a chirp (sometimes also referred to as a sweep signal).

[0048] The jammer 130 transmitting the jamming signal 131 may generate sufficient power over a large enough bandwidth to adversely impact the device 120 or the communication between the device 120 and the UE 110. Thus, to avoid any adverse impacts on the device 120 or on the communication between the UE 110 and the device 120, it is beneficial to detect the presence of the jamming signal 131 and / or understand one or more characteristics of the jamming signal 131.

[0049] It is proposed herein to use a receiver 121, preferably a wide-band receiver 121, to detect the presence of the jamming signal 131 and thereby detect the presence of the jammer 130 in the communication network 100. As indicated in Figure 1, the receiver 121 may for example be comprised in the device 120, or the device 120 may employ an external receiver 121 to detect the jamming signal 131. At least some embodiments proposed herein exploit knowledge of signal structure of the jamming signal 131 to establish that the detected signal is indeed a jamming signal. One or more parameters of the detected signal may be used to determine the characteristics of the jamming signal 131, such as bandwidth of the jamming signal 131, center frequency of the jamming signal 131, and / or sweep period of the jamming signal 131, as will be described later in the application. For example, a chirp jamming signal (i.e., a jamming signal sweeping across a frequency band) with a bandwidth exceeding the bandwidth of the receiver 121 may give rise to a train of pulses at the receiver 121. The duration and relative timing of the pulses may depend on the chirp jamming signal as well as the parameters of the receiver 121. For a chirp jamming signal having up-chirps and down-chirps, the pulses detected at the receiver 121 may be divided into two subsets, one including even-numbered pulses and the other including odd-numbered pulses. These two subsets may correspond to up-chirps and down-chirps, respectively. The average of the pulse durations in the two sub-sets may be determined using the information of the detected chirp signal and the receiver 121. These average durations may then be used to determine properties / characteristics of the chirp signal, for example including which chirp signal sweep pattern (e.g., triangular, sawtooth, sinusoidal, and exponential) was detected as the pulses at the receiver 121.

[0050] In this way, at least some embodiments proposed herein provide an efficient method for detecting and determining the presence of a jammer 130 in a communication network 100.

[0051] Figure 2 shows a graph of an example jamming signal having a sawtooth chirp signal pattern. As may be seen in Figure 2, the horizontal axis represents time and the vertical axis represents frequency. The chirp signal is a signal in which the frequency increases with time, i.e. up-chirp, and / or decreases with time, i.e. down-chirp. The increase and / or decrease of the instantaneous frequency is repeated according to some pattern. The sawtooth chirp signal sweeps through frequencies in a proportional manner with time (either proportionally increasing with time or proportionally decreasing with time). In a sawtooth chirp signal comprising only up-chirps, there is a pattern of signals which have an increasing frequency with time. Similarly, in a sawtooth chirp signal comprising only down-chirps, there is a pattern of signals which have a decreasing frequency with time. A sawtooth chirp signal pattern comprising up-chirps is illustrated in Figure 2 as an example of a jamming signal (or a potential jamming signal).

[0052] Consider a scenario where the sawtooth chirp signal in Figure 2 is employed as jamming signal 131 to interfere with the communication of the device 120 in the communication network 100 shown in Figure 1. Assume that the receiver 121 monitors a frequency band having a bandwidth Br with a center frequency fcr. The frequency band monitored by the receiver 121 is herein referred to as ‘monitored frequency band’. Let fh be the highest frequency of the monitored frequency band and fl be the lowest frequency of the monitored frequency band. The monitored frequency band may, for example, have a bandwidth in the range of megahertz.

[0053] The sawtooth chirp signal sweeps through at least some portion of the monitored frequency band to cause jamming in the network 100. In other words, at least a portion of the bandwidth of the sawtooth chirp signal overlaps with the monitored frequency band.

[0054] The receiver 121 is capable of detecting a signal transmitted with a frequency falling within the monitored frequency band (including fh and fl). As a result, the receiver 121 detects the sawtooth chirp signal when the sawtooth chirp signal sweeps through the monitored frequency band. The receiver 121 detects an increased energy when the sawtooth signal chirp signal is present in the monitored frequency band. The receiver 121 may detect an energy pulse for a certain duration of time within the monitored frequency band indicating the presence of the sawtooth chirp signal within the monitored frequency band. Let til be the time duration for which the sawtooth chirp signal is present within the monitored frequency band.

[0055] Further, the sawtooth chirp signal may sweep through frequencies outside of the monitored frequency band, for a certain duration of time. During this time, the receiver 121 may not detect increased energy or an energy pulse within the monitored frequency band, indicating absence of the sawtooth chirp signal in the monitored frequency band. In other words, the receiver 121 may detect lesser energy in the monitored frequency band compared to when the sawtooth chirp signal was present in the monitored frequency band. Let tol be the time duration for which the sawtooth chirp signal is not present within the monitored frequency band.

[0056] The knowledge of the energy transitions due to the sawtooth chirp signal sweeping within the monitored frequency band to sweeping outside of the monitored frequency band (or vice versa) may be exploited to determine the time durations til and tol. For example, the start time and the end time of the energy pulse may be obtained and the time duration for which the sawtooth chirp signal is present (or not present) in the monitored frequency band may be determined.

[0057] A simple example of how to determine the durations tol and til has been described above. However, as will be described further below, more sophisticated methods may be used to deal with the fact that communication signals 111 may typically coexist with the jamming signal 131 in the communication system 100. Such coexistence may make it more difficult to detect when the jamming signal 131 enters and exists the monitored frequency band, since energy detected in the monitored frequency band could originate from legitimate communication 111. Also, there may be noise as well as other interference sources than the jamming signal 131 in the monitored frequency band.

[0058] Based on the determined time durations til and tol and the knowledge of the receiver 121 ’s frequency parameters (such as Br, fh, fl, fcr), one or more characteristics of the sawtooth chirp signal may be determined, as will now be explained. The characteristics of the sawtooth chirp signal may, for example, be the period, bandwidth and center frequency of the sawtooth chirp signal. Let the sawtooth chirp signal have a time period tp, bandwidth Bj and a center frequency fcj as shown in Figure 2.

[0059] The receiver 121 bandwidth Br is given by

[0060] Br = fh - fl. (1)

[0061] The center frequency of the frequency band monitored by the receiver 121 is given by

[0062] Typically, the parameters Br and fcr may already be known or pre-defined for a receiver 121 that is used for monitoring the presence of a jamming signal (or a potential jamming signal).

