WO2019220444A1

WO2019220444A1 - Authentication mechanism for drones

Info

Publication number: WO2019220444A1
Application number: PCT/IL2019/050556
Authority: WO
Inventors: Ben NASSI; Adi Shamir; Yuval Elovici
Original assignee: B. G. Negev Technologies And Applications Ltd., At Ben-Gurion University; Yeda Research And Development Co. Ltd
Priority date: 2018-05-16
Filing date: 2019-05-16
Publication date: 2019-11-21

Abstract

A method and a system are provided for authenticating in-coming unmanned aerial vehicles (UAVs) entering an airspace of a restricted site, including receiving at a site control unit (SCU) an electric signal from one or more of a microphone and a video camera, wherein the electric signal includes a modulated UAV signal originating from an airborne UAV, identifying the modulated UAV signal in the electric signal, processing the identified modulated UAV signal to extract a UAV-generated, time-based password, determining that the airborne UAV is a legitimate UAV when an SCU-generated, time-based password matches the UAV-generated, time -based password, and determining that the airborne UAV is not a legitimate UAV when the SCU- and UAV-generated, time -based passwords do not match.

Description

AUTHENTICATION MECHANISM FOR DRONES

FIELD OF THE INVENTION

[0001] The present invention relates to surveillance systems for monitoring air traffic of unmanned aerial vehicles and in particular to methods and systems for drone authentication.

BACKGROUND

[0002] The last few years have seen a proliferation of small unmanned aerial vehicles (UAVs), i.e., drones, for applications that include commercial, military, and hobbyist activities. The US Federal Aviation Administration (FAA) estimated that in 2017 there were three million drones purchased worldwide, thirty percent in the U.S. alone. Urban planners are beginning to plan smart cities with UAV flight zones and UAV highways. Applications for drones at commercial sites have begun to include perimeter surveillance and package delivery. Drones have the potential to gradually replace many types of ground-based delivery vehicles, meaning that traditional shipping and receiving docks will soon be serving drone fleets in addition to delivery trucks.

[0003] However, as drones proliferate, they also present new threats. They may be used maliciously by criminals and terrorists, either to cause physical attacks or to infiltrate organization premises for espionage. Drone defense is further complicated by the possibility of multiple drones attacking a site at once, potentially overwhelming systems for both detection and response and increasing the need for a rapid response to potential attackers. Methods of defending against malicious drones may include lasers to generate a high power weapon beam. However, differentiating enemy attackers from friendly drones is still problematic. SUMMARY

[0004] There is therefore provided, by embodiments of the present invention, a system for authenticating an airborne unmanned aerial vehicle (UAV) in a given airspace, the system including: a video camera oriented to acquire video of the given airspace and to generate a real time video stream; and a site control unit (SCU), comprising an SCU processor and associated SCU memory storage including SCU instructions that when executed by the SCU processor execute: receiving the video stream, identifying in the video stream a modulated light signal from the airborne UAV, processing the identified modulated light signal to extract a UAV-generated, time-based password, and determining that the airborne UAV is a legitimate UAV when an SCU-generated, time -based password matches the UAV-generated, time-based password.

[0005] The system may further include a UAV authentication module, configured to operate on a UAV to perform steps of: generating the UAV-generated, time -based password, based on a secret key shared with the SCU; and driving a UAV light emitter to emit a modulated light signal conveying the time-based password.

[0006] The time -based passwords of the UAV and of the SCU may be implemented by an algorithm such as: a “Hashed Message Authentication Code (HMAC), One-time password" (HOTP) generator, as specified in RFC 4226 of the Internet Engineering Task Force (IETF); a "Time-Based One-Time Password Algorithm" (TOTP) generator, specified in RFC 4226 of the IETF; or an RSA SecurlD™ authentication code generator.

[0007] The UAV may be further configured to identify a restricted site where the authentication is performed and to begin emitting the modulated light signal before entering the airspace of the restricted site. The UAV may be further configured to make deliveries, and the restricted site may be a delivery destination of the UAV. [0008] The UAV light emitter may be a visible light or infrared LED. In some embodiments, the UAV light emitter may be an LED that is continuously on, and the light emitted may be modulated by varying a voltage on an electrochromic filter.

