Abstract: In an acoustic apparatus, an acoustic transducer is arranged in a substrate. Multiple acoustic pathways in the substrate have predetermined lengths, wherein a proximal end of each pathway forms an opening in a front surface of the substrate, and a distal end terminates at the acoustic transducer. The predetermined lengths of the acoustic pathways are designed to form an acoustic spatial filter that selectively passes acoustic signals from or to different locations. The transducer can convert electric energy to acoustic energy when the apparatus operates as a speaker, or the the transducer can convert acoustic energy to electric energy and operate as a microphone.

Abstract: A pairwise distance computation transforms first and second signals using an absolute distance preserving mapping, such that a k-norm distance between the first mapped signal and the second mapped signal represents an absolute distance between the first signal and the second signal. The absolute distance preserving mapping maps an element of a first or a second signal to a vector having a size equal to a cardinality of the finite alphabet of the signals. The absolute distance preserving mapping determines a position N of the element in an ordered sequence of symbols of the finite alphabet and determines values for each of N elements of the vector as a fractional power 1/k of positive increments in the finite alphabet. The values for subsequent elements of the vector are determined as zero.

Abstract: A method reconstructs and an uncompressed signal by first obtaining an encoded signal corresponding to the uncompressed signal, wherein the encoded signal includes universally quantized dithered linear measurements of the signal, and wherein each universally quantized dithered linear measurement is a quantized dithered linear measurement of the signal missing one or more significant bits. Side information about the signal is obtained, and the side information is used to obtain a prediction of the signal. The missing one or more significant bits are determined from the encoded signal using the prediction of the signal and the missing one or more significant bits are combined with the encoded signal to produce quantized dithered linear measurements of the signal. Then, the signal can be reconstructed as a reconstructed signal using the quantized dithered linear measurements.

Abstract: A method determines values of the airflow measured in the conditioned environment during the operation of the air-conditioning system and selects, from a set of regimes predetermined for the conditioned environment, a regime of the airflow matching the measured values of the airflow. The method selects, from a set of models of the airflow predetermined for the conditioned environment, a model of airflow corresponding to the selected regime and models the airflow using the selected model. The operation of the air-conditioning system is controlled using the modeled airflow.

Abstract: A system for determining a depth image representing distances to points of a scene includes a light source for generating and directing a set of light pulses toward the scene and a set of sensors for detecting and integrating light reflected from the scene. Each sensor is associated with a point in the scene and generates a signal indicative of an amount of light reflected from the point in the scene and reaching the sensor during a period of time. The system includes at least one processor for coordinating operations of the light source and the set of sensors and for determining the depth image having intensity values representing the distances to the points in the scene. The processor commands the light source to generate the set of light pulses including a first pulse with a first intensity profile having portions of identical lengths with different integrated intensities.

Abstract: Signals received by an array of sensing elements are processed by first positioning the sensing elements in a uniform grid of L locations, wherein each location to include or not to include a sensing element is selected during a design phase. The sensing elements are selected and grouped into subsets, wherein each subset contains one or more sensing elements, and each sensing element is a member of one or more subsets. The signals in each subset are linearly combined to produce a combined signal, which is then sampled to form an output channel, which can detect objects.

Abstract: In a system and method for reconstructing a scene, a light source transmits an optical pulse unto the scene, and a set of sensors receive a reflected pulse to acquire sensed signals corresponding to the optical pulse. There is a fixed coded aperture in an optical path between the light source and set of sensors. Then, a processor performs a method to reconstruct the scene as a three-dimensional (3D) depth map using a scene model.

Abstract: A method for generating an embedding for a signal, which is invariant to an operator, by first measuring the signal with each vector from a set of measurement vectors to produce a corresponding set of coefficients, wherein the measuring compares and transforms each measurement vector and the signal using the operator. A set of one or more matching coefficients according to predefined criteria are selected. Then, the matching coefficients are combined to produce the embedding for the signal, wherein the embedding is operator invariant.

Abstract: Source signals emitted in a reverberant environment from different locations are processed by first receiving input signals corresponding to the source signals by a set of sensors. Then, a sparsity-based support estimation is applied to the input signals according to a reverberation model to produce estimates of the source signals and locations of a set of sources emitting the source signals.

Abstract: Reflectors in a scene are reconstructed using reflected signals. First, signals are transmitted to the scene, and the reflected signals are received by receivers are arranged in an array. The reflected signals are then processed to reconstruct the reflectors in the scene, wherein the processing enforces a model that a reflectivity of the scene in front and behind any reflector is equal to zero.

