Abstract: One use for delay adjustment circuit (32), coarse-grain delay offset circuit (34), and fine-grain delay synthesis circuit (36) may be as part of a delay replication circuit (30) used to replicate the frequency versus voltage behavior of an integrated circuit (29). Also, a circuit (30) and method for determining optimal power and frequency metrics of integrated circuit (29) is also described. In addition, a method for determining programmable coefficients to replicate frequency and supply voltage correlation is described.

Abstract: Behavior of an electrical circuit can be modeled using a trained neural network. For example, using one or more neural networks, power consumption, including leakage power and switching energy, can be estimated. Also, a profile of current versus time can be generated for the electrical circuit. A hierarchy of neural networks may be used to model the circuit at different levels. In one embodiment, a circuit behavior is modeled using one or more neural networks, cluster values, and cluster probabilities.

Abstract: A process for deriving a power transfer function of a circuit. The power transfer function can be used to represent the real time power consumption of a circuit based on the status of the inputs. In one embodiment, the power transfer function is derived from frequency domain analysis of signals applied to the inputs of a circuit during tests of the circuit. In one embodiment, the inputs of the circuit are grouped in groups based on a commonality of power consumption of the signals. The inputs may be grouped by clustering squared coherencies associated with the inputs. The transfer function may be implemented in a power monitoring circuit having inputs coupled to the inputs of the circuit to provide a real time estimation of power consumption of the circuit.

Abstract: Behavior of an electrical circuit can be modeled using a trained neural network. For example, using one or more neural networks, power consumption, including leakage power and switching energy, can be estimated. Also, a profile of current versus time can be generated for the electrical circuit. A hierarchy of neural networks may be used to model the circuit at different levels. In one embodiment, a circuit behavior is modeled using one or more neural networks, cluster values, and cluster probabilities.