Method, apparatus and system for performing rapid initialization of a receiver

Abstract

A method, apparatus and system for performing rapid initialization of a receiver include applying an estimate of receiver position, time and radio signals received by the signal receiver to a neural network that is trained with signal environment signatures to generate a plurality of receiver direction of motion hypotheses, generating a plurality of phasor sequences each forming a phasor hypothesis comprising a signal phase representing one of the direction of motion hypotheses, correlating a local signal with at least one received signal to produce at least one correlation result, compensating a phase of at least one of the local signal, a received signal or at least one correlation result based on the plurality of phasor hypotheses, determining a preferred hypothesis in the plurality of hypotheses that optimizes the at least one correlation result, and improving reception of the receiver using the phase compensated correlation result optimized by the preferred hypothesis.

Description

METHOD, APPARATUS AND SYSTEM FOR PERFORMING RAPID INITIALIZATION OF A RECEIVERBACKGROUNDField

[0001] Embodiments of the present principles generally relate to radio signal receivers and, in particular, to a method, apparatus and system for performing rapid initialization of a radio signal receiver.Description of the Related Art

[0002] Positioning signal receivers such as receivers for global satellite navigation systems (GNSS) signals have become ubiquitous in mobile devices. A GNSS receiver (e.g., receivers for GPS, GLONASS, GALILEO, BEIDOU, etc. satellite signals or a combination thereof) receive signals from satellites, process the received signals and determine the position of the receiver from information contained in the received signals. The typical accuracy of a consumer receiver without the assistance of an inertial measurement unit (IMU) can range from 5m to 50m. To provide inertial navigation in a typical mobile device, an IMU typically comprises a magnetometer, a gyroscope and an accelerometer (i.e., traditional IMU sensors). The signals from these three sensors (typically, MEMS-based sensors) are used to augment the GNSS receiver’s positioning computation such that the receiver accuracy can be improved to about 20cm. However, the additional accuracy comes with a substantial cost of the IMU sensors and additional computational complexity.

[0003] In other forms of radio receivers, it can be advantageous for signal reception to know the receiver’s motion such that the received signals can be motion compensated by removing the frequency and / or phase error caused by receiver motion from the signal correlation process. Compensating for the receiver motion improves receiver signal reception and / or lowers the cost of receiver components.

[0004] In some instances, receivers can use a signal processing technique known as SUPERCORRELATION™ for improving radio signal reception using receiver motion compensated signal processing and is described in commonly assigned US patent 9,780,829, issued 3 October 2017; US patent 10,321 ,430, issued 11 June 2019; US patent 10,816,672, issued 27 October 2020; US patent publication2020 / 0264317, published 20 August 2020; and US patent publication 2020 / 0319347, published 8 October 2020, which are hereby incorporated herein by reference in their entireties. In such instances, when a receiver has a cold start or has the view of the sky blocked (e.g., inside a building) for a period of time, the initialization process can take time while the SUPERCORRELATION™ technique tests various position and motion hypotheses before determining the actual receiver direction of motion.

[0005] Therefore, there is a need for a method, apparatus and system for performing rapid initialization in a radio signal receiver that uses SUPERCORRELATION™ techniques to enhance signal processing.SUMMARY

[0006] Embodiments of the present invention generally relate to a method, apparatus, and system for performing rapid initialization of a radio signal receiver as shown in and / or described in connection with at least one of the figures.

[0007] Various advantages, aspects and features of the present disclosure, as well as details of an illustrated embodiment thereof, are more fully understood from the following description and drawings.BRIEF DESCRIPTION OF THE DRAWINGS

[0008] So that the manner in which the features of the present invention can be understood in detail, a particular description of the invention, may be had by reference to embodiments, some of which are illustrated in the appended drawings. It is to be noted, however, that the appended drawings illustrate only typical embodiments of this invention and are therefore not to be considered limiting of its scope, for the invention may admit to other equally effective embodiments.