[0009] Determining that the airborne UAV is legitimate may include issuing an authorization notice, and determining that the UAV is not legitimate may include issuing a security alert. The system may also include a drone defense subsystem. Issuing the security alert may include alerted the drone defense subsystem to intercept the UAV. The subsystem may include one or more interception technologies comprising one or more of an anti-drone laser and an anti-drone net launcher.

[0010] The video camera's frame rate may be at least 100 frames per second (FPS). The video camera may be located at a perimeter of a restricted area.

[0011] The video camera may be one of multiple video cameras, and the SCU may be configured to simultaneously process multiple video streams from the multiple video cameras. In some embodiments, identifying the modulated light signal may include simultaneously identifying multiple modulated light signals of multiple airborne UAVs in the video stream.

[0012] In further embodiments, the SCU may be further configured to process the video stream to determine a speed of the UAV or to determine a position of the UAV and may process video streams from multiple video cameras to triangulate the position of the UAV.

[0013] In further embodiments of the present invention, a system is provided for authenticating an unmanned aerial vehicles (UAVs) in a given airspace, the system including: a microphone collocated at the restricted site; and a site control unit (SCU) having an SCU processor and associated SCU memory storage, including SCU instructions that when executed by the SCU processor execute steps including: receiving an audio stream from the microphone, identifying a modulated audio signal from an airborne UAV in the audio stream, processing the identified modulated audio signal to extract a UAV- generated, time -based password, determining that the airborne UAV is a legitimate UAV when an SCU-generated, time-based password matches the UAV-generated, time-based password. Converserely, the SCU may determine that the airborne UAV is not a legitimate UAV when the SCU- and UAV-generated, time-based passwords do not match. The system may further include a UAV authentication module, configured to operate on a UAV to perform steps of: generating the UAV-generated, time-based password, based on a secret key shared with the SCU; and driving a UAV audio speaker to emit a modulated audio signal conveying the time-based password.

[0014] Also provided by embodiments of the present invention is a computer-based method for authenticating unmanned aerial vehicles (UAVs) in a given airspace, the method including: receiving, at a site control unit (SCU), an electric signal from one or more of a microphone and a video camera, wherein the electric signal includes a modulated UAV signal originating from an airborne UAV; identifying the modulated UAV signal in the electric signal; processing the identified modulated UAV signal to extract a UAV- generated, time -based password; and determining that the airborne UAV is a legitimate UAV when an SCU-generated, time-based password matches the UAV-generated, time- based password. The method may also include performing at one or more UAVs, each UAV comprising one or more of a UAV light emitter and a UAV audio speaker, further steps including: generating the UAV-generated, time-based password, based on a secret key shared with the SCU; and driving at least one of the UAV light emitters and UAV audio speakers to emit the modulated UAV signal conveying the time -based password BRIEF DESCRIPTION OF DRAWINGS

[0015] For a better understanding of various embodiments of the invention and to show how the same may be carried into effect, reference will now be made, by way of example, to the accompanying drawings. Structural details of the invention are shown to provide a fundamental understanding of the invention, the description, taken with the drawings, making apparent to those skilled in the art how the several forms of the invention may be embodied in practice. In the figures:

[0016] Fig. 1 is a schematic diagram of a system for authenticating drones, according to some embodiments of the present invention;

[0017] Fig. 2 is a schematic diagram of an alternative configuration of the system for authenticating drones, according to some embodiments of the present invention; and

[0018] Fig. 3 is a flow diagram, depicting a process of authenticating drones, according to some embodiments of the present invention.

DETAILED DESCRIPTION

[0019] Drones can be used for malicious purposes, including terrorist attacks and espionage. It is therefore necessary to protect sites that may have frequent drone operations against malicious drones. Such sites may include businesses that have frequent in-coming and out-going drone deliveries, such as factory sites. Agriculture areas, where drones may be used to monitor field conditions and to disburse pesticides, are another example of where it may be difficult to differentiate malicious drones from legitimate drones of the operators.

