Claims:
1. A method for drone signal detection in presence of multiple signals, the method comprising:
receiving, by a capturing unit (502), a plurality of RF communication signals from surrounding environment and converting the received RF communication signals into a plurality of digitized samples (602);
generating, by a spectrogram computation and filtering unit (504), spectrogram of the digitized samples and filtering the generated spectrogram to reduce noise, wherein the generated spectrogram of the digitized samples comprises one or more frequency bins and one or more time bins (604);
detecting, by a signal detection unit (506), a plurality of signals with varying power levels by applying a unique dynamic threshold for each frequency bin in the filtered spectrogram of the digitized samples (606);
identifying, by a merged signal identification unit (508), presence of merged signals in the plurality of detected signals and splitting the merged signals for detecting signals which are overlapped in at least one of time domain and frequency domain of the filtered spectrogram (608);
estimating, by a signal parameters estimation unit (510), parameters of the detected signals by applying image processing techniques on the detected signals (610);
comparing, by a signal parameters comparison unit (512), the estimated signal parameters to signal parameters stored in a database, wherein the stored signal parameters are the signal parameters associated with one of different classes of RF communication signals and different types of drone devices (612); and
detecting, by the signal parameters comparison unit (512), a drone signal, when the estimated signal parameters of at least one of the detected signals matches with the stored signal parameters of at least one type of drone device (614).
2. The method as claimed in claim 1, wherein detecting the signals with varying power levels by applying the unique dynamic threshold comprises:
calculating a difference between a maximum power and a mean power for each frequency bin in the filtered spectrogram of the digitized samples;
smoothening the calculated maximum and mean power difference values to eliminate the noise;
estimating a noise floor by using mode of the smoothened maximum and mean power difference values to determine the frequency bins with the signals;
detecting valid peaks in the smoothened maximum and mean power difference values for accurate estimation of the frequency bins where the signals are present; and
applying the unique dynamic threshold for each frequency bin of the detected valid peaks depending on a minimum power, the mean power and the maximum power in each frequency bin for accurate estimation of the time bins where the signals are present and thereby obtaining a binarized image.
3. The method as claimed in claim 1 and 2, wherein estimating the parameters of the detected signals comprises:
performing dilation and erosion on the binarized image to eliminate the noise;
extracting contours in the binarized image and enforcing conditions on area, mean and variance of the extracted contours for selecting all valid contours; and
estimating the parameters of the detected signals using all the valid contours.
4. The method as claimed in claim 3, wherein for all the valid contours, top left and bottom right corner details of a rectangle enclosing the valid contours are used to compute a start time and an end time of each signal, a start frequency and an end frequency of each signal for estimating the parameters of the detected signals, and the estimated parameters of the detected signals comprise of bandwidth, time duration and centre frequency.
5. The method as claimed in claim 1, 3 and 4, wherein estimating the parameters of the detected signals further comprise merging of split signals of low power by enforcing constraints on the time duration and the bandwidth of the detected signals.
6. The method as claimed in claim 1, wherein the multiple signals comprise of the plurality of interfering RF communication signals which include signals from multiple drone devices.
7. The method as claimed in claim 1, wherein identifying and splitting the merged signals comprises of analysing a bandwidth variation of the merged signals along a time axis and a time duration variation of the merged signals along a frequency axis in the filtered spectrogram for estimating the correct signal parameters of the drone signal merged with the multiple signals.
8. A system for drone signal detection in presence of multiple signals, the system comprising:
a capturing unit (502), configured to receive a plurality of RF communication signals from surrounding environment and convert the received RF communication signals into a plurality of digitized samples;
a spectrogram computation and filtering unit (504) configured to receive the digitized samples from the capturing unit (502), generate spectrogram of the digitized samples and filter the generated spectrogram to reduce noise, wherein the generated spectrogram of the digitized samples comprises one or more frequency bins and one or more time bins;
a signal detection unit (506) configured to receive the filtered spectrogram of the digitized samples from the spectrogram computation and filtering unit (504) and detect a plurality of signals with varying power levels by applying a unique dynamic threshold for each frequency bin in the filtered spectrogram of the digitized samples;
a merged signal identification unit (508) configured to receive the plurality of detected signals from the signal detection unit (506), identify presence of merged signals in the plurality of detected signals and split the merged signals for detecting signals which are overlapped in at least one of time domain and frequency domain of the filtered spectrogram;
a signal parameters estimation unit (510) configured to receive the detected signals and estimate parameters of the detected signals by applying image processing techniques on the detected signals;
a signal parameters comparison unit (512) configured to receive the estimated signal parameters from the signal parameters estimation unit (510), compare the estimated signal parameters to signal parameters stored in a database, wherein the stored signal parameters are the signal parameters associated with one of different classes of RF communication signals and different types of drone devices and detect a drone signal, when the estimated signal parameters of at least one of the detected signals matches with the stored signal parameters of at least one type of drone device; and
an alert generation unit (514) configured to generate an alert signal on detection of the drone signal.