Abstract: A scene is reconstructed by transmitting pulses into a scene from an array of transmitters so that only one pulse is transmitted by one transmitter at any one time. The one pulse is reflection by the scene and received as a set of signals. Each signal is sampled and decomposed to produce frequency coefficients stacked in a set of linear systems modeling a reflectivity of the scene. Then, a reconstruction method is applied to the set of linear systems. The reconstruction method solves each linear system separately to obtain corresponding solutions, which are shared and combined to reconstruct the scene.

Abstract: A method plans paths of a set of mobile sensors with changeable positions and orientations in an environment. Each sensor includes a processor, an imaging system and a communication system. A desired resolution of coverage of the environment is defined, and an achieved resolution of the coverage is initialized. For each time instant and each sensor, an image of the environment is acquired using the imaging system. The achieved resolution is updated according to the image. The sensor is moved to a next position and orientation based on the achieved resolution and the desired resolution. Then, local information of the sensor is distributed to other sensors using the communication system to optimize a coverage of the environment.

Abstract: A pairwise distance computation transforms first and second signals using an absolute distance preserving mapping, such that a k-norm distance between the first mapped signal and the second mapped signal represents an absolute distance between the first signal and the second signal. The absolute distance preserving mapping maps an element of a first or a second signal to a vector having a size equal to a cardinality of the finite alphabet of the signals. The absolute distance preserving mapping determines a position N of the element in an ordered sequence of symbols of the finite alphabet and determines values for each of N elements of the vector as a fractional power 1/k of positive increments in the finite alphabet. The values for subsequent elements of the vector are determined as zero.

Abstract: A method plans paths of a set of mobile sensors with changeable positions and orientations in an environment. Each sensor includes a processor, an imaging system and a communication system. A desired resolution of coverage of the environment is defined, and an achieved resolution of the coverage is initialized. For each time instant and each sensor, an image of the environment is acquired using the imaging system. The achieved resolution is updated according to the image. The sensor is moved to a next position and orientation based on the achieved resolution and the desired resolution. Then, local information of the sensor is distributed to other sensors using the communication system to optimize a coverage of the environment.

Abstract: A synthetic aperture radar image is generated by directing randomly a beam of transmitted pulses at an area using a steerable array of antennas, wherein the area is uniformly by the transmitted pulses while the array of antennas moves along a path. A sparse reconstruction procedure is applied to received signals from the area due to reflecting the transmitted pulses to generate the image corresponding to the area. The radar system can operate in either sliding spotlight mode, or scan mode. The area can be of an arbitrary shape, and a resolution of the image can be increased.

Abstract: A method extracts a low-rank descriptor of a video acquired of a scene by first extracting a set of descriptors for each image in the video. The sets of descriptors for the video are aggregated to form a descriptor matrix. Iteratively, a low-rank descriptor matrix is determined from the descriptor matrix, as well as a selection matrix that associates each column in the descriptor matrix to a corresponding column in the low-rank descriptor matrix. The low-rank descriptor matrix is output upon convergence.

Abstract: A method generates a 3D synthetic aperture radar (SAR) image of an area by first acquiring multiple data sets from the area using one or more SAR systems, wherein each SAR system has one or more parallel baselines and multiple pulse repetition frequency (PRF), wherein the PRF for each baseline is different. The data sets are registered and aligned to produce aligned data sets. Then, a 3D compressive sensing reconstruction procedure is applied to the aligned data sets to generate the 3D image corresponding to the area.

Abstract: Scale-invariant features are extracted from an image. The features are projected to a lower dimensional random projection matrix by multiplying the features by a matrix of random entries. The matrix of random projections is quantized to produce a matrix of quantization indices, which form a query vector for searching a database of images to retrieve metadata related to the image.

Abstract: A synthetic aperture radar image is generated by directing randomly a beam of transmitted pulses at an area using a steerable array of antennas, wherein the area is uniformly by the transmitted pulses while the array of antennas moves along a path. A sparse reconstruction procedure is applied to received signals from the area due to reflecting the transmitted pulses to generate the image corresponding to the area. The radar system can operate in either sliding spotlight mode, or scan mode. The area can be of an arbitrary shape, and a resolution of the image can be increased.

Abstract: A hash of signal is determining by dithering and scaling random projections of the signal. Then, the dithered and scaled random projections are quantized using a non-monotonic scalar quantizer to form the hash, and a privacy of the signal is preserved as long as parameters of the scaling, dithering and projections are only known by the determining and quantizing steps.