Circuitry comprises a prediction register storing a plurality of entries each having respective data values for association with one or more branch instructions; prediction circuitry to detect, using prediction data derived by a mapping function from the stored data values associated with a given branch instruction, whether or not a branch represented by the given branch instruction is predicted to be taken; update circuitry to modify the stored data values associated with the given branch instruction in dependence upon a resolution of whether the branch represented by the given branch instruction is taken or not; and control circuitry configured to selectively alter one or more of the data values other than data values associated with the given branch instruction.

The disclosure concerns a method implemented by a processing device. The method includes performing a first execution by the processing device of a computing function based on one or more initial parameters stored in a first memory device. The execution of the computing function generates one or more modified values of at least one of the initial parameters, wherein during the first execution the one or more initial parameters are read from the first memory device and the one or more modified values are stored in a second memory device. The method also includes performing a second execution by the processing device of the computing function based on the one or more initial parameters stored in the first memory device.

A method is provided for performing a cryptographic operation in a white-box implementation on a mobile device. The cryptographic operation is performed in the mobile device for a response to a challenge from a mobile device reader. The mobile device reader includes a time-out period within which the cryptographic operation must be completed by the mobile device. In accordance with an embodiment, a first time period to complete the cryptographic operation on the mobile device is determined. A predetermined number of dummy computations are added to the cryptographic operation to increase the first time period to a second time period. The second time period is only slightly less than the time-out period by a predetermined safety value to make it less likely a relay attack with be successful.

A method of training a submodule and preventing capture of an AI module is disclosed. Input data received from an input interface is transmitted through a blocker module to an AI module, which computes a first output data by executing a first model. A submodule in the AI system trained using methods steps processes the input data to identify an attack vector from the input data. The submodule executes the first model and at least a second model. The first model and the second model have a first and second set of network parameters and hyper-parameters respectively. The identification information of the attack vector is sent to the information gain module.

There is provided a device of protecting an Integrated Circuit from perturbation attacks. The device includes a sensing unit configured to detect a perturbation attack, the sensing unit comprising a set of digital sensors comprising at least two sensors, the sensors being arranged in parallel. Each digital sensor provides a digitized bit output having a binary value, in response to input data, the sensing unit being configured to deliver at least one binary vector comprising a multi-bit value, the multi-bit value comprising at least two bit outputs provided by the set of digital sensors. The sensing device further comprising an analysis unit, the analysis unit being configured to receive at least one binary vector provided by the sensing unit, the analysis unit being configured to detect a perturbation attack from the at least one binary vector.

A calculation is performed on a first number and a second number. For each bit of the second number a first function is performed. The first function inputs include contents of a first register, contents of a second register and the first number. A result of the first function is placed in a third register. For each bit of the second number, a second function is performed which has as inputs contents of the third register and the contents of a selected one of the first and the second register according to a state of a current bit of the second number. A result of the second function is stored in the selected one of the first and second register.

A digital encrypting and decrypting unit (PMEU) that operates according to a Rivest-Shamir-Adleman (RSA) cryptosystem based on Residue Numeral System (RNS) and Chinese Reminder Theorem (CRT). The unit includes two modular exponentiation calculating units (MES-1, MES-2) to process a two residual signals (X mod p; X mod q) to calculate a result of a modular exponentiation by a binary method. The calculating units have inputs (I-k[i], I-SM, I-MM) and outputs (O-k[i], O-SM, O-MM) for signals representing partial results of the modular exponentiation. A modular exponentiation controlling unit (MECU) is connected to the inputs and outputs of the calculating units to control flow of the signals representing the partial results of the modular exponentiation.

An integrated circuit (IC) includes first and second secure memory elements storing identical data and a memory management system that executes a memory operation on the first secure memory element and a control operation on the second secure memory element simultaneously. The control operation is associated with safety of the IC and is executed to enable error detection in the second secure memory element, fault injection for the second secure memory element, masking of a power profile associated with the memory operation, or a combination thereof. After the execution of the memory operation and the control operation, the memory management system copies the data of the first secure memory element to the second secure memory element to maintain sanity of the second secure memory element.

A true random number generator (TRNG) uses an analog circuit with a ring oscillator configured to collapse from an unstable oscillation state to a stable oscillation state at a random collapse time and counter counting a counter value representing the random collapse time. Various techniques are used to harden the TRNG including a truncator generating a true random number based on a truncation of the reference count value and a dedicated voltage regulator supplying power to the analog core including the ring oscillator. Techniques also include various solutions for drawing a constant current such as using a Gray code counter and adding noise current during and/or after the collapse event with a dummy inverter chain. Bit churning, bit obfuscation entropy enhancers and various post processing techniques may be employed to further harden the TRNG. An attack detection module may raise alerts when the TRNG is being attacked.

There is disclosed a countermeasure using the properties of the Montgomery multiplication for securing cryptographic systems such as RSA and DSA against, in particular, safe-error injection attacks. In the proposed algorithm, the binary exponentiation b=a.sup.d mod n is iteratively calculated using the Montgomery multiplication when the current bit d.sub.i of the exponent d is equal to zero. In that case, the Montgomery multiplication of the actual result of the exponentiation calculation by R is realized. Thanks to this countermeasure, if there is any perturbation of the fault injection type introduced during the computation, it will have visible effect on the final result which renders such attack inefficient to deduce the current bit d.sub.i of the private key d.