9. The system as claimed in claim 8, wherein the capturing unit (502) comprises at least one sensor configured to capture the plurality of RF communication signals in the surrounding environment.
, Description:
FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENTS RULES, 2003
COMPLETE SPECIFICATION
(See section 10, rule 13)

“A METHOD AND SYSTEM FOR DRONE SIGNAL DETECTION”
By

BHARAT ELECTRONICS LIMITED
WHOSE ADDRESS IS
OUTER RING ROAD, NAGAVARA, BANGALORE – 560045, KARNATAKA, INDIA

The following specification particularly describes the invention and the manner in which it is to be performed.

Field of the Invention

The present invention mainly relates to a field of monitoring technology and, more particularly, the present invention relates to a method and a system for drone signal detection in presence of multiple signals.

Background of the invention

In general, Unmanned Aerial Vehicles (UAVs) such as drones are extremely beneficial and possess capability of reaching most remote areas with little to no manpower needed and are adopted widely in military and commercial sectors. The drones were initially thought of as military aircrafts only. But, usage of the drones for commercial purposes increased tremendously as the drones became cheaper in recent days. With significant increase in number of the drones used for multi-purpose applications, the drones impose a serious risk to public safety and national security. For example, when the drones are used for unauthorized aerial surveillance, the drones may pose serious threat to organizations or facilities such as nuclear facilities, airports, stadiums, oil and gas refineries and prisons, and the like. Therefore, the drones pose a high risk in sensitive areas and detection of the drones in such secure areas is very crucial.
Multiple signals may be present in surrounding environment of the drones. When there is only one signal present, detection of signal is easier. When multiple signals are present, drone signal detection becomes challenging. So, one of the major challenges is to detect a drone signal in the presence of multiple signals, which could be interferers or signals from other drones.
One of the prior arts discloses a system, method, and apparatus for drone detection and classification. The method includes receiving, recording and processing a sound signal into a feature frequency spectrum. The method further includes applying broad spectrum matching to compare the feature frequency spectrum to at least one drone sound signature stored in a database.
Another prior art discloses methods, systems, and apparatus for drone-augmented emergency response services monitoring a predetermined geographic area that surrounds a particular property. The method includes detection of drone using one or more sensor signals such as (i) audio signals from a propeller of a drone device, (ii) video signals of nearby airspace, (iii) thermal signals generated from the drone device, (iv) radar detection of aerial speed of the drone device, (v) radiofrequency (RF) detection of oscillation in electronic circuits of the drone device, or (vi) RF communications. The method determines whether the drone device that is detected is an unauthorized drone device, and transmits a signal indicating the detection of the unauthorized drone.
Further, another prior art discloses a method of capturing the presence of a drone by collecting data associated with an aerial object using at least one sensor (cameras, microphones/acoustic sensors, and radiofrequency antennas), analyzing the data using a processor to determine at least one characteristic of the aerial object, accessing a library of stored characteristics of commercially available drones from a database, determining if at least one characteristic of the aerial object matches a characteristic of a commercially available drone and generating an indication of a positive match.
Further, another prior art discloses a system that prevents unmanned aerial systems (UAS) from flying into a defined airspace. The system detects UAS traffic and transmits a broadcast signal over a transmission region indicating a no-fly zone in which only UAS having an authorization are allowed to fly. Those UAS without clearance are diverted around the no-fly zone, denied Wi-Fi and/or RF connection, forced to return to home launch sites via activation of standard pre-programmed Return to Home (RTH) routines, or forced to land at specified locations where they may be captured.