[0063] The sawtooth chirp signal shown in Figure 2 only has up-chirps, and jumps back down in frequency without passing through the monitored frequency band. As a result, the sawtooth chirp signal is only detected by the receiver 121 on its way upwards through the monitored frequency band. The period tp of the sawtooth chirp signal may be determined as sum of the time durations til and tol determined above, tp = til + tol (3) where + denotes a summation operation.

[0064] The receiver 121 bandwidth Br, the period tp and the time duration ti 1 may be used to determine the bandwidth Bj of the sawtooth chirp signal as where * denotes a multiplication operation.

[0065] Furthermore, the bandwidth Bj of the sawtooth chirp signal overlaps in part with the bandwidth Br of the receiver 121 where the jamming takes place. Thus, the center frequency fcj of the sawtooth chirp signal may be determined as being within the range (5)

[0066] If Bj is much larger than Br, the term Br may be omitted to obtain an approximate determination of fcj as being within the range, (6)

[0067] In this way, one or more characteristics of the sawtooth chirp signal in Figure 2 may be determined.

[0068] Figure 3a illustrates a flowchart of a method 200 for determining one or more characteristics of a potential jamming signal. The method 200 may for example be performed by the device

[0069] 120 in Figure 1, using the receiver 121. While the reference numbers from Figure 1 will be used in the description of the method 200, it will be appreciated that Figure 1 only serves as an example, and that the method 200 need not necessarily be performed by the device 120. The method 200 comprises several steps, some of which steps are optional.

[0070] The method 200 comprises obtaining 201 , from a receiver 121 monitoring a frequency band, input regarding a potential jamming signal. The receiver 121 may, for example, be monitoring a frequency band over which a UE 110 is communicating with a network 100. The receiver

[0071] 121 may, for example, be comprised in the device 120. As another example, the receiver 121 and the device 120 may be comprised in a communication system 100, but the receiver 121 may be arranged separately from the device 110. The receiver 121 may detect presence of a signal at the monitored frequency band. This detected signal is referred to as a ‘potential jamming signal’ since it has not yet been determined if the detected signal is indeed a jamming signal. A jammer 130 intending to interfere with the communication of the UE 110 and the network 100 may transmit a jamming signal 131 that is now detected by the receiver 121 as the potential jamming signal. The receiver 121 may thus detect the potential jamming signal and provide input regarding the potential jamming signal.

[0072] In an embodiment, the potential jamming signal is the sawtooth signal described above in relation to Figure 2.

[0073] The receiver 121 may provide the input to the device 120. The input may, for example, be obtained at the device 120. The input may, for example, be obtained at a communication system 100 comprising the device 120. The input obtained from the receiver 121 at step 201 may, for example, comprise samples of the potential jamming signal at the monitored frequency band. The input obtained from the receiver 121 may, for example, comprise samples of energy detected at the monitored frequency band, such as the energy of the potential jamming signal at the monitored frequency band. As another example, the obtained input may be an indication that a potential jamming signal is detected (or not detected) within the monitored frequency band.

[0074] The potential jamming signal may sweep the monitored frequency band for a certain duration on a periodic basis. This is a typical feature of a chirp jammer which may be exploited to determine or estimate the presence of a jamming signal.

[0075] The method 200 comprises estimating 202, based on the input obtained at step 201, when the potential jamming signal is present within the monitored frequency band and when the potential jamming signal is not present within the monitored frequency band. The input provided by the receiver 121 at step 201 may be used to estimate when the potential jamming signal is present within the frequency band monitored by the receiver 121 (herein referred to as ‘monitored frequency band’) and when the potential jamming signal is not present within the monitored frequency band.

[0076] In one or more embodiments, the input obtained at step 201 may comprise a sequence of zeros and ones over time, wherein a zero (in other words, a ‘0’) indicates the absence of the potential jamming signal within the monitored frequency band at that time instant and a one (in other words, a ‘1’) indicates the presence of the potential jamming signal within the monitored frequency band at that time instant.

[0077] In one or more embodiments, energy samples (or amplitude samples) of the detected potential jamming signal are provided as the input at step 201. The estimation at step 202 may include estimating an increased energy in the monitored frequency band due to the presence of the potential jamming signal.

[0078] The estimation at step 202 may, for example, include dividing the input into two sets. The first set is a zero-hypothesis set (HO) comprising those one or more inputs corresponding to the potential jamming signal not being present in the monitored frequency band. The second set is a one-hypothesis set (Hl) comprising those one or more inputs corresponding to the potential jamming signal being present in the monitored frequency band. An intra-set variance minimizing algorithm such as Otsu’s method may be used to perform the estimation at 202, as will be described later in the application. As another example, a peak detection / threshold technique such as the z-test method may be used to perform the estimation at step 202, as will be described later in the application.

[0079] Once it has been estimated at step 202 when the potential jamming signal is present or not within the monitored frequency band, it may be needed to understand the characteristics and / or signal patterns of the potential jamming signal to get a better understanding of the type of jammer being used. To achieve this, a next step would be to determine the time duration for which the potential jamming signal sweeps the monitored frequency band and the time duration for which the potential jamming signal does not sweep the monitored frequency band.

[0080] The method 200 comprises determining 203 a duration of a first time span for when the potential jamming signal is present within the monitored frequency band. The duration of the first time span may for example be til as described above in relation to Figure 2. The duration of the first time span may correspond to the duration of the time spent by the potential jamming signal in the monitored frequency band before exiting the monitored frequency band. In other words, the duration of the first time span may correspond to time duration of the potential jamming signal being present in the receiver 121 bandwidth. The potential jamming signal may sweep up or down in frequency within a single time period. So, the duration of the first time span may correspond to the duration of the time spent by, for example, the up-chirp of the potential jamming signal in the monitored frequency band as exemplified by til in Figure 2.

[0081] The receiver 121 may for example detect the potential jamming signal within the monitored frequency band based on an increased energy in the monitored frequency band during the duration of the first time span. Further, the receiver 121 may detect that the energy decreases when the potential jamming signal sweeps past and outside of the monitored frequency band, after the duration of the first time span referred to at step 203 (and exemplified by til in Figure 2).

