Abstract: Systems, methods, computer program products, and devices provide for computing an eigensystem from a first data set; computing updated eigenvalues that approximate an eigensystem of at least a second data set based on the eigensystem of the first data set; and evaluating a plurality of features in each of the first and at least second data sets using a cost function. The cost function can comprise fewer than the total number of eigenvalues and can include a condition number. The cost function can comprise a coarse approximation of the eigenvalues to de-select at least one of the data sets. This can be useful for learning and/or online processing in an artificial neural network.

Abstract: Enhanced discrete Fourier transform spread Orthogonal Frequency Division Multiplexing (DFT-s-OFDM) selects spreading code roll-off factors based on available spectrum resources in a wireless network and power efficiency needs of at least one wireless device. Excess spectral resources can be used to increase processing gain and reduce peak-to-average power ratio (PAPR), both of which improve a wireless device's power efficiency. Each layer can employ a portion of the DFT-s-OFDM code space in an OFDM symbol for orthogonal multiple access or non-orthogonal multiple access, allowing multiple layers to share the same OFDM symbol. Both uplink and downlink transmissions can benefit from frequency diversity and low PAPR due to DFT-s-OFDM spreading. Since DFT-s-OFDM codes are cyclic shifts of each other, a DFT-s-OFDM discrete-time signal can be synthesized from cyclic shifts of a kernel discrete-time waveform.

Abstract: An unmanned aerial vehicle (UAV) uses a first baseband processor to establish a first communication link with a ground network cell and a second baseband processor that establishes a second communication link with a user device. The second baseband processor is communicatively coupled to the first baseband processor such that the user device exchanges communication data with the core network via the first communication link and the second communication link. Flight-control hardware steers the UAV along a flight trajectory that is determined by a ground-based UAV controller based at least on a geolocation of the user device. The second baseband processor establishes the second communication link with the user device while the first baseband processor maintains the first communication link with the ground network cell.

Abstract: Systems, methods, and apparatuses for analyzing and synthesizing wireless communication signals are provided. A receiver might transform a received signal into a basis in which the transformed signal is sparse, which can reduce the complexity of joint detection by facilitating message passing algorithm (MPA) decoding. A transmitter might employ dense codewords in a first basis, which may facilitate certain signal-processing operations and may provide a transmission with a low peak-to-average-power ratio. The codewords can be designed to be sparse when transformed to a second basis. The codewords may be configured for non-orthogonal multiple access.

Abstract: An artificial neural network (ANN) generates a base expanded matrix that represents an output of a layer of the ANN, such as the output layer. Values in each row are grouped with respect to a set of network parameters in a previous layer, and a sum of the values in each row produces an output vector of activations. The ANN updates the values in at least one column of the expanded matrix according to parameter updates, which results in an updated expanded matrix or an update expanded matrix. An error or a total cost can be computed from the updated expanded matrix or the update expanded matrix. Nonlinear activation functions can be modeled as piecewise linear functions, and a change in an activation function's slope can be modeled as a linear update to an expanded matrix. Parameter updates can be constrained to a restricted value set in order to simplify update operations performed on the expanded matrices.

Abstract: A cooperative multi-user multiple input, multiple output (MIMO) antenna array comprises a MIMO subspace processing system communicatively coupled by a first network to a first set of antennas residing on multiple geographically distributed wireless terminals in a mobile radio network. The first set of antennas is updated to produce a second set of antennas, and the first network is reconfigured to connect the MIMO subspace processing system to the second set of antennas. MIMO subspace processing is reconfigured to employ channel state information for the second set of antennas to generate a plurality of non-interfering subspace channels occupying a common frequency in the mobile radio network.

Abstract: Systems, methods, and apparatuses for analyzing a wireless communication signal are provided. A set of linear operations is performed on a received signal vector, which comprises values of a transmitted signal received by a receiver. The set of linear operations is configured to produce an expanded matrix having multiple rows and multiple columns. The column values in each row of the base expanded matrix are summed to produce a processed signal vector.

Abstract: A wireless client device in a wireless network listens to transmissions from other wireless client devices to determine a local wireless network topology of wireless client devices, and transmits the local wireless network topology to a server in a content delivery network. The wireless client device transmits a request for content to the server and receives a message from the server via the wireless network, the message indicating at least one source wireless client device in the local wireless network topology that contains the content. The wireless client device can transmit a request for the content to the at least one source wireless client device. The wireless client device receives the content from the at least one source wireless client device via a peer-to-peer link in the wireless network.

Abstract: In a multi-user multiple antenna system, a central processor is communicatively coupled to a plurality of geographically distributed access points via a network. The central processor selects two or more of the distributed access points to serve each of a plurality of user devices based on signal power of wireless links between each user device and the distributed access points. The central processor performs subspace processing on the signals transmitted and/or received across the plurality of distributed access points for producing a plurality of non-interfering spatial subchannels.

Abstract: In a multi-user multiple antenna system, a central processor is communicatively coupled to a plurality of geographically distributed access points via a network. The central processor, the geographically distributed access points, or a plurality of client devices served by the system computes channel estimates of wireless channels between the geographically distributed access points and the client devices. The central processor computes access-point weights from the channel estimates to synthesize an antenna array from the plurality of geographically distributed access points, and the access-point weights are used to implement spatial multiplexing.

Abstract: Systems, methods, computer program products, and devices reduce computational processing performed by at least one computer processor that computes an eigensystem from a first data set; computes updated eigenvalues that approximate an eigensystem of at least a second data set based on the eigensystem of the first data set; and evaluates a plurality of features in each of the first and at least second data sets using a cost function; wherein reducing the computational processing of the at least one computer processor is achieved by at least one of selecting the cost function to comprise fewer than the total number of eigenvalues and employing a coarse approximation of the eigenvalues to de-select at least one of the data sets. This is especially useful for learning and/or online processing in an artificial neural network.