Practically, a signal detection system or method uses fixed thresholding for the signal detection. However, the signal detection system using a fixed threshold is extremely sensitive to noise. That is, the fixed thresholding when used for the signal detection has various limitations, especially in cases where signal power is comparable with noise power. So, the fixed threshold cannot be used for detecting a signal presence. This is because, there is a possibility of receiving multiple signals with varying power levels and one fixed threshold solution doesn’t fit for all the signals. If a fixed low threshold is set to ensure detection of a low power drone signal, there is a risk of detecting noise as drone signal and false alarms may increase. In other words, if focus is on increasing drone detection accuracy, then false alarm/false alarm rate for drone detection increases. If focus is on decreasing the false alarm rate, then the drone detection accuracy becomes poor.
Therefore, there is a need in the art with a method and system for drone signal detection when the multiple signals are present, either from the interferers or from multiple drones and to solve the above mentioned limitations.

Objective of the invention

The main objective of the present invention is to provide a method and system for drone signal detection in presence of multiple signals and to achieve a proper balance between drone detection accuracy and false alarm rate by incorporating dynamic thresholding.

Summary of the invention

An aspect of the present invention is to address the above-mentioned problems and/or disadvantages and to provide at least the advantages described below.
Accordingly, one aspect of the present invention relates to a method for drone signal detection in presence of multiple signals, the method comprising: receiving, by a capturing unit, a plurality of RF communication signals from surrounding environment and converting the received RF communication signals into a plurality of digitized samples, generating, by a spectrogram computation and filtering unit, spectrogram of the digitized samples and filtering the generated spectrogram to reduce noise, wherein the generated spectrogram of the digitized samples comprises one or more frequency bins and one or more time bins, detecting, by a signal detection unit, a plurality of signals with varying power levels by applying a unique dynamic threshold for each frequency bin in the filtered spectrogram of the digitized samples, identifying, by a merged signal identification unit, presence of merged signals in the plurality of detected signals and splitting the merged signals for detecting signals which are overlapped in at least one of time domain and frequency domain of the filtered spectrogram, estimating, by a signal parameters estimation unit, parameters of the detected signals by applying image processing techniques on the detected signals, comparing, by a signal parameters comparison unit, the estimated signal parameters to signal parameters stored in a database, wherein the stored signal parameters are the signal parameters associated with one of different classes of RF communication signals and different types of drone devices and detecting, by the signal parameters comparison unit, a drone signal, when the estimated signal parameters of at least one of the detected signals matches with the stored signal parameters of at least one type of drone device.
Another aspect of the present invention relates to a system for drone signal detection in presence of multiple signals, the system comprises a capturing unit configured to: receive a plurality of RF communication signals from surrounding environment and convert the received RF communication signals into a plurality of digitized samples, a spectrogram computation and filtering unit configured to: receive the digitized samples from the capturing unit, generate spectrogram of the digitized samples and filter the generated spectrogram to reduce noise, wherein the generated spectrogram of the digitized samples comprises one or more frequency bins and one or more time bins, a signal detection unit configured to: receive the filtered spectrogram of the digitized samples from the spectrogram computation and filtering unit and detect a plurality of signals with varying power levels by applying a unique dynamic threshold for each frequency bin in the filtered spectrogram of the digitized samples, a merged signal identification unit configured to: receive the plurality of detected signals from the signal detection unit, identify presence of merged signals in the plurality of detected signals and split the merged signals for detecting signals which are overlapped in at least one of time domain and frequency domain of the filtered spectrogram, a signal parameters estimation unit configured to: receive the detected signals and estimate parameters of the detected signals by applying image processing techniques on the detected signals, a signal parameters comparison unit configured to: receive the estimated signal parameters from the signal parameters estimation unit, compare the estimated signal parameters to signal parameters stored in a database, wherein the stored signal parameters are the signal parameters associated with one of different classes of RF communication signals and different types of drone devices and detect a drone signal, when the estimated signal parameters of at least one of the detected signals matches with the stored signal parameters of at least one type of drone device and an alert generation unit configured to generate an alert signal on detection of the drone signal.
Other aspects, advantages, and salient features of the invention will become apparent to those skilled in the art from the following detailed description, which, taken in conjunction with the annexed drawings, discloses exemplary embodiments of the invention.