[0082] The start time and the end time of the energy pulse within the monitored frequency band may be used to determine the duration of the first time span. As an example, in case of dividing the input into HO and Hl sets, the Hl set includes multiple occurrences of detections of the potential jamming signal within the monitored frequency band. The duration of the first time span may correspond to one part of the Hl set within a single time period. The length (or duration) of the occurrences of detections, within the Hl set, of the potential jamming signals within the monitored frequency band, may be used to determine the duration of the first time span til . The energy transition from the potential jamming signal sweeping the monitored frequency band to the potential jamming signal sweeping outside of the monitored frequency band (or vice versa) may be used to determine the duration of the first time span. For example, the time instances of transition from HO set to Hl set and from Hl set to HO set may be used to determine the duration of the first time span.

[0083] The method 200 comprises determining 204 a duration of a second time span for when the potential jamming signal is not present within the monitored frequency band. The duration of the second time span may for example be tol described above in relation to Figure 2. The duration of the second time span may correspond to the duration of the time when the potential jamming signal is not present in the monitored frequency band, possibly after a first detection of the potential jamming signal by the receiver 121. In other words, the duration of the second time span may correspond to time duration of the potential jamming signal not being present in the receiver 121 bandwidth. The duration of the second time span may correspond to the duration of the time spent by, for example, the up-chirp of the potential jamming signal outside the monitored frequency band as exemplified by tol in Figure 2.

[0084] The method 200 comprises determining 205 one or more characteristics of the potential jamming signal based on the durations of the first and second time spans determined at steps 203 and 204. The duration of the first time span and the second time span may, for example, be used to determine one or more characteristics of the potential jamming signal such as center frequency, bandwidth and / or period. As exemplified above in equations 3-6, it is possible to determine the bandwidth Bj and period tp of the sawtooth shaped jamming signal in Figure 2 using a first duration ti 1 and a second duration to 1 , and it is also possible to get a rough estimate of where the center frequency fcj of the jamming signal may be located using the first duration til and the second duration tol. As will be described further below, by determining characteristics of the potential jamming signal, one may also identify the signal pattern and / or type of the potential jamming signal, at least in some scenarios. The potential jamming signal may, for example, have a chirp signal pattern of the type sawtooth as shown in Figure 2. As another example, the potential jamming signal may have a triangular chirp signal pattern as shown in Figure 4. As still another example, the potential jamming signal may have a sinusoid chirp signal pattern as shown in Figure 5. As yet another example, the potential jamming signal may have an exponential chirp signal pattern as shown in Figure 6. As exemplified above with reference to Figure 2, it may be sufficient to determine the first and the second time spans (as in steps 203-204) in order to determine certain characteristics of the different types of the potential jamming signal, for example in case of symmetric potential jamming signals. However, it is possible to determine additional characteristics of the potential jamming signal by determining further time durations, such as durations of a third span and a fourth time span. For example, such additional time durations may be employed to distinguish between different chirp signal patterns, for example differentiate between triangular, sinusoidal and exponential chirp signals.

[0085] Let us now consider the example of triangular chirp signal pattern shown in Figure 4. It may be seen that the triangular chirp signal comprises an up-chirp as well as a down-chirp within a single period tp of the triangular chirp signal. The triangular chirp signal would be detected by the receiver 121 twice within the monitored frequency band, once during the up-chirp and once during the down-chirp. For this reason, there are two durations during which the triangular chirp signal sweeps the monitored frequency band, a duration of a first time span til and a duration of a third time span ti2. Furthermore, there are two durations during which the triangular chirp signal does not sweep the monitored frequency band or sweeps outside of the monitored frequency band, a duration of a second time span tol and a duration of a fourth time span to2.

[0086] Therefore, the period tp of the triangular chirp signal may be seen as the sum of four durations - duration of the first time span til, duration of the second time span tol, duration of the third time span ti2 and duration of the fourth time span to2, as shown in Figure 4. That is, tp = til + tol + ti2 + to2, (7) wherein + denotes a summation operation

[0087] The knowledge of the energy transitions due to the triangular chirp signal sweeping within the monitored frequency band to sweeping outside of the monitored frequency band and vice versa, may be exploited to determine the four time durations. For example, the start time and the end time of the energy pulse for the first detection (during up-chirp) and the second detection (during down-chirp) by the receiver 121 within the monitored frequency band may be obtained, and the time duration for which the triangular chirp signal is present (or not present) in the monitored frequency band may be determined. Further, the durations of the first and the second time spans may, for example, be determined according to the similar procedures as described above for the sawtooth chirp signal of Figure 2. The durations of the first and the second time spans til and tol may, for example, be determined according to the steps 203 and 204 of Figure 3 a respectively. The durations of the third and the fourth time spans ti2 and to2 may for example be determined in a similar way as the durations of the first and the second time spans til and tol.

[0088] Once the four durations of the time spans have been obtained, various characteristics of the triangular chirp signal may be determined. The duration of the first time span til and the duration of the third time span ti2 may be used to determine the linear sweep rates as follows:

[0089] (8)

[0090] (9) where rl indicates the linear sweep rate for the up-chirp part of the triangular chirp signal and r2 indicates the linear sweep rate for the down-chirp part of the triangular chirp signal. Br as described above is the bandwidth of the monitored frequency band.

[0091] The bandwidth Bj of the triangular chirp signal may thus be determined by

[0092] (10) where * denotes multiplication and tp is the period of the triangular chirp signal.

[0093] If the center frequency fcj of the triangular chirp signal is not equal to the center frequency fcr of the frequency band monitored by the receiver 121, then the time durations of the out-of- band time spans, tol and to2, will be unequal. The frequency offset fo between fcj and fcr may be determined by

[0094] The center frequency of the triangular chirp signal may be determined by (12)

[0095] In this way one or more characteristics of the triangular chirp signal may be determined, such as the center frequency fcr, the bandwidth Bj and the period tp.

[0096] Based on the techniques exemplified above with reference to Figure 4, a number of optional steps will now be described for the method 200 from Figure 3a. These optional steps are illustrated in Figure 3b. In some embodiments, the method 200 comprises the optional step of determining 206 a duration of a third time span for when the potential jamming signal is present within the monitored frequency band. As shown above in reference to the example of the triangular chirp signal of Figure 4, the potential jamming signal may comprise a down-chirp part in addition to the up-chirp part within its period tp. The duration of the third time span, herein referred to as ti2, may correspond to a duration when the down-chirp part of the potential jamming signal is present within the monitored frequency band. The determination of the one or more characteristics of the potential jamming signal at step 205 may for example be further based on the duration on the third time span.