[0020] In embodiments of the present invention, a“white list” method is used to register legitimate drones. Registered drones are programmed to broadcast a time -based, encrypted message, for example by emitting a modulated visible light signal or modulated audio signal. Being time-based, the message cannot be easily imitated by illegitimate drones. At the site being protected, a site control unit decodes signals from legitimate drones and classifies drones as legitimate when a message can be authenticated. Drones that are not registered (i.e., not "whitelisted") will not send a message that can be authenticated and may be classified as illegitimate, which may then elicit a response to a potentially malicious, or "enemy," drone. As opposed to radio communications, light and audio signals are more easily associated with a specific airborne drone, such that illegitimate drones can be more quickly identified and intercepted by anti-drone methods.

[0021] Fig. 1 is a schematic diagram, depicting components of a system 10 for authenticating drones, according to some embodiments of the present invention. In an area in which there are legitimate drone activities, such as a business site receiving drone shipments or dispatching drone deliveries, system 10 is configured to protect the area against illegitimate drones. Hereinbelow, the area to be protected is referred to as a restricted area. Legitimate drones may be configured to perform autonomous activities, such that they can carry out activities, such as package delivery, without human operators guiding their activity in real time. To operate autonomously, legitimate drones may also include devices for identifying their own location, such as visual location identification systems and Global Positioning System (GPS) receivers.

[0022] In embodiments of the present invention, drones that are to enter or operate in the restricted area, that is, "legitimate" drones, such as drone 20, configured with one or more light emitters 22. The light emitters 22 may be LED devices, configured to emit light at one or more electromagnetic frequencies, typically in the visible or infrared range of the light spectrum. The drone 20 also includes a processor, such as a microcontroller, configured with an associated authentication module. The authentication module is a software or hardware component associated with the drone 20. The authentication module is programmed to generate a message and to drive the light emitters 22 to emit a modulated light signal including the message.

[0023] The generated message may be, for example, a time-based password (also referred to herein as a "time -based code"). A time-based password hinders attackers from emulating the modulated signal, that is, it hinders "replay attacks" by which a signal from a legitimate drone is recording and later replayed buy an illegitimate drone. Time -based password generation algorithms may include one of: a“Hashed message authentication code (HMAC), one-time password" (HOTP) generator, specified in RFC 4226 of the Internet Engineering Task Force (IETF); a "Time-Based One-Time Password Algorithm" (TOTP) generator, specified in RFC 4226 of the of the Internet Engineering Task Force (IETF); and an RSA SecurlD authentication code generator, a proprietary hardware or software generation method available from RSA Security LLC. Time-based password generators require that the sender and receiver of the message share a secret key, also referred to as a token or seed. The sender processes the share seed and the current time to generate the password, which is transmitted to the receiver. The receiver also processes the share seed and the current time to generate the password, and, if the passwords match, the sender is authenticated. Hereinbelow, a "match" is meant to mean that the passwords are the same, or can be correlated by further processing, such as by a pre-defined algorithm executed by the receiver.

[0024] An organization may assign a unique seed to each drone. Alternatively a common seed may be used for all drones. Drones that visit sites of other organizations, such as delivery drones, may have multiple secret keys (seeds), for example one key for each independent organization to which deliveries are made. The seed is typically stored in the memory of the authentication module of the drone. The drone operating system may also be secured to prevent hackers from accessing software and the seeds of the time -based password generator.

[0025] The light signal modulation of the light emitter may include any of a variety or combination of modulation methods. For example, on-off modulation may be employed at a predetermined bit-rate frequency. An on-off modulation frequency up to the frame rate of the video cameras may be employed, to ensure that the rate of bit pulses can be captured by the frame rate of the video cameras.

[0026] On-off modulation may be implemented either by turning on and off the light itself, or by setting the light behind an electrochromic filter. The transparency of the electrochromic filter may be controlled by modulating a voltage on the filter to hide or expose the light, the light being maintained in a constantly on state.

[0027] The drones may be configured to emit the modulated light signal constantly, or at regular intervals. Typically, the time code used for the time -based password generation is set to change at increments of 30 seconds or 1 minute, but can also be set to change at a rate sufficiently high to ensure that each message transmitted by a drone is generated with a different time code.

