Abstract: A system and method for processing predictions in the presence of adversarial perturbations in a sensing system. The processor receives inputs from sensors and runs a neural network having a network function that generates, as outputs, predictions of the neural network. The method generates from a plurality of outputs a measurement quantity (m) that may be, at or near a given input, either (i) a first measurement quantity M1 corresponding to a gradient of the given output, (ii) a second measurement quantity M2 corresponding to a gradient of a predetermined objective function derived from a training process for the neural network, or (iii) a third measurement quantity M3 derived from a combination of M1, and M2. The method determines whether the measurement quantity (m) is equal to or greater than a threshold. If greater than the threshold, one or more remedial actions are performed to correct for a perturbation.

Abstract: A method of sensing a target in a target detection system having processing circuitry and a multiplexer coupled to the processing circuitry and to a plurality NT of transmit antennas forming a sparse transmit uniform linear array (ULA), the multiplexer being configured to generate multiplexed and phase modulated transmit signals (T1 . . . TNT) based on signals from a local oscillator. The processing circuitry receives signals via a plurality NR of receive antennas forming a dense receive ULA. The method includes transmitting the transmit signals via the transmit antennas as a general radiation pattern corresponding to a block circulant probing signal matrix, and receiving via the receive antennas receive signals resulting from backscattering of the transmit signals transmitted towards K targets. The method further includes processing the received reflection signals to determine the presence, range and/or angular position of a target within a field of view of the transmit antennas.