[0097] The time spent by the potential jamming signal in the monitored frequency band may, thus, be split into the time spent in the monitored frequency band during the up-chirp parts of the signal (corresponding to the first time span til) and the time spent in the monitored frequency band during the down-chirp parts of the signal (corresponding to the third time span ti2). The up- chirp parts will then correspond to every second time the potential jamming signal sweeps the monitored region (for example all odd-numbered occasions), and the down-chirp parts will correspond to the remaining times the potential jamming signal visits the monitored region (for example all even-numbered occasions). As an example, the time spent by the potential jamming signal in the monitored frequency band may be represented by the one-hypothesis set Hl mentioned above. By forming averages over the even-numbered occasions and odd-numbered occasions, respectively, more robust / accurate / reliable values for the first time span til and the third time span ti2 may be obtained. Such averages (formed over multiple periods of the jamming signal) may for example be used in the determination step 205 instead of directly using values for the first time span til and the third time span ti2 obtained from a single period of the jamming signal.

[0098] Hence, in some embodiments, method 200 comprises the optional step 207 which includes determining an average of the durations of odd-numbered time spans during which the potential jamming signal is present in the monitored frequency band, and determining an average of the durations of even-numbered time spans during which the potential jamming signal is present in the monitored frequency band. In such embodiments, the determination of the one or more characteristics of the potential jamming signal at step 205 may be based on the determined averages. That is, the determination of the one or more characteristics may be based on the average of the durations of the odd-numbered time spans during which the potential jamming signal is present in the monitored frequency band and the average of the durations of the even- numbered time spans during which the potential jamming signal is present in the monitored frequency band. Time spans (or occasions) during which the potential jamming signal is present in the monitored frequency band may be enumerated according to the order in which they appear in time. The odd-numbered and even numbered time spans referred to in step 207 may refer to such an enumeration.

[0099] In some embodiments, the method 200 comprises the optional step of determining 208 a duration of a fourth time span for when the potential jamming signal is not present within the monitored frequency band. In a single period of the potential jamming signal comprising an up-chirp and a down-chirp (for example the triangular chirp signal pattern of Figure 4) there may be two durations during which the potential jamming signal does not sweep the monitored frequency band. The first duration during which the potential jamming signal is not present in the monitored frequency band is the duration of the second time span described above in relation to step 204 (and exemplified in Figure 4 by tol). The second duration during which the potential jamming signal is not present in the monitored frequency band is the duration of the fourth time span referred to at step 208 (and exemplified in Figure 4 by to2). As described above using the example of triangular chirp signal pattern of Figure 4, the duration of the fourth time span may be used to determine one or more characteristics of the potential jamming signal, such as for example the center frequency, bandwidth and / or period.

[0100] The time during which the potential jamming signal is not present in the monitored frequency band includes the second time span (exemplified by tol in Figure 4) and the fourth time span (exemplified by to2 in Figure 4). As an example, the time during which the potential jamming signal is not present in the monitored frequency band may be represented by the zerohypothesis set HO mentioned above. The time during which the potential jamming signal is not present in the monitored frequency band may be divided into two subsets in a comb-like fashion because the potential jamming signal may be above or below the monitored frequency band, and these two different situations typically occur every second time the potential jamming signal is outside of the monitored frequency band. The first subset of the time during which the potential jamming signal is not present in the monitored frequency band may comprise every second occasion (for example all odd-numbered occasions) while the second subset comprises the remaining occasions (for example all even-numbered occasions). By forming averages over the even-numbered occasions and odd-numbered occasions, respectively, more robust / accurate / reliable values for the second time span tol and the fourth time span to2 may be obtained. Such averages (formed over multiple periods of the jamming signal) may for example be used in the determination step 205 instead of directly using values for the second time span tol and the fourth time span to2 obtained from a single period of the jamming signal.

[0101] Hence, in some embodiments, the method 200 comprises the optional step 209, which includes determining an average of the durations of odd-numbered time spans during which the potential jamming signal is not present in the monitored frequency band, and determining an average of the durations of even-numbered time spans during which the potential jamming signal is not present in the monitored frequency band. In such embodiments, the determination of the one or more characteristics of the potential jamming signal at step 205 may be based on the determined averages. That is, the determination of the one or more characteristics may be based on the average of the durations of the odd-numbered time spans during which the potential jamming signal is not present in the monitored frequency band and the average of the durations of even-numbered time spans during which the potential jamming signal is not present in the monitored frequency band. Time spans (or occasions) during which the potential jamming signal is not present in the monitored frequency band may be enumerated according to the order in which they appear in time. The odd-numbered and even numbered time spans referred to in the step 209 may refer to such an enumeration.

[0102] In some embodiments, the method 200 comprises the optional step of determining 210 a signal pattern of the potential jamming signal based on at least the durations of the following: the first time span, the second time span, the third time span and the fourth time span (which may be determined at steps 203, 204, 206 and 208, respectively). These four durations enable different jamming signal patters to be distinguished from each other. For example, the asymmetric shape of the triangular chirp signal in Figure 4 causes til to differ from ti2 and tOl to differ from t02, which is not compatible with the sawtooth chirp signal in Figure 2 nor the sinusoidal chirp signal of Figure 5. The signal pattern determined at step 210 may for example be a triangular chirp, a saw-tooth chirp, a sinusoidal chirp or an exponential chirp.

[0103] It may be desired in certain situations to avoid unnecessary computations of the signal characteristics, for instance, if a potential jamming signal is not indeed a jamming signal. Thus, in order to reduce the processing overhead and thereby save computational costs and energy, the method 200 may comprise the optional step of determining 211 that the potential jamming signal is not a jamming signal. The determination at step 211 may, for example, be based on the estimation performed at step 202. For example, it may be estimated that the energy detected due to the presence of the potential jamming signal in the monitored frequency band does not exceed a pre-defined energy threshold, to qualify as a jamming signal. The determination step 211 may for example be carried out concurrently with, or as part of, step 202.