[0028] The drones may be configured to transmit continuously when operating within a restricted area's airspace, such that at any moment the legitimacy of a drone may be verified. The drones may also be configured to begin transmission when they approach a restricted area, for example, on return from flights to distant areas, such as return from customer delivery flights. Alternatively or additionally, drones may be configured to begin emitting modulated light in response to signaling from a site control unit, for example, in response to a signal broadcast by broadcast radio or Wi-Fi communication protocols. [0029] The system 10 also includes one or more video cameras 24, collocated within or at the perimeter of the restricted area and oriented with a view of the airspace above and surrounding the restricted area. Typically the video cameras are high speed cameras with a frame rate of at least a 100 frames per second (FPS).

[0030] Video streams from the video cameras are transmitted to a site control unit (SCU) 26, which may be located at the site or in a remote location. The SCU includes a processor subsystem 28, including a processor and memory. The SCU is programmed to receive the video streams from the one or more video cameras and to process each video stream to identify modulated light signals, the light modulation in a given video stream being identified by identifying pixel changes over multiple frames. When a modulated light signal is identified, the SCU processor subsystem then decodes the modulated message while also generating a time-based password, using the same generator as that installed on legitimate drones. If the decoded message matches the SCU-generated time -based password, the drone that transmitted the modulated light signal is determined to be legitimate. If the decoded message does not match, the drone is determined to be illegitimate.

[0031] When each drone is configured with a unique secret key (i.e., time-based encryption seed) and with a unique identifier, the drones may also issue the identifier in a manner that is not encrypted, allowing the SCU to first determine which corresponding seed to apply to generate the time-based password. Alternatively, the SCU may be configured to generate, on a continuous basis, time -based passwords for all seeds of registered drones, and then match the results with the received message.

[0032] The processor subsystem 28 may be programmed to issue a security alert if no message can be decoded or if a decoded, time-based password is not valid (that is, does not match a time-base password generated by the processor subsystem 28). Alternatively or additionally, the processor subsystem 28 may be programmed to issue an "authorization notice," confirming that a drone is legitimate, if a valid password is identified (by matching a processor subsystem-generated time-based password). The security alerts and authorization notices may be transmitted to a drone defense subsystem 30, which may provide enemy drone interception mechanisms, such as net launchers, radio jammers, anti drone drones, and/or lasers. Alternatively or additionally, the drone defense subsystem may perform additional functions, such as issuing alarms to human operators and notifying external security forces. Processing of signals may also include additional audio or visual processing of the received data streams, in order to determine additional features of drone flights, such as determining a position and speed of a drone. Processing the audio and/or video stream may include processing streams received by the SCU from multiple video camera and/or microphones to triangulate a position of a UAV.

[0033] Fig. 2 is a schematic diagram of an alternative configuration 40 of the system 10 for authenticating drones, according to some embodiments of the present invention. As shown, legitimate drones 50 may be configured with audio loudspeakers 52 (also referred to hereinbelow as speakers, or "sound emitters"), which may be instead of, or in addition to, the light emitters 22 described above with respect to Fig. 1.

[0034] The speakers may be directional speakers, such as parametric ultrasound speakers. Legitimate (i.e., registered) drones may be programmed to drive their speakers 22 to emit a modulated sound. The modulation encodes the time-based password generated by the drone (i.e., by the processor of the drone), as described above. [0035] The drones may be configured to emit the modulated audio signal constantly, or at regular intervals, or only when in the airspace of a site, or only after receiving a signal from the SCU, as in the manner described above with respect to Fig. 1.

[0036] The configuration 40 also includes one or more microphones 54, collocated within or at the perimeter of the restricted area and oriented to receive the audio transmissions.

[0037] Digital or analog audio streams from the microphones are received at the SCU 26, which may be configured to process either or both of input analog and video transmissions. The SCU processor subsystem 28 is configured to process each audio stream to identify modulated audio signals. The audio signal modulation may include any of a variety or combination of modulation methods. For example, two -level frequency modulation may be employed to switch between two different audio frequencies at a predetermined bit-rate frequency. The bit rate frequency is set to conform to the audio sampling rate of the SCU, which may be in the range, for example, of 20-100 HZ, though rates of thousands of pulses per second may also be implemented.

