Detection of spoofing and meaconing for geolocation positioning system signals
A computer architecture for geolocation spoofing/meaconing detection is disclosed. According to some aspects, a computer accesses an incoming geolocation positioning signal. The computer determines, using a signal characteristics calculation subsystem, geolocation positioning signal characteristics for the incoming geolocation positioning signal. The computer provides, using a geolocation positioning spoofing/meaconing detection subsystem, the geolocation positioning signal characteristics as an input vector to a neural network, wherein the neural network determines whether the incoming geolocation positioning signal is legitimate or fake. If the incoming geolocation positioning signal is determined to be fake: the computer computes, using a Bayesian inference subsystem, a likelihood and a severity of a geolocation positioning technology based attack. The computer provides, as a digital transmission, an indication of whether the incoming geolocation positioning signal is legitimate or fake.
1. A geolocation spoofing/meaconing detection apparatus, the apparatus comprising:
processing circuitry and memory; the processing circuitry to:
access an incoming geolocation positioning signal;
determine, using a signal characteristics calculation subsystem, geolocation positioning signal characteristics for the incoming geolocation positioning signal;
provide, using a geolocation positioning spoofing/meaconing detection subsystem, the geolocation positioning signal characteristics as an input vector to a neural network, wherein the neural network determines whether the incoming geolocation positioning signal is a legitimate geolocation positioning signal or a fake geolocation positioning signal, wherein the neural network comprises at least an input layer, one or more hidden layers, and an output layer;
in a case where the incoming geolocation positioning signal is determined to be the fake geolocation positioning signal: compute, using a Bayesian inference subsystem, a likelihood and a severity of a geolocation positioning technology based attack; and
provide, as a digital transmission, an indication of whether the incoming geolocation positioning signal is the legitimate geolocation positioning signal or the fake geolocation positioning signal and, in the case where the incoming geolocation positioning signal is determined to be the fake geolocation positioning signal, the likelihood and the severity of the geolocation positioning technology based attack.
2. The apparatus of claim 1 , wherein the geolocation positioning signal characteristics comprise one or more of: a geolocation positioning transmitter (GPT) location, a signal-to-noise ratio (SNR), a pseudo-range (PR), a Doppler shift (DO), and a carrier phase shift (CP).
3. The apparatus of claim 2 , wherein the GPT location is read from the incoming geolocation positioning signal.
4. The apparatus of claim 3 , wherein the GPT location comprises a satellite vehicle number (SVN).
5. The apparatus of claim 2 , wherein the SNR is computed based on Eigenvalues of a received signal covariance matrix of the incoming geolocation positioning signal.
6. The apparatus of claim 2 , wherein the PR is computed based on a travel time of a signal from the GPT to a geolocation positioning receiver (GPR), wherein the travel time is computed by cross-correlating a Gold code of the GPT and a replica of the Gold code generated at the GPR.
7. The apparatus of claim 2 , wherein the DO is computed based on an amplitude and a phase of the incoming geolocation positioning signal and an amplitude and a phase of a reference signal.
8. The apparatus of claim 2 , wherein the CP is computed based on a phase difference between the incoming geolocation positioning signal and a reference signal.
9. The apparatus of claim 1 , wherein the neural network comprises a plurality of neurons, wherein each neuron has an activation function.
10. The apparatus of claim 9 , wherein the activation function is a hyperbolic tangent (tanh) function.
11. The apparatus of claim 1 , wherein the neural network implements an online machine learning algorithm, wherein the online machine learning algorithm uses the geolocation positioning signal characteristics as the input vector.
12. The apparatus of claim 1 , wherein the Bayesian inference subsystem computes the likelihood and the severity of the geolocation positioning technology based attack based on a prior probability distribution of geolocation positioning technology based attacks and an attack presence probability, and wherein the prior probability distribution and the attack presence probability are updated based on processing of signals received via a geolocation positioning receiver.
13. The apparatus of claim 1 , further comprising a geolocation positioning receiver to:
receive the incoming geolocation positioning signal; and
provide the incoming geolocation positioning signal for access by the processing circuitry.
14. The apparatus of claim 11 , wherein the geolocation positioning receiver comprises a global positioning system (GPS) receiver.
15. A non-transitory machine-readable medium for geolocation spoofing/meaconing detection, the machine-readable medium storing instructions which, when executed by processing circuitry of one or more machines, cause the processing circuitry to:
access an incoming geolocation positioning signal;
determine, using a signal characteristics calculation subsystem, geolocation positioning signal characteristics for the incoming geolocation positioning signal;
provide, using a geolocation positioning spoofing/meaconing detection subsystem, the geolocation positioning signal characteristics as an input vector to a neural network, wherein the neural network determines whether the incoming geolocation positioning signal is a legitimate geolocation positioning signal or a fake geolocation positioning signal, wherein the neural network comprises at least an input layer, one or more hidden layers, and an output layer;
in a case where the incoming geolocation positioning signal is determined to be the fake geolocation positioning signal: compute, using a Bayesian inference subsystem, a likelihood and a severity of a geolocation positioning technology based attack; and
provide, as a digital transmission, an indication of whether the incoming geolocation positioning signal is the legitimate geolocation positioning signal or the fake geolocation positioning signal and, in the case where the incoming geolocation positioning signal is determined to be the fake geolocation positioning signal, the likelihood and the severity of the geolocation positioning technology based attack.
16. The machine-readable medium of claim 15 , wherein the geolocation positioning signal characteristics comprise one or more of: a geolocation positioning transmitter (GPT) location, a signal-to-noise ratio (SNR), a pseudo-range (PR), a Doppler shift (DO), and a carrier phase shift (CP).
17. The machine-readable medium of claim 15 , wherein the neural network implements an online machine learning algorithm, wherein the online machine learning algorithm uses the geolocation positioning signal characteristics as the input vector.
18. A geolocation spoofing/meaconing detection method comprising:
accessing an incoming geolocation positioning signal;
determining, using a signal characteristics calculation subsystem, geolocation positioning signal characteristics for the incoming geolocation positioning signal;
providing, using a geolocation positioning spoofing/meaconing detection subsystem, the geolocation positioning signal characteristics as an input vector to a neural network, wherein the neural network determines whether the incoming geolocation positioning signal is a legitimate geolocation positioning signal or a fake geolocation positioning signal, wherein the neural network comprises at least an input layer, one or more hidden layers, and an output layer;
in a case where the incoming geolocation positioning signal is determined to be the fake geolocation positioning signal: computing, using a Bayesian inference subsystem, a likelihood and a severity of a geolocation positioning technology based attack; and
providing, as a digital transmission, an indication of whether the incoming geolocation positioning signal is the legitimate geolocation positioning signal or the fake geolocation positioning signal and, in the case where the incoming geolocation positioning signal is determined to be the fake geolocation positioning signal, the likelihood and the severity of the geolocation positioning technology based attack.
19. The method of claim 18 , wherein the geolocation positioning signal characteristics comprise one or more of: a geolocation positioning transmitter (GPT) location, a signal-to-noise ratio (SNR), a pseudo-range (PR), a Doppler shift (DO), and a carrier phase shift (CP).
20. The method of claim 18 , wherein the neural network implements an online machine learning algorithm, wherein the online machine learning algorithm uses the geolocation positioning signal characteristics as the input vector.