[0104] The determination at step 211 may, for example, be based on some or all of the durations determined at steps 203, 204, 206, and 208 (such as til, tol, ti2 and / or to2). For example, if these durations are not compatible with the expected jamming signal pattem(s) (for example one or more of the chirp signal patterns in Figure 2, Figure 4, Figure 5 and Figure 6), it may be determined that the potential jamming signal is not to be considered as a jamming signal. The determination step 211 may for example be carried out concurrently with, or as part of, steps 203, 204, 205, 206 and / or 208.

[0105] Figure 5 illustrates a frequency-time graph of a jamming signal in the form of a sinusoid signal pattern according to one or more embodiments.

[0106] The procedure of the method steps described above in relation to Figure 3a and Figure 3b, namely the method steps 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, may also be applicable to embodiments involving jammer signals of the type shown in Figure 5. For example, the determination of the durations of the first time span, the second time span, the third time span and the fourth time span may be performed according to Steps: 203 of Figure 3a, 204 of Figure 3a, 206 of Figure 3b and 208 of Figure 3b, respectively. The one or more characteristics of the sinusoid signal may be determined, exemplifying the step 205 of Figure 3a, the step 206 of Figure 3b and the step 208 of Figure 3b, as follows:

[0107] Referring to Figure 5, let,

[0108] The period tp of the sinusoid signal is obtained by summing the durations of the first time span, the second time span, the third time span and the fourth time span, as

[0109] Further, let the sweep rates be, Further, the receiver bandwidth is given by

[0110] Br = r1 - r2 (18)

[0111] Inserting equations (16) and (17) in (18), the bandwidth of the sinusoid signal (or the jamming signal) Bj may be determined by,

[0112] The center frequency fcj of the sinusoid signal may be determined by,

[0113] Thus, the device may determine one or more jamming characteristics of the sinusoid signal (or the jamming signal) such as the center frequency, bandwidth and the period based on equations (20), (19) and (15) described above. The determination may, thus, be based on time durations of the first time span til, the second time span tol, the third time span ti2 and the fourth time span to2.

[0114] Figure 6 illustrates a frequency-time graph of a jamming signal in the form of an exponential signal pattern according to one or more embodiments.

[0115] The procedure of the method steps described above in relation to Figure 3a and Figure 3b, namely the method steps 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, may also be applicable to embodiments involving jammer signals of the type shown in Figure 6. For example, the determination of the durations of the first time span, the second time span, the third time span and the fourth time span may be performed according to Steps: 203 of Figure 3a, 204 of Figure 3 a, 206 of Figure 3b and 208 of Figure 3b, respectively.

[0116] Measurements show that several commercial jammers have exponential sweep characteristics. Such sweeps may typically be generated by Resistor-Capacitor (RC) circuits. Figure 7 illustrates one such example of an RC circuit comprising a resistor R and a capacitor C, connected to a voltage supply and a voltage-controlled oscillator (VCO). Figure 8 illustrates another example of an RC circuit wherein a resistor Rc is used for charging procedure while a different resistor Rd is used for the discharging procedure. If different resistors are used for charging and discharging procedures, then more general values of the one or more jamming characteristics, such as jammer bandwidth, are possible.

[0117] Now referring to Figure 7, the RC circuit is described as follows. If the switch S has been in the lower position, i.e. in SI, for a long time, then the capacitor C will be completely discharged. At time t=0, the switch S is moved to the upper i.e. S2 and the capacitor starts to charge. The voltage over the capacitor at time t>0 is given by,

[0118] Vc = Vs * (1 — exp (— t / tf) (21) where Vs is the voltage of the voltage supply and T is the time constant r = R*C. At time t=5* T, Vc = 0.99 Vs and we can consider the capacitor fully charged.

[0119] If the switch S has been in S2 position for a long time, the capacitor is fully charged i.e. Vc = Vs. If the switch S is moved to the lower position SI at time t=0, then the voltage over the capacitor at time t>0 will be given by,

[0120] Vc = Vs * exp (— t / r) (22)

[0121] Now at time t=5* T, VC=0.01 VS and we can consider the capacitor discharged.

[0122] The voltage over the capacitor is fed to the VCO. The VCO outputs an instantaneous frequency given by,

[0123] / (t) = yo + K * Vin (23) where fO is the quiescent (dormant) frequency, K is a constant and Vin is the input voltage, here Vin = Vc. Let fs be the frequency swing of the VCO if the RC circuit is allowed to reach fully charged and discharges states, then fs = K*Vs. Assume that the capacitor is completely charged and discharged before the switch position is changed, the up-sweep frequency may determined as fu(t) = f0 + fs * ((1 — exp (—t / r)). (24)

[0124] The down-sweep frequency may be determined as fd(t) = f0 + fs * (exp (—t / t)). (25)

[0125] In Figure 8, there are two different resistors Rc and Rd for charging and discharging, respectively. In comparison to Figure 7, two different time constants, TCfor charging and Ta for discharging, are obtained. The up-sweep frequency and down-sweep frequency may thus be determined by replacing T in (24) and (25) with TCand Td respectively.

[0126] In the description above, it was assumed that the capacitor is allowed to charge and discharge completely. If this is not the case, i.e., if, for example, the position of the switch is changed with a frequency greater than 1 / (5* T), then the capacitor will not charge and discharge fully. The input voltage to the VCO may not reach its end values (0 and Vs), and the output frequency swing may not be the full frequency range.

[0127] As indicated in Figure 6, the slope of the frequency change (and corresponding voltage change) is largest just after the switch position is changed (from charging to discharging) and the slope changes within the receiver bandwidth. Depending on how the exponential signal (or jamming signal) bandwidth and the receiver bandwidth overlap, the receiver may be affected by different parts of the exponential sweep signal.

[0128] If the receiver center frequency may be changed (retuned) and / or the receiver bandwidth may be changed, the parameters of the RC circuit may be estimated and the bandwidth of the jamming signal may be determined by fitting the time durations to the exponential curves shown in Figure 6. Retuning may also be used to determine the jamming center frequency.

[0129] Figure 9 illustrates a method 900 according to one or more embodiments of the invention. The method 900 may be performed by the device 120 or the receiver 121 or by a combination of the device 120 and the receiver 121. The method 900 comprises one or more of the following steps, some of which are optional:

[0130] Step 901 (optional) - The potential jamming signal is detected in the monitored frequency band by the receiver 121 or the device 120. The receiver may send an input regarding the potential jamming signal as described in step 201 of Figure 3a. The input may, for example, be energy (or amplitude) samples of the potential jamming signal. The energy (or amplitude) samples of the potential jamming signal are filtered to reduce the impact of noise. For this purpose, a filter such as a square filter or raised cosine filter may be used.