Abstract: Applications of CI processing to ad-hoc and peer-to-peer networking significantly improve throughput, network capacity, range, power efficiency, and spectral efficiency. CI-based subscriber units perform network-control functions to optimize network performance relative to channel conditions, network loads, and subscriber services. CI codes are used to identify and address network transmissions. Channel characteristics of communication links are employed to encode, address, and authenticate network transmissions. CI transceivers used as relays and routers employ unique characteristics of transmission paths to code and decode network transmissions. A central processor is adapted to perform array processing with signals received from, and transmitted by, a plurality of subscriber units in a wireless network.

Abstract: Systems, methods, and apparatuses for synthesizing a wireless communication signal are provided. A base expanded matrix is generated, wherein a sum of values in each row of the base expanded matrix produces a base signal vector. Values in at least one column of the base expanded matrix are updated to produce an updated expanded matrix. The values in each row of the updated expanded matrix are summed to produce an updated signal vector. The updated signal vector is transmitted over a wireless channel.

Abstract: A wireless client device in a wireless network listens to transmissions from other wireless client devices to determine a local wireless network topology of wireless client devices, and transmits the local wireless network topology to a server in a content delivery network. The wireless client device transmits a request for content to the server and receives a message from the server via the wireless network, the message indicating at least one source wireless client device in the local wireless network topology that contains the content. The wireless client device can transmit a request for the content to the at least one source wireless client device. The wireless client device receives the content from the at least one source wireless client device via a peer-to-peer link in the wireless network.

Abstract: Enhanced discrete Fourier transform spread Orthogonal Frequency Division Multiplexing (DFT-s-OFDM) selects spreading code roll-off factors based on available spectrum resources in a wireless network and power efficiency needs of at least one wireless device. Excess spectral resources can be used to increase processing gain and reduce peak-to-average power ratio (PAPR), both of which improve a wireless device's power efficiency. Each layer can employ a portion of the DFT-s-OFDM code space in an OFDM symbol for orthogonal multiple access or non-orthogonal multiple access, allowing multiple layers to share the same OFDM symbol. Both uplink and downlink transmissions can benefit from frequency diversity and low PAPR due to DFT-s-OFDM spreading. Since DFT-s-OFDM codes are cyclic shifts of each other, a DFT-s-OFDM discrete-time signal can be synthesized from cyclic shifts of a kernel discrete-time waveform.

Abstract: Low-complexity computational processing provides a set of selective mapping weights for reducing peak-to-average-power ratio (PAPR) in transmitted Multiple Input, Multiple Output Orthogonal Frequency Division Multiplexing (MIMO-OFDM) signals. A MIMO precoder and invertible transform generates a base discrete-time MIMO-OFDM signal from a set of data symbols and MIMO precoding weights. A matrix multiplier multiplies the set of data symbols with a sparse update weight matrix, and the resulting product is MIMO-precoded and modulated onto an OFDM signal to produce a discrete-time update signal. A linear combiner sums the discrete-time update signal and the base discrete-time MIMO-OFDM signal to produce an updated discrete-time MIMO-OFDM signal from which the PAPR can be measured.

Abstract: Applications of CI processing to ad-hoc and peer-to-peer networking significantly improve throughput, network capacity, range, power efficiency, and spectral efficiency. CI-based subscriber units perform network-control functions to optimize network performance relative to channel conditions, network loads, and subscriber services. CI codes are used to identify and address network transmissions. Channel characteristics of communication links are employed to encode, address, and authenticate network transmissions. CI transceivers used as relays and routers employ unique characteristics of transmission paths to code and decode network transmissions. A central processor is adapted to perform array processing with signals received from, and transmitted by, a plurality of subscriber units in a wireless network.

Abstract: An OFDM transmitter spreads original data symbols with a complex-valued spreading matrix derived from a discrete Fourier transform. Spread data symbols are mapped to OFDM subcarriers. Spreading and mapping are configured to produce a transmitted spread-OFDM signal with a low peak-to-average power ratio (PAPR) and orthogonal code spaces. In MIMO systems, the complex-valued spreading matrix can comprise a MIMO precoding matrix, and the code spaces can comprise MIMO subspaces. In Cooperative-MIMO, a combination of low code-space cross correlation and low PAPR can be achieved.

Abstract: An Orthogonal Frequency Division Multiplexing (OFDM) transmitter generates OFDM multiple-access signals with low Peak-to-Average-Power Ratio (PAPR). A code-division multiplexer arranges original data symbols from different data streams inside each length-N symbol block, which is spread by a Discrete Fourier Transform (DFT) spreader. The arrangement of the original data symbols configures the DFT spreader to spread each original data symbol into a predetermined spread-DFT code division multiple access channel. The resulting DFT-spread data symbols are mapped to OFDM subcarriers, and an inverse discrete Fourier transform (IDFT) operates on the DFT-spread data symbols to generate an OFDM transmission signal having low PAPR.

Abstract: Systems, methods, computer program products, and devices reduce computational processing performed by at least one computer processor that computes an eigensystem from a first data set; computes updated eigenvalues that approximate an eigensystem of at least a second data set based on the eigensystem of the first data set; and evaluates a plurality of features in each of the first and at least second data sets using a cost function; wherein reducing the computational processing of the at least one computer processor is achieved by at least one of selecting the cost function to comprise fewer than the total number of eigenvalues and employing a coarse approximation of the eigenvalues to de-select at least one of the data sets. This is especially useful for learning and/or online processing in an artificial neural network.