[0009] FIG. 1 depicts a communication environment in which an embodiment of a rapid initialization process can be implemented to improve initialization of a radio signal receiver in accordance with at least one embodiment of the present principles;

[0010] FIG. 2 depicts a functional block diagram of the radio signal receiver 104 of FIG. 1 in accordance with at least one embodiment of the present principles;

[0011] FIG. 3 depicts a block diagram of a computing device programmed to function as a rapid initialization module of the receiver of FIG. 2 in accordance with at least one embodiment of the invention;

[0012] FIG. 4 depicts a flow diagram of a method of training the neural network 226 in accordance with at least one embodiment of the present principles; and

[0013] FIG. 5 depicts a flow diagram of a method of operation of a receiver in accordance with at least one embodiment of the present principles.

[0014] To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures. The figures are not drawn to scale and may be simplified for clarity. It is contemplated that elements and features of one embodiment may be beneficially incorporated in other embodiments without further recitation.DETAILED DESCRIPTION

[0015] Embodiments of the present principles generally relate to an apparatus, method, and system for performing rapid initialization of radio signal receivers that use motion compensated signal processing. Such receivers can include positioning systems (e.g., a GNSS receiver) and / or communications receivers (e.g., WiFi, cellular, Bluetooth communications receivers). While the concepts of the present principles are susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and are described in detail below. It should be understood that there is no intent to limit the concepts of the present principles to the particular forms disclosed. On the contrary, the intent is to cover all modifications, equivalents, and alternatives consistent with the present principles and the appended claims. For example, although embodiments of the present principles will be described primarily with respect to specific receivers and received signals from specific sources, embodiments of the present principles can be applied to substantially any receiver receiving signals from substantially any source.

[0016] In one exemplary embodiment of the present principles, a neural network (or other machine learning technique) is trained with a catalog of example environments (e.g., street layouts and building models) and GNSS signals thattogether form an image or signature of the signal environment for a given position and time. The neural network learns how to infer the likely street layout (an inference output) using only satellite visibility information (inference input) (e.g., satellite locations in the sky and signal strength received from each satellite).

[0017] Applying the current signal environment (e.g., provided positions and signal strengths of all the satellites that are currently above the horizon) to the neural network enables a quick determination of the probable road layout and hence set of potential directions of travel of the receiver. The receiver only needs to know its approximate position within a few kilometers and time to within a few minutes in order to correctly place each satellite in the sky (correct azimuth and elevation being sent into the neural network). In essence, the neural network is performing pattern matching between its learned signal environment patterns and the current pattern being seen by the receiver.

[0018] In some embodiments, a radio signal receiver of the present principles uses the SUPERCORRELATION™ technique to test hypotheses of receiver motion that are constrained near the direction of motion estimate from the neural network. In such embodiments, constraining the number of hypotheses used in the SUPERCORRELATION™ technique, expedites the initialization process of a receiver. That is, in some embodiments, there can be four main hypotheses to test in the SUPERCORRELATION™ technique, for example, two in a first direction and two in a second direction, orthogonal to the first direction, such as two directions along a road and two directions across a road. Consequently, rather than testing many hypotheses regarding the motion of the receiver, the receiver only needs to test 4 hypotheses (i.e., two along the road and two across the road). As such, the receiver quickly determines its direction of travel and can quickly begin using SUPERCORRELATION™.

[0019] More specifically, satellite-based positioning systems and communications receivers utilize encoded digital signals including a deterministic digital code to facilitate signal acquisition, e.g., Gold codes. Such a digital code is determined by the receiver and repeatedly broadcast by the transmitter to enable receivers to acquire and process transmitted signals. Using such deterministic codes combined with anaccurate motion model of the receiver, embodiments of the present principles can enable a receiver to improve its position computation accuracy and / or signal reception without using traditional IMU data from a full complement of IMU sensors. The technique for improving radio signal reception using receiver motion compensated signal processing is known as SUPERCORRELATION™ and is described in commonly assigned US patent 9,780,829, issued 3 October 2017; US patent 10,321 ,430, issued 11 June 2019; US patent 10,816,672, issued 27 October 2020; US patent publication 2020 / 0264317, published 20 August 2020; and US patent publication 2020 / 0319347, published 8 October 2020, which are hereby incorporated herein by reference in their entireties. Previously, the motion model was derived using IMU data; however, in embodiments of the present principles, the motion model can be derived using only gyroscope data, controller area network (CAN) bus data, visual odometry data, or a combination of data from any of these motion data sources. In one embodiment, the receiver has no motion information except for a position from a GNSS receiver. Also, motion information determined by the SUPERCORRELATION™ technique can be provided as feedback to a motion module to correct and / or update the motion model.

