Abstract:
A system and method for storing a power delay profile in compressed form. In some embodiments, the method includes: identifying, in a power delay profile including a first group of taps, a first arrival tap corresponding to a first arrival path; identifying a first subset of taps, the first subset being a subset of the first group and including the first arrival tap; identifying a second subset of taps, the second subset being a subset of the first group and being disjoint from the first subset; and storing, in a memory, the union of the first subset and a proper subset of the second subset.
Abstract:
A method and system for selecting a symbol detector are herein provided. A method includes extracting a first set of features for a k-th resource element (RE), where k is an integer greater than one, extracting a second set of features from a first RE to a (k−1)th RE, and selecting a symbol detector for the k-th RE using a reinforcement learning (RL) neural network based on the extracted first set of features and the extracted second set of features.
Abstract:
A method and system for multiple-input multiple-output (MIMO) detector selection using a neural network is herein disclosed. According to one embodiment, a method includes generating a labelled dataset of features and detector labels, training a multi-layer perceptron (MLP) network using the generated labelled dataset, and selecting a detector class from a plurality of detector classes based on outputs of the trained MLP network.
Abstract:
A system, method and device for wireless communication is provided. The method includes receiving, by a receiver, data from a transmitter, storing the data in the receiver, and determining, by the receiver, a probability of a bit stored in the data and a probability of a symbol based on the probability of the bit, wherein determining the probability of the bit includes moving a decision boundary associated with a constellation diagram.
Abstract:
A system and a method are disclosed for normalizing Log-Likelihood Ratios for bits of a transport block. For a narrowband channel estimation (NBCE), a normalization factor is selected based on an estimated delay spread, a cyclic prefix, an estimated Doppler spread, a rank, a modulation and coding scheme (MCS) and a signal-to-noise ratio (SNR) for the transport block, and an input LLRin for the individual bits of the transport block are scaled to respectively form an output LLRout for the individual bits of the transport block using the normalization factor. For a wideband channel estimation, a normalization factor is selected based on the rank/MCS/SNR of the transport block, and the input LLRin for the individual bits of the transport block are scaled to respectively form the output LLRout for the individual bits of the transport block using the normalization factor.
Abstract:
A method and system for training a neural network are herein provided. According to one embodiment, a method includes generating a first labelled dataset corresponding to a first modulation scheme and a second labelled dataset corresponding to a second modulation scheme, determining a first gradient of a cost function between a first neural network layer and a second neural network layer based on back-propagation using the first labelled dataset and the second labelled dataset, and determining a second gradient of the cost function between the second neural network layer and a first set of nodes of a third neural network layer based on back- propagation using the first labelled dataset. The first set of nodes of the third neural network layer correspond to a first plurality of detector classes associated with the first modulation scheme.
Abstract:
A system and method for receiving a quadrature amplitude modulation symbol. In some embodiments, the symbol has a plurality of bits and is associated with a point in a constellation of quadrature amplitude modulation points, each point of the constellation having associated with it a binary word. The method includes receiving a first analog signal carrying a modulation; performing initial estimation, to generate a first initial modulation estimate, for a portion of the first analog signal carrying a modulation; identifying, based on the first initial modulation estimate, a row of an initial candidate lookup table, the row corresponding to a region of the constellation; and reading from the row of the initial candidate lookup table a first plurality of initial candidate points of the constellation.
Abstract:
A method is provided for operating an electronic device. The method includes displaying GUI items; receiving a first touch input in respect to a first GUI; receiving a second touch input in respect to a second GUI; storing first information about a first external input object and a first drawing attribute, corresponding to the first GUI, associated with the first external input object in response to the first touch input; storing second information about a second external input object associated with a second drawing attribute, corresponding to the second GUI, in response to the second touch input; receiving a third touch input; determining whether the third touch input is input using the first external input object or the second external input object; identifying a drawing attribute for the third touch input from the stored information, based on the determination; and applying the identified drawing attribute to the third touch input.
Abstract:
A method and an apparatus are provided for performing a function corresponding to a recognized user input through a touch screen. The method includes displaying a window of an application; displaying a setting menu of the application; receiving a selection for at least one item in the setting menu; recognizing an input object that performed the selection; registering the input object as an object for executing a function corresponding to the at least one item selected by the input object; detecting an input by the input object in the application window; and executing the function in response to the input by the input object.
Abstract:
A system and method for configuring an RF network based on machine learning. In some embodiments, the method includes: receiving, by a first neural network, a first state and a first state transition, the first state including: one or more identifiers for available active ports, and a set of available connections between two or more circuit elements, each of the circuit elements being one of: (1) a first circuit type, (2) a second circuit type that operatively connects a circuit element of the first circuit type to one of the available active ports, and (3) the available active ports; and generating, by the first neural network, a first estimated quality value, for the first state transition.