[0038] When a modulated audio signal is identified, the SCU decodes the transmitted message, as described above, which must match an SCU-generated time-based password to be valid. The processor subsystem 28 may be programmed to issue a security alert if no message can be decoded or if the decoded time -based password is not valid. Alternatively or additionally, the processor subsystem 28 may be programmed to issue an "authorization notice," confirming that a drone is legitimate, that is, if a valid message is identified. The security alerts and authorization notices may be transmitted to a drone defense subsystem

30, as described above. [0039] Fig. 3 is a flow diagram, depicting a process 300 of authenticating drones, according to some embodiments of the present invention. Steps of the process 300 are as follows.

[0040] At an initialization step 302, drones that are to be operated within a site are registered and configured for the authentication process. This includes both drones that are part of an internal drone fleet and drones that make site "visits," that is, drones operated by external parties, such as delivery drones. Initialization includes providing the drone with a one-time password generator that is also installed on the SCU, so that a one-time password transmit by the drone, by either a light or sound emitter, can be authenticated by the SCU.

[0041] At a subsequent step 304, the drone, while airborne, generates a time-based code. As described above, the drone may be configured to continuously generate the code, or only while approaching or within the airspace of a restricted site. The drone may also be configured to generate a code in response to an SCU query broadcast.

[0042] After generated a code, the drone, at a step 306, transmits the code by a modulated signal as described above. The method may be one or both of a visual or audio transmission method.

[0043] At a step 308, the SCU receives, from either a site video camera or a microphone, a respective video or audio data stream including the modulated signal (i.e., the signal transmitted by the drone). Using video or audio processing techniques, the SCU extracts the modulated signal from the respective data stream, and then decodes the signal to determine the included message. The SCU may also identify multiple drones in the video and or audio data streams, and may subsequently independently process the light and/or audio signals from each. [0044] At a step 310, the SCU then determines whether the received message is a valid time-based code. If the code is valid, that is, if the SCU can reproduce the same code, the SCU may issue an authentication notice, precluding defensive steps that may be automatically initiated if a drone is not authenticated. Additionally or alternatively, if a drone is not authenticated, the SCU may issue an alert, which may initiate measures to stop an illegitimate drone. Typically, to reproduce the codes of multiple drones, the SCU is configured with a list of all the time-based codes of all legitimate drones. Alternatively, an SCU at the restricted site may communicate with a remote SCU server configured with all time-based codes of legitimate drones.

[0045] It is to be understood that all or part of a process and of a system implementing the process of the present invention may be implemented in digital electronic circuitry, or in computer hardware, firmware, software, or in combinations thereof. All or part of the process and system may be implemented as a computer program product, tangibly embodied in an information carrier, such as a machine -readable storage device or in a propagated signal, for execution by, or to control the operation of, data processing apparatus, such as a programmable processor, computer, or deployed to be executed on multiple computers at one website or distributed across multiple websites. Memory storage may also include multiple distributed memory units, including one or more types of storage media. Examples of storage media include, but are not limited to, magnetic media, optical media, and integrated circuits. A computer configured to implement the process may access, provide, transmit, receive, and modify information over wired or wireless networks. The computing may have one or more processors and one or more network interface modules. Processors may be configured as a multi -processing or distributed processing system. Network interface modules may control the sending and receiving of data packets over networks.

[0046] It is to be further understood that the scope of the present invention includes variations and modifications thereof which would occur to persons skilled in the art upon reading the foregoing description and which are not disclosed in the prior art.

Claims

1. A system for authenticating unmanned aerial vehicles (UAVs) in a given airspace, comprising:

a video camera oriented to acquire video of the given airspace and to generate a real time video stream; and

a site control unit (SCU), comprising an SCU processor and associated SCU memory storage including SCU instructions that when executed by the SCU processor execute steps comprising:

receiving the video stream,

identifying in the video stream a modulated light signal from an airborne UAV, processing the identified modulated light signal to extract a UAV-generated, time-based password,

determining that the airborne UAV is a legitimate UAV when an SCU- generated, time-based password matches the UAV-generated, time-based password.

