In a first method, a wireless device estimates a delay profile of a channel impulse response, CIR, for a channel between a network node and the wireless device, compresses the delay profile using a compression function, and transmits the compressed delay profile. The compression function includes a first function and a quantizer. The first function is configured to receive input data and reduce a dimension of the input data. In a second method, a network node receives a compressed delay profile of CIR for a channel between a network node and a wireless device, decompresses the compressed delay profile using a decompression function, and estimates a position of the wireless device based on at least the decompressed delay profile. The decompression function includes a first function which is configured to receive input data and provide output data in a higher dimensional space than the input data.

Systems and methods for encoding and decoding neural network data is provided. A method includes: obtaining an independent neural network with a topology; encoding the independent neural network with the topology such as to obtain a neural network representation (NNR) bitstream; and sending the NNR bitstream to a decoder, wherein the NNR bitstream includes a group of NNR units (GON) that represents the independent neural network with the topology, and the GON includes an NNR model parameter set unit, an NNR layer parameter set unit, an NNR topology unit, an NNR quantization unit, and an NNR compressed data unit.

A decompression system has a plurality of decompression devices in an array or chain layout for decompressing respective compressed data values of a compressed data block. A first decompression device is connected to a next decompression device, and a last decompression device is connected to a preceding decompression device. The first decompression device decompresses a compressed data value and reduces the compressed data block by extracting a codeword of the compressed data value and removing the compressed data value from the compressed data block, retrieving a decompressed data value out of the extracted codeword, and passing the reduced compressed data block to the next decompression device. The last decompression device receives a reduced compressed data block from the preceding decompression device and decompresses another compressed data value by extracting a codeword of the other compressed data value, and retrieving another decompressed data value out of the extracted codeword. Elected for publication; FIG. 8.

A device configured to compress a tensor including a plurality of cells includes: a quadtree generator configured to generate a quadtree searching for a non-zero cell included in the tensor and extract at least one parameter value from the quadtree; a mode selector configured to determine a compression mode based on the at least one parameter; and a bitstream generator configured to generate a bitstream by compressing the tensor based on the compression mode.

Apparatus and methods for training a neural network accelerator using quantized precision data formats are disclosed, and in particular for storing activation values from a neural network in a compressed format having lossy or non-uniform mantissas for use during forward and backward propagation training of the neural network. In certain examples of the disclosed technology, a computing system includes processors, memory, and a compressor in communication with the memory. The computing system is configured to perform forward propagation for a layer of a neural network to produced first activation values in a first block floating-point format. In some examples, activation values generated by forward propagation are converted by the compressor to a second block floating-point format having a non-uniform and/or lossy mantissa. The compressed activation values are stored in the memory, where they can be retrieved for use during back propagation.

A method and a device for matching a quantization table of data representative of a radio signal received by a radio antenna of a mobile network. The method includes: obtaining an item of information representative of a channel decoding error rate of a decoded quantized demodulated signal from a demodulation of the radio signal received by the antenna, the demodulated radio signal having been quantized by the quantization table, and the quantized demodulated radio signal having undergone a channel decoding; matching the quantization table when the channel decoding error rate is higher than a determined threshold; and transmitting an item of information representative of the matching of the quantization table to a channel decoding device or to a demodulation device.

In some examples, a method for compressing a spectral reflectance dataset may be performed through compression circuitry. The method may include computing a principal component analysis basis for the spectral reflectance dataset; projecting the spectral reflectance dataset onto the principal component analysis basis to obtain a weight matrix; quantizing the weight matrix; performing a Huffman encoding process on the quantized weight matrix to generate a Huffman table and Huffman codes for the quantized weight matrix; and providing compressed spectral reflectance data as the principal component analysis basis, the Huffman table, and the Huffman codes.

A distributed radio frequency communication system facilitates communication between a wireless terminal and a core network. The system includes a remote radio unit (RRU) coupled to at least one antenna to communicate with the wireless terminal. The RRU includes electronic circuitry to perform at least a first portion of a first-level protocol of a radio access network (RAN) for communicating between the wireless terminal and the core network. The system also includes a baseband unit (BBU) coupled to the core network, and configured to perform at least a second-level protocol of the RAN. A fronthaul link is coupled to the BBU and the RRU. The fronthaul link utilizes an adaptive fronthaul protocol for communication between the BBU and the RRU. The adaptive fronthaul protocol has provisions for adapting to conditions of the fronthaul link and radio network by changing the way data is communicated over the fronthaul link.

Embodiments of the present invention provide a system for fast parallel graph compression based on identifying a set of large cliques, which is used to encode the graph. The system provides both permanently-stored and in-memory graph encoding and reduces the space needed to represent and store a graph, the I/O traffic to use the graph, and the computation needed to perform algorithms involving the graph. The system thereby improves computing technology and graph computation. During operation, the system obtains data indicating vertices and edges of a graph. The system executes a clique-finding method to identify a maximum clique in the graph. The system then removes the clique from the graph, adds the clique to a set of found cliques, and generates a compressed representation of the graph based on the set of found cliques.

This disclosure relates to systems and methods for adaptively compressing data based on compression parameters. In one embodiment, a method for compressing a dataset is disclosed, including filtering a dataset based on occurrence of an event, and determining a quality of information index indicating a measure of quality of the dataset based on a quality of information estimation function. The method may include comparing the quality of information index with a list of indices stored in a lookup table to identify a target quality of information index and corresponding compression parameters, wherein the target quality of information index is indicative of a reference measure of quality of the dataset applicable for deriving statistical inferences based on analysis of the dataset. Also, the method may include inputting the compression parameters in a compression algorithm for compressing the dataset to achieve the target quality of information index for the analysis.