Brief description of the drawings
The above and other aspects, features, and advantages of certain exemplary embodiments of the present invention will be more apparent from the following description taken in conjunction with the accompanying drawings in which:
Figure 1 shows a block diagram of drone signal detection system according to one embodiment of the present/proposed invention.
Figure 2 illustrates a flow diagram for drone signal detection and classification according to one embodiment of the present invention.
Figure 3 illustrates a flow diagram for signal detection according to one embodiment of the present invention.
Figure 4 illustrates a flow diagram for estimation of signal parameters according to one embodiment of the present invention.
Figure 5 illustrates various units/blocks in the drone signal detection system according to one embodiment of the present invention.
Figure 6 illustrates a flow diagram of a method for drone signal detection according to one embodiment of the present invention.
Persons skilled in the art will appreciate that elements in the figures are illustrated for simplicity and clarity and may have not been drawn to scale. For example, the dimensions of some of the elements in the figure may be exaggerated relative to other elements to help to improve understanding of various exemplary embodiments of the present disclosure/invention.
Throughout the drawings, it should be noted that like reference numbers are used to depict the same or similar elements, features, and structures.

Detailed description of the invention

The following description with reference to the accompanying plots/drawings is provided to assist in a comprehensive understanding of exemplary embodiments of the invention as defined by the claims and their equivalents. It includes various specific details to assist in that understanding but these are to be regarded as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. In addition, descriptions of well-known functions and constructions are omitted for clarity and conciseness.
The terms and words used in the following description and claims are not limited to the bibliographical meanings but are merely used by the inventor to enable a clear and consistent understanding of the invention. Accordingly, it should be apparent to those skilled in the art that the following description of exemplary embodiments of the present invention are provided for illustration purpose only and not for the purpose of limiting the invention as defined by the appended claims and their equivalents.
It is to be understood that the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to “a component surface” includes reference to one or more of such surfaces.
Figures discussed below, and the various embodiments used to describe the principles of the present disclosure/invention in this patent document are by way of illustration only and should not be construed in any way that would limit the scope of the disclosure/invention. Those skilled in the art will understand that the principles of the present disclosure/invention may be implemented in any suitably arranged system. The terms used to describe various embodiments are exemplary. It should be understood that these are provided to merely aid the understanding of the description, and that their use and definitions in no way limit the scope of the invention. Terms first, second, and the like are used to differentiate between objects having the same terminology and are in no way intended to represent a chronological order, unless where explicitly stated otherwise. A set is defined as a non-empty set including at least one element.
The present invention mainly relates to a field of monitoring technology and drone signal detection. More particularly the present invention relates to a method and a system for drone signal detection in presence of multiple signals. The method and the system of the present invention detect presence of drone signals by passively listening to Radio Frequency (RF) signals in surrounding environment.
In one embodiment, the main objective of the present invention is to provide the method and the system for the drone signal detection in the presence of the multiple signals, which may be signals either from interferers or from multiple drones, wherein a proper balance between drone detection accuracy and false alarm rate is achieved by incorporating dynamic thresholding. The present invention includes applying the dynamic thresholding on spectrogram of the received or captured RF signals for detection of the signals with varying power levels. Further, the present invention also includes application of image processing techniques on the spectrogram for signal detection, identification and splitting of merged signals.
In one embodiment, the present invention discloses the method and the system for drone detection and classification in presence of the interferers. The interferers may comprise the signals such as but not limited to Bluetooth, Wi-Fi, and the like.
In one embodiment, the present invention addresses the detection or the identification of the multiple signals overlapped in time domain and/or frequency domain (the merged signals) of the spectrogram of the received or captured RF signals effectively thereby increasing recognition accuracy of the drone detection. The identification of the merged signals is important for correct estimation of signal parameters (bandwidth, time duration, centre frequency, etc.) and for accurate detection of a drone signal merged with the multiple signals. The multiple signals may comprise the plurality of interfering RF communication signals which may also include the signals from multiple drone devices.