[0131] At Step 902 - The filtered signal samples are divided into two sets. The first set may be a zerohypothesis set (HO) corresponding to potential jamming signal not being present in the monitored frequency band. The second set may be a one-hypothesis set (Hl) corresponding to a potential jamming signal being present in the monitored frequency band. Step 902 is an example step of step 202 of Figure 3 a.

[0132] An intra-set variance minimizing algorithm, for example Otsu’s method, may be used at step 902 to separate the input signal into two sets.

[0133] An intra-set variance minimizing algorithm such as Otsu’s method may be used to divide the input and perform the estimation at step 202 and / or step 902. Otsu’s method is used to perform automatic thresholding. In the simplest form, Otsu's method returns a single threshold that separates the samples into two classes. This is done by computing a histogram and probabilities of each signal level in a set of the samples. The threshold that minimizes the intra-class intensity variance, or maximizes the inter-class variance, is exhaustively searched. The means and variances of the two sets are further determined. If the difference between means of the two sets is small compared to the mean values themselves, one may conclude that all samples belong to the same set (e.g., corresponding to the set HO, i.e., no jamming signal present). If the sum of the variances of the two sets, weighted by the size of the sets, is close to the variance of the whole set, then one may conclude that all samples belong to the same set corresponding to the set (e.g., HO, i.e., no jamming signal present).

[0134] It may be noted that while Otsu’s method works on a set of samples, an approximate version of Otsu’s method employed on a continuous stream of samples may be implemented in practice. Otsu’s method may be generalized to multiple classes using algorithms such as k- means algorithm.

[0135] As an alternative to Otsu’s method, a peak detection / threshold technique such as the z-test method may be used to divide the input signal from step 201 and / or step 901 into two sets and perform the estimation at step 202 and / or 902. A peak detection / threshold algorithm may return an empty Hl set. The z-test method compares a test value z (also referred to as ‘standard score’), to a predetermined value. The standard score may be computed as, where x is a test sample, m is the mean of the null hypothesis, and s is the standard deviation of the null hypothesis. The null hypothesis is the HO set.

[0136] If z exceeds a predetermined threshold, then the sample x belongs to the set Hl, otherwise the sample x belongs to the set HO. The threshold is selected such that exceeding the threshold value in a Gaussian distribution happens with a desired, small probability. The mean value and standard deviation used in the z-test may either be precomputed, computed from the block of samples at hand, or computed from continuously streaming data samples.

[0137] In the case of precomputing, prior knowledge of the channel or the communication system is utilized to estimate the mean and standard deviation. In the case of computing from the block of samples, an initial threshold from a previous block is used to identify the samples in the HO set, then the mean and standard deviation of the current samples in the HO set is computed. Optionally the sets are updated. In the case of streaming data samples, the mean and standard deviation are computed as moving averages of the samples that belong to the HO set. In other words, if the sample under consideration is classified as belonging to the HO set, then the mean and standard deviation is updated with this sample, otherwise not.

[0138] At Step 903 (Optional) - There may exist noise or other interferences due to which the potential jamming signal sweeping the monitored frequency band may spuriously make a transition to a frequency band outside of the monitored frequency for a very short period of time. In such a case, samples may be classified into the HO set for the duration of the spurious behaviour. If there are spurious transitions between the two sets, then such spurious transitions are removed. One way to implement this is to create an index vector of the same length as the sample vector. In the index vector, let a 1 indicate that the sample belongs to the Hl set and a 0 indicate that the sample belongs to the HO set. The index vector contains a series of 1 s when the potential jamming signal is in the monitored frequency band and a series of Os when the potential jamming signal is outside the monitored frequency band. Spurious samples can either be in the series or at the transitions between the series of Os and 1 s (both from Os to Is and 1 s to Os). The average durations of the series of Is (or Os) may be determined, and these averages may be used to detect and remove spurious samples. As an example, if in the middle of a series of Os, a single sample is 1 or a number of samples, significantly smaller than the mean duration of the series of Is, are Is, then this is considered a spuriosity and the Is are set to Os. At transitions (from HO to Hl or vice versa), rapid, repeating transitions between the two sets are removed. E.g., if at a transition from Os to Is, there is a pattern of detections as 000 1 0 1 1 1, the detections are changed to 0 0 0 0 1 1 1 1.

[0139] Further, if the series of 0s and Is are short and have irregular lengths, i.e., the variance of the lengths is high, then it may be an indication that the splitting of the set may have been carried out erroneously. Hence, the spuriosity removal at step 903 can serve as a sanity check on step 902.

[0140] A filter may be used to remove the spurious sample. The parameters for the filter, for example filter used in step 901 and / or filter used in step 903, may need to be adapted according to a present value of Jamming signal-to-Noise ratio (JNR), otherwise there may be a risk of removing actual jamming signal samples of the input data. Prior knowledge about the periodicity or pseudo-periodicity of the potential jamming signal may be used (e.g., knowing that there is always a given number of peaks within any fixed interval). At Step 904 - It is determined whether a potential jamming signal is a jamming signal or not. If a threshold test / peak detection is used in step 902, the presence of the potential jamming signal is determined by determining if the Hl set is empty (or of a sufficiently small size, which may be considered as a false alarm). On the other hand, if a variance-minimizing algorithm is used, the presence of the potential jamming signal is determined by comparing the difference between the variance of the complete input signal and the weighted sum of the variances of sets HO and Hl. If a potential jamming signal is determined to be a jamming signal, then the procedure continues to step 905. If it is determined that the potential jamming signal is not a jamming signal, the procedure stops.

[0141] This step 904 has an advantage that it enables the procedure to terminate upon determining that the potential jamming signal is not a jamming signal, thereby enabling savings of computational resources and reduction of processing costs.

[0142] At Step 905 - The time / index of the transitions between the sets HO and Hl is determined. The time may be read from a system clock. The index is a counter of the sample number, which multiplied with the sample period provides a time.

[0143] At Step 906 - The duration of times spent in set HO and Hl is determined. The times spent in Hl corresponds to the time duration of the jamming signal being present in the communication system receiver bandwidth. The time spent in HO corresponds to the time duration of the jamming signal not being present in the communication system receiver bandwidth.

