Abstract: Apparatuses and methods for transmission mode adaptation in New Radio (NR) with AI/ML assistance. A base station includes a transceiver configured to receive a set of input metrics. The set of input metrics comprises at least one metric derived from a channel state information (CSI) report. The base station further includes a processor operably coupled to the transceiver, the processor configured to determine, based on the set of input metrics, a first multiple-input multiple-output (MIMO) mode throughput prediction and a second MIMO mode throughput prediction, generate, based on the first MIMO mode throughput prediction, a predicted first MIMO mode throughput result, generate, based on the second MIMO mode throughput prediction, a predicted second MIMO mode throughput result, and select a MIMO mode based on the predicted first MIMO mode throughput result and the predicted second MIMO mode throughput result.

Abstract: A method includes identifying ACF information by: obtaining channel information including multiple channels of expected operation scenarios; and based on the channel information for each of the channels, determining MMSE channel estimation (CE) weights expressed in a form of ACFs and an SNR, and covariance matrices. The method includes clustering the MMSE CE weights into K clusters. A center ACF weight of each of the K clusters represents a codeword. The method includes determining a distance metric based on a cluster distance after a re-clustering. The method includes, in response to a determination that cluster distances before and after the clustering differ from each other by a non-negligibly, iteratively re-clustering the ACF information thereby updating the center ACF weights and cluster distances. The method includes generating the codebook to include an index k of each of the K clusters and the center ACF weight of each of the K clusters.

Abstract: Methods and apparatuses for throughput prediction. A base station includes a transceiver configured to receive a set of input metrics and a processor operably coupled to the transceiver. The processor is configured to calculate, based on the set of input metrics, a transmit antenna selection (TAS) throughput prediction and further configured to generate, based on the TAS throughput prediction, a predicted TAS throughput result.

Abstract: Methods and apparatuses for mobility classification and channel state information (CSI) prediction. A method for operating a base station includes receiving, from a user equipment (UE), a channel status state information (CSI) report comprising a precoding matrix indicator (PMI), a channel quality information (CQI), and a rank indicator (RI). The method further includes identifying a CSI configuration of the UE and performing a metric smoothing operation on the PMI. The metric smoothing operation results in a smoothed PMI. The metric smoothing operation comprises a scaling function based on the RI and a discrete Fourier transform (DFT) vector length. The method further includes determining a mobility range classification of the UE based on the smoothed PMI, other metrics in the CSI feedback report, and the CSI configuration.

Abstract: A method and device for self-tuning scales of variables for processing in fixed-point hardware. The device includes a sequence of fixed-point arithmetic circuits configured to receive at least one input signal and output at least one output signal. The circuits are preconfigured with control scales associated with each of the input and output signals. A first circuit in the sequence is configured to receive a first input signal having a dynamic true scale that is different from the control scale associated with the first input signal. Each of the circuits is further configured to determine, for each of the output signals, an adaptive scale from the control scale associated with the output signal based on the true scale of the first input signal and the control scale associated with the first input signal, and generate, from the input signal, the output signal having the associated adaptive scale.

Abstract: A method includes determining estimated features comprising second order statistics based on at least one received signal. The method also includes classifying, using a machine learning network, each channel of the at least one received signal into a channel profile based on the estimated features. The method also includes obtaining multiple minimum mean square error (MMSE) channel estimation weights from a database based on the estimated features, the database storing (i) representative MMSE estimation weights and (ii) channel cluster representatives indexed by the estimated features. The method also includes applying a respective MMSE channel estimation weight for each channel.

Abstract: A system for using radio frequency (RF) communication signals to extract situational awareness information thorough deep learning. Pre-processing is performed to maximally preserve discriminating features in spatial, temporal and frequency domains. A specially designed neural network architecture is used for handling complex RF signals and extracting spatial, temporal and frequency domain information. Data collection and training is used so that the learning system desensitizes from features orthogonal to the underlying classification problem.