The present invention involves a database creation wherein individual signals or classes of the RF communication signals and different types of drone devices are independently captured, analyzed and signal parameters or signal features such as bandwidth, time duration, centre frequency, etc. are stored in a database. The individual signals or the classes of the RF communication signals may comprise but is not limited to the signals such as Bluetooth, Wi-Fi, or may be the signals of the different types of the drone devices, and the like. Further, the signal parameters of the detected signals are extracted or estimated by applying the image processing techniques on the spectrogram. The present invention also involves drone signal classification which involves matching the extracted or the estimated signal parameters of the detected signals against the stored parameters in the database. An alert signal is generated on detection of a drone signal. The present invention also has provision to further add new drone signatures for classification.
In one embodiment, the present invention discloses a system for drone signal detection comprising of sensor(s), FPGA board(s) and embedded processor(s), wherein the system detects the presence of the drone signals by passively listening to the RF signals in the surrounding environment.
Figure 1 shows a block diagram of drone signal detection system according to one embodiment of the present/proposed invention.
The figure shows the drone signal detection system or drone detection system, where the system comprises of one or more sensors 102 capable of receiving RF signals in surrounding environment or in vicinity and an analog receiver 104 which includes a local oscillator (LO) and a mixer to down-convert the received RF signals. The sensors 102 used in the system must also be capable of monitoring and receiving signals of interest, for example, drone RF communication signals. The RF signals may be, but not limited to RF communication signals such as Bluetooth, Wi-Fi, or may be signals of different types of drone devices, and the like. The RF signals received by the analog receiver 104 are digitized using an Analog to Digital Converter (ADC) after passing through an Automatic Gain Control (AGC) Unit 106. The ADC converts the received RF signals into digitized samples.
An embedded processor unit 110 issues commands/instructions to the analog receiver circuitry for setting local oscillator (LO) frequency, bandwidth and sampling rate of the received RF signals (the signals being received) depending on the signals of interest. The embedded processor unit 110 also sends commands to control the AGC. The received digitized samples are stored in a memory unit 108 as per instructions given by the embedded processor unit 110. The embedded processor unit 110 analyzes the digitized samples stored in the memory unit 108 to detect the presence of a drone signal in the surrounding environment or in surroundings. After detection of the drone signal, the system for the drone signal detection raises an alert using an output unit 112.
In an embodiment, one or more sensors 102 may be antennas capable of receiving or capturing or intercepting the RF signals in the surrounding environment and receiving the signals of interest (the drone RF communication signals) with sufficient bandwidth.
In an embodiment, the drone RF communication signals may comprise RF communications of the drone devices, for example, the RF communications to and from a drone device, and the like.
In an embodiment, the alert may be generated on detecting an unauthorized drone device. In an embodiment, the alert may comprise, but not limited to an email message, a text message, a visual alert, an audio alert or an audio warning, and the like.
In an embodiment, the output unit 112 may comprise, but not limited to a display device, a printer, a speaker configured to output the audio alert or warning, a light source configured to output the visual alert, and the like.
In an embodiment, the system for drone signal detection, detects the drone signal when multiple signals are present, wherein the multiple signals may comprise a plurality of interfering RF communication signals which may also include signals from multiple drone devices. The interfering RF signals may be, but not limited to signals such as Bluetooth, Wi-Fi, and the like.
Figure 2 illustrates a flow diagram for drone signal detection and classification according to one embodiment of the present invention.
The figure shows steps involved in the drone signal detection and classification performed by the embedded processor unit 110. In cases where signal strength is weak, it is almost impossible to detect presence of a signal by analyzing a time domain waveform. The same signal becomes evident when viewed in a frequency domain. Therefore, all the analysis for the drone signal detection and classification is done in the frequency domain by computing spectrogram for captured data 202. For example, the one or more sensors 102 capture RF communication signals in the surrounding environment or in the vicinity and the captured RF communication signals are converted into the digitized samples and the spectrogram of the digitized samples is generated.