2. The system of claim 1, further comprising a UAV authentication module, configured to operate on a UAV to perform steps of:

generating the UAV-generated, time-based password, based on a secret key shared with the SCU; and

driving a UAV light emitter to emit a modulated light signal conveying the time-based password.

3. The system of claim 2, wherein generating the time-based passwords of the UAV and of the SCU is implemented by one of: a“Hashed Message Authentication Code (HMAC), One-time password" (HOTP) generator, specified in RFC 4226 of the Internet Engineering Task Force (IETF); a "Time-Based One-Time Password Algorithm" (TOTP) generator, specified in RFC 4226 of the IETF; or an RSA SecurlD™ authentication code generator.

4. The system of claim 2, wherein the UAV is further configured to identify a restricted site and to begin emitting the modulated light signal before entering the airspace of the restricted site.

5. The system of claim 2, wherein the UAV is further configured to make deliveries, and wherein the restricted site is a delivery destination of the UAV.

6. The system of claim 2, wherein the UAV light emitter is one of a visible light LED or an infrared LED.

7. The system of claim 2, wherein the UAV light emitter is an LED that is continuously on, and wherein the light emitted is modulated by varying a voltage on an electrochromic filter.

8. The system of claim 1, wherein determining that the airborne UAV is legitimate further comprises issuing an authorization notice from the SCU, and not determining that the airborne UAV is legitimate comprises issuing a security alert from the SCU.

9. The system of claim 8, further including a drone defense subsystem and wherein issuing the security alert comprises alerting the drone defense subsystem to intercept the airborne UAV.

10. The system of claim 9, wherein the drone defense subsystem comprises one or more interception technologies comprising one or more of an anti-drone laser and an anti-drone net launcher.

11. The system of claim 1, wherein the video camera has a frame rate of at least 100 frames per second (FPS).

12. The system of claim 1, wherein the video camera is one of multiple video cameras, and wherein the SCU is configured to simultaneously process multiple video streams from the multiple video cameras.

13. The system of claim 1, wherein identifying the modulated light signal comprises simultaneously identifying multiple modulated light signals of multiple airborne UAVs in the video stream.

14. The system of claim 1, wherein the video camera is located at a perimeter of a restricted area.

15. The system of claim 1, wherein the SCU is further configured to process the video stream to determine at least one of a speed and a position of the UAV.

16. The system of claim 15, wherein processing the video stream comprises processing video streams from multiple video cameras to triangulate the position of the UAV.

17. A system for authenticating unmanned aerial vehicles (UAVs) in a given airspace comprising:

a microphone collocated at the restricted site; and

a site control unit (SCU) comprising an SCU processor and associated SCU memory storage including SCU instructions that when executed by the SCU processor execute steps of: receiving an audio stream from the microphone, identifying a modulated audio signal from an airborne UAV in the audio stream, processing the identified modulated audio signal to extract a UAV-generated, time-based password, determining that the airborne UAV is a legitimate UAV when an SCU-generated, time- based password matches the UAV-generated, time-based password.

18. The system of claim 17, further comprising a UAV authentication module, configured to operate on a UAV to perform steps of:

driving a UAV audio speaker to emit a modulated audio signal conveying the time-based password.

19. A computer-based method for authenticating unmanned aerial vehicles (UAVs) in a given airspace, comprising:

receiving at a site control unit (SCU) an electric signal from one or more of a microphone and a video camera, wherein the electric signal includes a modulated UAV signal originating from an airborne UAV;

identifying the modulated UAV signal in the electric signal;

processing the identified modulated UAV signal to extract a UAV-generated, time- based password; and

determining that the airborne UAV is a legitimate UAV when an SCU-generated, time-based password matches the UAV-generated, time-based password.

20. The method of claim 19, further comprising performing at one or more UAVs, each UAV comprising one or more of a UAV light emitter and a UAV audio speaker, further steps comprising:

generating the UAV-generated, time-based password, based on a secret key shared with the SCU; and driving at least one of the UAV light emitters and UAV audio speakers to emit the modulated UAV signal conveying the time-based password.