[0144] In one or more embodiments, the jamming signal is a chirp signal that sweeps up and down in frequency. In the case, the times spent in Hl is further divided into two subsets in a comb-like fashion, wherein the first subset is all odd-numbered times, and the second subset is all the even-numbered times. Let til be the average of the times in the first subset and ti2 be the average of the times in the second subset. Similarly, the times spent in HO is divided into two subsets of odd-numbered and even-numbered times. Let tol be the average of the times in the first subset and to2 be the average of the times in the second subset, til may for example be the duration of the first time span described above in step 203 of Figure 3a. tol may for example be the duration of the second time span described above in step 204 of Figure 3 a. ti2 may for example be the duration of the third time span described above in step 206 of Figure 3b. to2 may for example be the duration of the fourth time span described above in step 208 of Figure 3b. Let T1 be the average time it takes for the jamming signal to sweep the communication system receiver bandwidth in one direction (either up or down). Let T2 may be the average time it takes to sweep in the other direction. Thus, the jamming signal (or chirp signal) sweep period may be determined using the average time between every second transition from HO to Hl (and / or every second transition Hl to HO).

[0145] At Step 907 (optional) - The signal pattern (or the sweep pattern) of the jamming signal may be determined based on til, ti2, tol and to2. The signal pattern of the jamming signal may, for example, be sawtooth, triangular, sinusoidal or exponential.

[0146] In or more embodiments, a sawtooth jamming signal of Figure 2 may be determined based on til=ti2 and tol=to2.

[0147] In one or more embodiments, a general triangular jamming signal, such as that of Figure 4, may be fitted to all values of the times. This may also be the case for an exponential signal, such as that of Figure 6.

[0148] In or more embodiments, a sinusoidal jamming signal of Figure 5 may be determined based on til=ti2 while the tol and to2 being unequal.

[0149] At Step 908 - The one or more jamming characteristics are estimated according to procedures described above, for example step 205 of Figure 3a and / or Steps 206-209 of Figure 3b. The estimated one or more jamming characteristics may then be outputted and used for jamming signal classification and / or mitigating impact of the jamming signal on the communication system.

[0150] Figure 10 illustrates an implementation of the device 120 in a communication system 700 according to one or more embodiments. The device 120 may be implemented as an additional functional unit in an existing communication system, e.g., 3GPP, IEEE 802.11. The Radio Frequency (RF) front-end 1001 of the existing communication system may be used. The RF front-end 1001 may perform the function of processing the signal, e.g. the potential jamming signal or any signal detected in the monitored frequency band, at the original incoming radio frequency (RF), before it is converted to a lower intermediate frequency (IF). The RF frontend may include a band-pass filter, an RF amplifier, a local oscillator and a mixer. The Analog- to-Digital Converter (ADC) 1002 receives the analog output from the RF front end 1001 and digitizes the signal into In-phase Quadrature-phase (IQ) samples. The IQ samples are then sent to the device 120 that then performs a method according to one or more embodiments described above in relation to Figures 1 to 9. The IQ sample and / or output from the device 120 may then sent to a baseband processing block 1003 e.g., for working on Error Vector Magnitudes (EVM) values. EVM provides a comprehensive measure of the quality of the radio receiver or transmitter for use in digital communications. EVM captures noise, distorsion, phase noise and interference. EVM is commonly used in 3GPP to measure performance.

[0151] Figure 11 illustrates an implementation of the device 120 in a standalone system according to one or more embodiments. In this case, the standalone system requires its own RF front-end 1101 and ADC 1102.

[0152] Figure 12 illustrates a device 120 for determining one or more jamming characteristics of a potential jamming signal, comprising one or more processor(s) 1203 and memory 1201. The memory 1201 contains instructions which when executed on the one or more processor(s) 1203 cause the device 120 to perform a method described in relation to Figures 1-11, for example, method steps of Figure 3a and / or Figure 3b and / or Figure 9. The memory 1201 or computer readable storage medium 1202 may be configured to store data, programmatic software code and / or other information described herein. The instructions may be software (SW) or computer program 1204 associated with the device 120.

[0153] Thus, the device 120 may further comprise SW or computer program 1204, which is stored in, for example, the memory 1201 or the computer readable storage medium 1202 at the device 120, or stored in external memory, e.g., database, accessible by the device 120. The SW or computer program 1204 may be executable by the one or more processors 1203.

[0154] The device may comprise a receiver / transmitter (TX / RX) 1205, for example, the receiver 121. The receiver 121 may monitor a frequency band and detect the presence / absence of the potential jamming signal in the monitored frequency band. The receiver 121 may send an input on the potential jamming signal to the device 120.

[0155] The device 120 according to Figure 12 may have storage and / or processing capabilities. The device 120 may include one or more processors 1203 and a memory 1201 or a computer readable storage medium 1202. In particular, in addition to a traditional processor and memory, the device 120 may comprise integrated circuitry for processing and / or control, e.g., one or more processors and / or processor cores and / or Field Programmable Gate Array (FPGAs) and / or Application Specific Integrated Circuitry (ASICs) adapted to execute instructions. The processors) 1203 may be configured to access, e.g., write to and / or read from) the memory 1201 or the computer readable storage medium 1202, which may comprise any kind of volatile and / or nonvolatile memory, e.g., cache and / or buffer memory and / or Random Access Memory (RAM) and / or Read-Only Memory (ROM) and / or optical memory and / or Erasable Programmable Read-Only Memory (EPROM).

[0156] A computer program product in the form of a computer readable storage medium 1202 may comprise any form of volatile or non-volatile computer readable memory including, without limitation, persistent storage, solid-state memory, remotely mounted memory, magnetic media, optical media, RAM, ROM, mass storage media, for example: a hard disk, removable storage media for example: a flash drive, a Compact Disk (CD) or a Digital Video Disk (DVD), and / or any other volatile or non-volatile, non-transitory device 120 readable and / or computerexecutable memory devices that store information, data, and / or instructions that may be used by one or more processors 1203. Computer readable storage medium 1202 may store any suitable instructions, data or information, including a computer program 1204, software, an application including one or more of logic, rules, code, tables, etc. and / or other instructions capable of being executed by one or more processors 1203. Computer readable storage medium 1202 may be used to store any calculations made by one or more processors 1203. In some embodiments, one or more processors 1203 and computer readable storage medium 1202 may be considered to be integrated.