Pre-processing of acquired data is necessary to reduce noise present in the signal. The computed spectrogram is filtered 204 by passing through a 2D – low pass filter to reduce the noise in the signal. The filtered data/spectrogram is then passed on to signal detection module 206 to detect any signal present in the captured data. If no signal is detected at this stage, then there is no need to proceed further in process of the drone signal detection. In cases where the signal is detected, parameters of the signal (signal parameters) such as bandwidth, time duration, centre frequencies, etc. are estimated using signal parameters estimation module 208. Then, signal classification module 210 compares the estimated or extracted signal parameters with stored database and classifies the signal accordingly. The estimated signal parameters are compared to signal parameters stored in a database. The stored signal parameters are the signal parameters associated with one of different classes of the RF communication signals and different types of the drone devices. For example, the signal parameters corresponding to each class, like, Bluetooth, Wi-Fi, and the different types of the drone devices are stored in the database. So once the signal parameters of the detected signal are estimated, signal classification is done by comparing the estimated signal parameters to the signal parameters stored in the database. The drone signal is detected when the estimated signal parameters of at least one of the detected signals matches with the stored signal parameters of at least one type of drone device.
Figure 3 illustrates a flow diagram for the signal detection according to one embodiment of the present invention.
The figure shows steps involved in the signal detection in detail, wherein the proper balance between the drone detection accuracy and the false alarm rate is achieved by incorporating the dynamic thresholding in order to detect the signals effectively.
N point Fast Fourier transform (N point FFT) is used while computing the spectrogram, (i.e.), there are N frequency bins spanning the bandwidth of the captured RF signal. The factor ‘N’ is tunable while computing Short-Time Fourier Transform (STFT) depending on required resolution. For each frequency bin, a difference between a maximum power and a mean power is calculated 302. Intuitively, for any frequency bin, this difference between the maximum power and the mean power will be high if a signal is present and will be low if there is no signal. The calculated max mean power difference values are smoothened by averaging to eliminate very small spikes arising due to the noise 304. In order to determine the frequency bins with the signal/signals, noise floor is estimated 306. The noise floor is calculated using mode of the max mean power difference values. In other words, histogram of the max mean power difference values is estimated and the histogram mean value with maximum number of the frequency bins is used in estimating the noise floor.
When the signal is present at certain frequencies, some power gets leaked into the frequencies close by during detection. This may degrade bandwidth estimation accuracy if all the frequency bins with difference values greater than the estimated noise floor are considered for processing. To overcome this problem, valid peaks in the max mean difference plot/values are detected first 308. For every peak, the frequency bins for further processing are wisely selected based on a dynamic threshold set using the maximum difference value in the peak 310.
After selecting the frequency bins for processing, time bins with the signal/signals are to be detected pertaining to the selected frequency bins. In the present invention, a unique threshold is decided for each frequency bin (depending on minimum, mean and maximum power levels in the frequency bin) for detecting the signal along a time axis. At this stage, a binarized data/image is obtained from the spectrogram data by imposing various constraints through the dynamic thresholding.
The present invention comprises a method of the dynamic thresholding for detecting the signals with varying power levels present in the captured data by applying the unique dynamic threshold for each frequency bin of the spectrogram by computing the maximum power, the mean power and the minimum power levels in each frequency bin.
Figure 4 illustrates a flow diagram for estimation of signal parameters according to one embodiment of the present invention.
The figure shows steps involved in the estimation of signal parameters in detail. Correct estimation of the signal parameters is crucial in the signal detection and classification. The binarized data obtained from the signal detection module is treated as an image and image processing methodologies/techniques are efficiently utilized to extract the signal parameters. Dilation and erosion 402 are performed on the binarized image to remove the noise. Then, contours in the binarized image are extracted 404.
One major challenge is differentiating the noise and a signal of very short duration and very less bandwidth. Conditions on area, mean and variance of a contour 406 are enforced in order to select only valid contours thereby ignoring invalid signals. For all the valid contours, top left and bottom right corner details of a rectangle enclosing the contour are used to find a start time, an end time, a start frequency and an end frequency of the signal. Using these details, the signal parameters such as the bandwidth, the time duration, the centre frequency, etc. are estimated 408.
In low power signal cases, splitting of a signal may happen sometimes. Merging of such signals is done by enforcing constraints on time and bandwidth of the signals 410. After estimation of the signal parameters of detected signals, the merging of split signals of low power is done by enforcing constraints on the time duration and the bandwidth of the detected signals. In an embodiment, on estimating the parameters of the detected signals, two signals are merged if any one of the following two constraints are met: (a) both the signals occupy the same bandwidth and one of the signal’s start time is very close to the other signal’s end time (b) both the signals occupy the same time duration and one of the signal’s start frequency is very close to the other signal’s end frequency.
