Abstract:
A method of optimizing an iterative process defines a set of trainable parameters and a differentiable gating function to be applied to each parameter in the set of trainable parameters. A trainable model of the iterative process is built, wherein the iterative process is modified by using the value of the differentiable gating function applied to the parameters to compute a weighted sum of internal variables of the iterative process before and after each iteration. A machine learning-based optimization of the trainable model of the iterative process determines a subset of iterations of the iterative process to perform. The subset of iterations is determined such that an accuracy and a number of active iterations of the iterative process are jointly optimized. The method processes only the subset of the iterations to perform the iterative process. The method is applied to optimize the layered belief propagation algorithm for LDPC decoding.
Abstract:
The disclosure provides for interference mitigation for wireless signals in unlicensed spectrum. A wireless device may receive a combined signal including a first radio access technology (RAT) signal and a second RAT signal. The wireless device may generate, using a first RAT receiver in a first processing path, a channel estimate for the first RAT signal based on a previously decoded signal of the first RAT. The wireless device may reduce interference to the second RAT signal caused by the first RAT signal, in a second processing path, using the channel estimate. The wireless device may further decode the second RAT signal. The wireless device may remodulate the decoded signal using a transmitter to generate a remodulated second RAT signal. The remodulated second RAT signal may be canceled from the combined signal. The wireless device may decode a remaining portion of the combined signal including the first RAT signal.