Performance Analysis of Signature-Based Grant-Free Random Access Under Impersonation Attacks

In massive machine-type communications (mMTC), user-specific and non-orthogonal signatures can be assigned to wireless devices for uplink grant-free access. In this paper, we analyze the performance of signature-based grant-free random access in the presence of an adversary. To enhance the access security, we assume that an mMTC system renews the signatures of legitimate devices at each access secretly between the devices and a base station (BS). To gain unauthorized access to this mMTC system, we assume that the adversary attempts to impersonate legitimate devices by sending malicious signatures to the BS. By this impersonation attack, the adversary takes a chance to modify the data transmitted from legitimate devices and/or inject false data to the system. Moreover, this attack may deteriorate the performance of activity detection of legitimate devices through jamming. Under this scenario, we investigate the impacts of this impersonation attack on the performance of signature-based grant-free access theoretically and numerically, with respect to data modification, false authentication, and jamming.