There are various scenarios that arise when the multiple signals are present in the captured data such as – (a) two signals overlapped in time but separated in frequency; (b) two signals overlapped in the frequency but separated in the time; (c) two signals overlapped in both the time and the frequency and so on. The signals will remain separated in the spectrogram for the first two cases (a) and (b) as they are separated either in the time or the frequency. But the signals will be merged in the spectrogram for the case (c) as they are overlapped in both the time and the frequency. Estimation of the signal parameters may go wrong in such cases. So, identification of such overlapped cases is of prime importance.
In case of merged signals, there will be a bandwidth variation and a time duration variation, (i.e.), the bandwidth varies along the time and the time duration varies along the frequency for the merged signals. These variations are utilized in identification and splitting of the merged signals and thereby, enabling the estimation of correct signal parameters for the merged signals. In the signal classification module, the estimated signal parameters are matched against the stored parameters for different classes. If match is found, the system outputs the presence of matched signals.
The present invention comprises of a method of identifying and splitting the merged signals by analysing the bandwidth variation of the merged signals along time axis and the time duration variation of the merged signals along frequency axis in the spectrogram for estimating the correct signal parameters of the drone signal merged with the multiple signals.
Figure 5 illustrates various units/blocks in the drone signal detection system according to one embodiment of the present invention.
The figure shows the system for drone signal detection in presence of multiple signals where the system comprises a capturing unit 502, a spectrogram computation and filtering unit 504, a signal detection unit 506, a merged signal identification unit 508, a signal parameters estimation unit 510, a signal parameters comparison unit 512 and an alert generation unit 514.
The drone signal detection system comprises of the capturing unit 502 to capture and digitize the RF communication signals in the vicinity. For example, the capturing unit 502 is configured to receive a plurality of RF communication signals from the surrounding environment and convert the received RF communication signals into a plurality of digitized samples.
The spectrogram computation and filtering unit 504 computes the spectrogram of the captured data and filters the computed spectrogram to reduce noise. For example, the spectrogram computation and filtering unit 504 is configured to receive the digitized samples from the capturing unit 502, generate spectrogram of the digitized samples and filter the generated spectrogram to reduce noise, wherein the generated spectrogram of the digitized samples comprises one or more frequency bins and one or more time bins.
The signal detection unit 506 detects signals by applying the unique dynamic threshold for each frequency bin of the spectrogram using the maximum power, the mean power and the minimum power in each frequency bin. For example, the signal detection unit 506 is configured to receive the filtered spectrogram of the digitized samples from the spectrogram computation and filtering unit 504 and detect a plurality of signals with varying power levels by applying the unique dynamic threshold for each frequency bin in the filtered spectrogram of the digitized samples.
The merged signal identification unit 508 identifies the presence of the merged signals (signals overlapped in time domain and/or frequency domain) in order to detect the drone signals merged with other interfering signals. For example, the merged signal identification unit 508 is configured to receive the plurality of detected signals from the signal detection unit 506, identify presence of merged signals in the plurality of detected signals and split the merged signals for detecting signals which are overlapped in at least one of time domain and frequency domain of the filtered spectrogram.
The signal parameters estimation unit 510 extracts the parameters of the detected signals’ such as but not limited to the bandwidth, the time duration, the centre frequency, etc. by applying the image processing techniques on the spectrogram. For example, the signal parameters estimation unit 510 is configured to receive the detected signals and estimate parameters of the detected signals by applying image processing techniques on the detected signals.
The signal parameters comparison unit 512 compares the extracted signal parameters against the stored signal parameters in the database. For example, the signal parameters comparison unit 512 is configured to receive the estimated signal parameters from the signal parameters estimation unit 510, compare the estimated signal parameters to signal parameters stored in the database, wherein the stored signal parameters are the signal parameters associated with one of different classes of the RF communication signals and the different types of the drone devices and detect the drone signal, when the estimated signal parameters of at least one of the detected signals matches with the stored signal parameters of at least one type of the drone device.
The alert generation unit 514 to alert the user in case a match is found with at least one of the drone signatures stored in the database. For example, the alert generation unit 514 is configured to generate an alert signal on detection of the drone signal.