Abstract: A method includes storing multiple signals received from a user equipment (UE) in a queue. The method also includes estimating a sounding reference signal (SRS) signal-to-noise-ratio (SNR) and determining a filtered SNR based on the received signals. The method also includes computing one or more features based on the filtered SNR and at least some of the received signals in the queue. The method also includes determining (i) a channel condition of the UE and (ii) a speed range of the UE based on the one or more computed features, wherein the channel condition of the UE comprises line-of-sight (LoS) or non-line-of-sight (NLoS). The method also includes determining a transmission configuration based on the channel condition of the UE and the speed range of the UE.

Abstract: A method for operating a base station comprises receiving channel information from a plurality of UEs; determining, based on the channel information, one or more UEs on which to base a channel prediction; computing a first set of metrics and a second set of metrics corresponding to the plurality of UEs, wherein computing the first set of metrics has a lower complexity than computing the second set of metrics; performing a selection process on the plurality of UEs based on the first and second set of metrics associated with the plurality of UEs; selecting a first subset of UEs from the plurality of UEs based on a first set of metrics; selecting, from the first subset, a second subset of UEs based on a second set of metrics; and performing the channel prediction based on the second subset of UEs.

Abstract: Methods and apparatuses for a channel estimation and prediction operation in a wireless communication systems. A method of a BS comprises: receiving an SRS; partitioning, based on a partition policy, a frequency band of the SRS into sub-bandwidths in a frequency domain; generating, based on previously stored CSI in memory and the partition policy, a set of chunks corresponding to respective sub-bandwidths; performing CHPD operations corresponding to the respective sub-bandwidths to generate channel parameters, wherein different CHPD operations are applied to the respective sub-bandwidths; combining the channel parameters predicted from the respective sub-bandwidths in the frequency domain; and performing, based on the combined channel parameters, a channel estimation and prediction operation.

Abstract: Methods and apparatuses for a BS in a communication system. The method comprises: identifying antenna groups; identifying channel coefficients for each of the antenna groups to perform a channel tracking and prediction operation; receiving, from a user equipment (UE), an uplink signal to perform the channel tracking and prediction operation; and performing, based at least in part on the received uplink signal, a channel coefficient tracking operation for the channel coefficients of the antenna groups, respectively, the channel coefficient tracking operation including a channel subspace parameter tracking operation and a subspace coefficient tracking operation.

Abstract: Simultaneous 2-D room shape reconstruction and self-localization is accomplished using no pre-established infrastructure. A mobile device with co-located microphone and loudspeaker is used to collect echoes reflected by the walls. The system uniquely recovers arbitrary 2-D convex room shape as well as the position of mobile device 10 by collecting and processing distances between three consecutive measurement points as well as acoustic echoes from the device. A practical algorithm for room shape reconstruction and self-localization in the presence of noise and higher order echoes is proposed. Experimental results are provided to demonstrate the effectiveness of the approach.

Abstract: A method for operating a base station comprises receiving channel information from a plurality of UEs; determining, based on the channel information, one or more UEs on which to base a channel prediction; computing a first set of metrics and a second set of metrics corresponding to the plurality of UEs, wherein computing the first set of metrics has a lower complexity than computing the second set of metrics; performing a selection process on the plurality of UEs based on the first and second set of metrics associated with the plurality of UEs; selecting a first subset of UEs from the plurality of UEs based on a first set of metrics; selecting, from the first subset, a second subset of UEs based on a second set of metrics; and performing the channel prediction based on the second subset of UEs.

Abstract: A system for using radio frequency (RF) communication signals to extract situational awareness information thorough deep learning. Pre-processing is performed to maximally preserve discriminating features in spatial, temporal and frequency domains. A specially designed neural network architecture is used for handling complex RF signals and extracting spatial, temporal and frequency domain information. Data collection and training is used so that the learning system desensitizes from features orthogonal to the underlying classification problem.

Abstract: Simultaneous 2-D room shape reconstruction and self-localization is accomplished using no pre-established infrastructure. A mobile device with co-located microphone and loudspeaker is used to collect echoes reflected by the walls. The system uniquely recovers arbitrary 2-D convex room shape as well as the position of mobile device 10 by collecting and processing distances between three consecutive measurement points as well as acoustic echoes from the device. A practical algorithm for room shape reconstruction and self-localization in the presence of noise and higher order echoes is proposed. Experimental results are provided to demonstrate the effectiveness of the approach.