Systems, methods, and computer-readable media for call classification and for training a model for call classification, an example method comprising: receiving DTMF information from a plurality of calls; determining, for each of the calls, a feature vector including statistics based on DTMF information such as DTMF residual signal comprising channel noise and additive noise; training a model for classification; comparing a new call feature vector to the model; predicting a device type and geographic location based on the comparison of the new call feature vector to the model; classifying the call as spoofed or genuine; and authenticating a call or altering an IVR call flow.

Systems, methods, and computer-readable media for call classification and for training a model for call classification, an example method comprising: receiving DTMF information from a plurality of calls; determining, for each of the calls, a feature vector including statistics based on DTMF information such as DTMF residual signal comprising channel noise and additive noise; training a model for classification; comparing a new call feature vector to the model; predicting a device type and geographic location based on the comparison of the new call feature vector to the model; classifying the call as spoofed or genuine; and authenticating a call or altering an IVR call flow.

In a telephone switching apparatus equipped with an automatic call recording function, when recording of a call is started, a PB signal cannot be sent by operating a button any longer. Further, a user mistakenly assumes that there is a problem because the PB signal cannot be sent. Means capable of selecting, when recording means starts recording of a call, whether a PB signal is sent or not is provided, thus making it possible to send a PB signal.

In a telephone switching apparatus equipped with an automatic call recording function, when recording of a call is started, a PB signal cannot be sent by operating a button any longer. Further, a user mistakenly assumes that there is a problem because the PB signal cannot be sent. Means capable of selecting, when recording means starts recording of a call, whether a PB signal is sent or not is provided, thus making it possible to send a PB signal.

Systems, methods, and computer-readable media for call classification and for training a model for call classification, an example method comprising: receiving DTMF information from a plurality of calls; determining, for each of the calls, a feature vector including statistics based on DTMF information such as DTMF residual signal comprising channel noise and additive noise; training a model for classification; comparing a new call feature vector to the model; predicting a device type and geographic location based on the comparison of the new call feature vector to the model; classifying the call as spoofed or genuine; and authenticating a call or altering an IVR call flow.

Methods described herein relate to identifying an audio rogue base station intercept of a call using a modified distance bounding algorithm. Methods may include: establishing a shared resource for a session with a remote end device; selecting a nonce and provide the nonce to the remote end device; generating a sequence of bits; dividing the sequence of bits into a predetermined number of sub-sequences; performing a bit exchange with the remote end device using dual tone multi-frequency (DTMF) tones; determining, from the bit exchange, an average time of transmission; and identifying the presence or absence of an adversary intercepting the session.

Systems, methods, and computer-readable media for call classification and for training a model for call classification, an example method comprising: receiving DTMF information from a plurality of calls; determining, for each of the calls, a feature vector including statistics based on DTMF information such as DTMF residual signal comprising channel noise and additive noise; training a model for classification; comparing a new call feature vector to the model; predicting a device type and geographic location based on the comparison of the new call feature vector to the model; classifying the call as spoofed or genuine; and authenticating a call or altering an IVR call flow.

Systems, methods, and computer-readable media for call classification and for training a model for call classification, an example method comprising: receiving DTMF information from a plurality of calls; determining, for each of the calls, a feature vector including statistics based on DTMF information such as DTMF residual signal comprising channel noise and additive noise; training a model for classification; comparing a new call feature vector to the model; predicting a device type and geographic location based on the comparison of the new call feature vector to the model; classifying the call as spoofed or genuine; and authenticating a call or altering an IVR call flow.