In an embodiment, the capturing unit 502 comprises of one or more sensors 102 configured to capture the plurality of RF communication signals in the surrounding environment. The sensors 102 are capable of intercepting the RF communication signals in the vicinity and receiving the signals of interest - drone RF communication signals with sufficient bandwidth.
In an embodiment, the embedded processor unit 110 may comprise the spectrogram computation and filtering unit 504, the signal detection unit 506, the merged signal identification unit 508, the signal parameters estimation unit 510 and the signal parameters comparison unit 512.
In an embodiment, the alert generation unit 514 may be the output unit 112.
In an embodiment, the embedded processor unit 110 may comprise the alert generation unit 514 for creating alerts on detecting the presence of the drone signal in the surroundings.
Figure 6 illustrates a flow diagram of a method for drone signal detection, according to one embodiment of the present invention. Figure 6 shows the method for the drone signal detection in the presence of the multiple signals. Referring to Figure 6, at step 602, the method comprises receiving the plurality of RF communication signals from the surrounding environment and converting the received RF communication signals into the plurality of digitized samples by the capturing unit 502. The capturing unit 502 comprises of one or more sensors 102 capable of intercepting the RF communication signals in the surrounding environment or in the vicinity and receiving the signals of interest – the drone RF communication signals with sufficient bandwidth. At step 604, the method comprises generating the spectrogram of the digitized samples, wherein the spectrogram of the digitized samples comprises one or more frequency bins and one or more time bins and filtering the generated spectrogram to reduce noise by the spectrogram computation and filtering unit 504.
At step 606, the method comprises detecting the plurality of signals with varying power levels by applying the unique dynamic threshold for each frequency bin in the filtered spectrogram of the digitized samples by the signal detection unit 506.
At step 608, the method comprises identifying the presence of the merged signals in the plurality of the detected signals and splitting the merged signals for detecting signals which are overlapped in at least one of the time domain and the frequency domain of the filtered spectrogram by the merged signal identification unit 508.
At step 610, the method comprises estimating parameters of the detected signals by applying the image processing techniques on the detected signals by the signal parameters estimation unit 510.
At step 612, the method comprises comparing the estimated signal parameters to the signal parameters stored in the database by the signal parameters comparison unit 512, wherein the stored signal parameters are the signal parameters associated with one of different classes of the RF communication signals and the different types of the drone devices. Further, at step 614, the signal parameters comparison unit 512 detects the drone signal, when the estimated signal parameters of at least one of the detected signals matches with the stored signal parameters of at least one type of the drone device.
In an embodiment, the alert generation unit 514 generates the alert signal on detection of the drone signal.
In an embodiment, the steps involved in the method for drone signal detection may be performed by the embedded processor unit 110.
In an embodiment, the embedded processor unit 110 may comprise the spectrogram computation and filtering unit 504, the signal detection unit 506, the merged signal identification unit 508, the signal parameters estimation unit 510 and the signal parameters comparison unit 512.
Unlike conventional methods and systems, the proposed invention detects the drone signal in the presence of the multiple signals and achieves the proper balance between the drone detection accuracy and the false alarm rate by incorporating the dynamic thresholding. Further, in the proposed invention, the presence of the merged signals is identified for estimating the correct signal parameters of the drone signal merged with the multiple signals.
Figures are merely representational and are not drawn to scale. Certain portions thereof may be exaggerated, while others may be minimized. Figures illustrate various embodiments of the invention that can be understood and appropriately carried out by those of ordinary skill in the art.
In the foregoing detailed description of embodiments of the invention, various features are grouped together in a single embodiment for the purpose of streamlining the disclosure/invention. This device or unit or arrangement of disclosure/invention is not to be interpreted as reflecting an intention that the claimed embodiments of the invention require more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive subject matter lies in less than all features of a single disclosed embodiment. Thus, the following claims are hereby incorporated into the detailed description of embodiments of the invention, with each claim standing on its own as a separate embodiment.
It is understood that the above description is intended to be illustrative, and not restrictive. It is intended to cover all alternatives, modifications and equivalents as may be included within the spirit and scope of the invention as defined in the appended claims. Many other embodiments will be apparent to those of skill in the art upon reviewing the above description. The scope of the invention should, therefore, be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled. In the appended claims, the terms “including” and “in which” are used as the plain-English equivalents of the respective terms “comprising” and “wherein,” respectively.