Detection of unauthorized objects on ATM

A LoRa-wave sensing information-security process to protect against unauthorized objects on an automated teller machine (ATM) comprising the steps of:

• transmitting, by a LoRa modulator using a spread-spectrum modulation technique, a slot chirp signal to an ATM card slot and a keypad chirp signal to an ATM keypad;
• receiving, by the LoRa modulator, a card-slot signal pattern of the slot-chirp signal rebounding off of the ATM card slot and a keypad signal pattern of the keypad-chirp signal rebounding off the ATM keypad;
• scanning, by a transparent laser-guided system, the ATM slot;
• receiving, by the transparent laser-guided system, a received signal rebounded off of the ATM slot;
• creating, by the transparent laser-guided system based on the received signal, a 3D image of the ATM slot;
• transmitting, by the LoRa modulator to a sensor, the card-slot signal pattern and the keypad-signal pattern, said sensor infused with a DCNN/StNet-based algorithm trained on three-dimensional (3D) dimensions for the ATM card slot, 3D dimensions of the ATM keypad, and a weight of the ATM keypad;
• transmitting, by the transparent laser-guided system to the sensor, the 3D image;
• and
• detecting, by the sensor based on the 3D image as well as the DCNN/StNet-based algorithm, any dimensional change in the ATM slot;
• detecting, by the sensor, a first frequency-disrupted signal for the card-slot signal pattern if a skimmer is on the ATM card slot and a second frequency-disrupted signal for the keypad-signal pattern if a fake keypad is on the ATM keypad, said first frequency-disrupted signal and said second frequency-disrupted signal based on thermal variances, humidity variances, pressure variances, vibration, motion detection, and electrical current drawn from the ATM card slot exceeding a normal usage level;
• determining, by the sensor, a normal weight of the ATM keypad;
• determining, by the sensor, a sensed weight of the ATM keypad;
• detecting, by the DCNN/STNet-based algorithm in the sensor, whether the fake keypad is on the ATM keypad if the sensed weight constantly exceeds the normal weight;
• determining, by the sensor, a weight distribution for the ATM keypad when a key on the ATM keypad is depressed;
• detecting, by the DCNN/STNet-based algorithm in the sensor, whether the fake keypad is on the ATM keypad if the weight distribution is distributed equally over the ATM keypad;
• determining, by the sensor, that the fake keypad is not on the ATM keypad if the weight distribution is unevenly distributed over the ATM keypad;
• determining, by the sensor, whether a consensus exists that the fake keypad is on the ATM keypad or that the skimmer is on the ATM card slot;
• and
• triggering, by the sensor if the consensus exists, NB-IoT LoRa-waves to initiate an alarm process, and disabling the ATM.