[0157] The device 120 may, for example, be a radio network node capable of transmitting and receiving a communication message. The device 120 may, for example, be a Third Generation Partnership Project (3GPP) cellular communication node such as a Sixth Generation (6G) node, a Fifth Generation (5G) node, a Long-Term Evolution (LTE) node, a Third Generation node (3G) or a Second Generation node (2G). More specifically, the device 120 may be a communication node in the radio access network (RAN) of any one of the above communication networks. In an embodiment, the device 120 is a gNodeB (gNB). The device 120 may for example be implemented in a RAN node.

[0158] The device 120 may also be a communication node operating in an Internet of Things (loT) network or a Low Power Wide Area Network (LPWAN). In other words, the device 120 may, for example, operate in any one of the 3GPP Narrowband loT (NB-loT), SigFox, LoRaWAN communication networks. Additionally or alternatively, the device 120 may be a communication node in the RAN of any of the above communication networks.

[0159] In some embodiments, the device 120 may be an edge node. In some embodiments where the device 120 is a gNB, the gNB may comprise both the gNB-Control Unit (gNB-CU) and the gNB-Distributed Unit (gNB-DU) in the device 120. In another embodiment, the device 120 is the gNB-DU wherein the gNB-CU resides in another radio node. In yet another embodiment, the device 120 is the gNB-CU wherein the gNB-DU is deployed separately and the gNB-DU interacts with a user equipment (UE). In each of the embodiments, the gNB-DU and the gNB- CU communicate with each other over the Fl interface.

[0160] In one or more embodiments, the device 120 may communicate with one or more radio base stations to obtain input on the potential jamming signal. In some other embodiments the device 120 may directly monitor the receiver band for the detecting the potential jamming signal.

Claims

CLAIMS:

1. A method (200) for determining one or more characteristics of a potential jamming signal, the method comprising: obtaining (201), from a receiver monitoring a frequency band (Br), input regarding a potential jamming signal; estimating (202), based on the input, when the potential jamming signal is present within the monitored frequency band and when the potential jamming signal is not present within the monitored frequency band;- determining (203) a duration of a first time span for when the potential jamming signal is present within the monitored frequency band;- determining (204) a duration of a second time span for when the potential jamming signal is not present within the monitored frequency band; and determining (205) one or more characteristics of the potential jamming signal based on the durations of the first and second time spans.

2. The method according to claim 1, further comprising determining (206) a duration of a third time span for when the potential jamming signal is present within the monitored frequency band, and wherein the determination of the one or more characteristics of the potential jamming signal is further based on the duration of the third time span.

3. The method according to claim 2, wherein the first time span and the third time span correspond to an odd-numbered time span and an even-numbered time span respectively, in a consecutive enumeration of time spans during which the potential jamming signal is present in the monitored frequency band.

4. The method according to claim 3, further comprising:- determining (207) an average of the durations of odd-numbered time spans during which the potential jamming signal is present in the monitored frequency band; and- determining (207) an average of the durations of even-numbered time spans during which the potential jamming signal is present in the monitored frequency band,wherein the determination of the one or more characteristics of the potential jamming signal is based on the determined averages.

5. The method according to any of claims 1-4, further comprising determining (208) a duration of a fourth time span for when the potential jamming signal is not present within the monitored frequency band, and wherein the determination of the one or more characteristics of the potential jamming signal is further based on the duration of the fourth time span.

6. The method according to claim 5, wherein the second time span and the fourth time span correspond to an odd-numbered time span and an even-numbered time span respectively, in a consecutive enumeration of time spans during which the potential jamming signal is not present in the monitored frequency band.

7. The method according to claim 6, further comprising: determining (209) an average of the durations of odd-numbered time spans during which the potential jamming signal is not present in the monitored frequency band; and determining (209) an average of the durations of even-numbered time spans during which the potential jamming signal is not present in the monitored frequency band, wherein the determination of the one or more characteristics of the potential jamming signal is based on the determined averages.

8. The method according to any of claims 1-7, comprising determining (210) a signal pattern of the potential jamming signal based on at least the durations of the following: the first time span, the second time span, the third time span and the fourth time span.

9. The method according to claim 8, wherein the signal pattern is at least one of: a triangular chirp, a saw-tooth chirp, a sinusoidal chirp and an exponential chirp.

10. The method according to any of claims 1-9, comprising determining (211), based on the estimation, that the potential jamming signal is not a jamming signal.

11. The method according to any of claims 1-10, wherein the estimating is based on an intra-set variance minimizing algorithm.

12. The method according to any of claims 1-11, wherein the estimating is based on a peak detection or threshold-based algorithm.

13. The method according to any of claims 1-12, wherein the estimating includes performing filtering of the input or of an intermediate result obtained based on the input.

14. The method according to any of claims 1-13, wherein the input comprises squared magnitude of samples of the potential jamming signal.

15. The method according to any of claims 1-14, wherein the one or more characteristics of the potential jamming signal includes at least one of: center frequency of the potential jamming signal (fcj), bandwidth of the potential jamming signal (Bj), period of the potential jamming signal (tp).

16. The method according to any of claims 1-15, wherein the method is performed by Radio Access Network, RAN, equipment or a RAN node.

17. A device (100) for determining one or more characteristics of a potential jamming signal, the device comprising a memory (1101) and a processor (1103), the memory containing instructions which when executed on the processor, cause the device to:- obtain, from a receiver monitoring a frequency band (Br), input regarding a potential jamming signal; estimate, based on the input, when the potential jamming signal is present within the monitored frequency band and when the potential jamming signal is not present within the monitored frequency band; determine a duration of a first time span for when the potential jamming signal is present within the monitored frequency band; determine a duration of a second time span for when the potential jamming signal is not present within the monitored frequency band; anddetermine one or more characteristics of the potential jamming signal based on the durations of the first and second time spans.

18. The device (100) according to claim 17, the memory containing instructions which when executed on the processor, cause the device to perform the method according to one or more of claims 2-16.

19. A computer program (1104) comprising instructions which, when executed on a device (100), cause the device to carry out the method according to one or more of claims 1-16.

20. A computer program product (1102) comprising a computer readable storage means on which the computer program (1104) according to claim 19 is stored.

Images