[0020] In one embodiment, the radio receiver is embedded in a moving platform such as, for example, but not limited to, an automobile, motorcycle, airplane, helicopter, drone, bicycle, person (e.g., a person carrying a smartphone, tablet, computer, internet of things (loT) device, etc.), and the like. In operation, the at least one received signal is correlated with at least one locally generated signal to produce at least one correlation result. As is described in detail below, embodiments of the present principles use a rapid initialization technique to determine the motion information used to motion compensate the correlation results to enable the receiver to improve signal processing.

[0021] FIG. 1 depicts a communication environment 100 in which an embodiment of a rapid initialization process can be implemented to improve initialization of a radio signal receiver in accordance with at least one embodiment of the present principles. In the communication environment 100 of FIG. 1 , a person 102 carrying a receiver 104 exits (arrow 106) a building 108. While within the building 108, the receiver 104 does not have a clear view of the sky and the satellite signals 110 broadcast from aplurality of satellites 112-1 , 112-2, . . . 112-N. Consequently, the receiver 104 may not be able to accurately know its direction of travel when the person 102 leaves the building. An embodiment of the rapid initialization technique of the present principles can be used, as described below, to quickly determine the direction of travel (arrow 114) by reducing the number of travel direction hypotheses to, for example, four (i.e. , two along the road 116 and two across the road 116 (represented by arrows 118-1 , 118-2, 118-3 and 118-4)).

[0022] In the embodiment of FIG. 1 , the receiver 104 can be a component (e.g., a GNSS receiver) within user equipment such as mobile phones, tablets, laptop computers, loT devices, and the like. For simplicity, the device is described herein as a receiver. Those skilled in the art will understand that the receiver can be a standalone receiver or can be a portion or component within user equipment.

[0023] FIG. 2 depicts a functional block diagram of the radio signal receiver 104 of FIG. 1 in accordance with at least one embodiment of the present principles. In the embodiment of FIG. 2, the receiver 104 comprises an antenna 210, a front end 212, a GNSS signal processor 214, motion compensation processor 216, a motion module 218 and a rapid initialization module 224. When carried by a person 102 (or other transport platform such as a vehicle), the receiver 104 and its antenna 210 are an indivisible unit where the antenna 210 moves with the person 102. The SUPERCORRELATION™ technique operates based upon determining a component of motion of the signal receiving antenna that is in the direction of the source of a received signal. Any mention of motion herein refers to the motion of the antenna 210. In most scenarios, the motion of the person 102 is the same as the motion of the antenna 210 and, as such, the following description assumes the motion of the person 102 and antenna 210 are the same.

[0024] In the embodiment of FIG. 2, the receiver’s front end 212 downconverts, filters, and samples (digitizes) the received signals in a manner that is well-known to those skilled in the art. The output of the receiver front end 212 is a digital signal containing data. The data of interest for performing motion compensation is a deterministic code, e.g., Gold code, used by the GNSS signal processor 214 to synchronize the receiver 104 to the GNSS transmission.

[0025] The GNSS signal processor 214 correlates the received code from each satellite with locally generated codes to produce correlation results. The correlation results are processed as is well-known in the art to generate position information 226, e.g., the correlation results are used to determine pseudoranges for each satellite and the pseudoranges are processed to compute the receiver position. The motion compensation processor 216 performs the SUPERCORRELATION™ processing to provide signals (phasor sequences) to phase adjust the correlation results such that the coherent integration period is extended, e.g., extended to one or more seconds. The phasor sequence is a time sequence of phase offsets where each phasor in the sequence adjusts the phase of a signal sample. The adjustment can be performed by adjusting the phase of each sample of the received signals, the locally generated signals or the correlation results themselves. In some embodiments, the least computationally intensive adjustment process adjusts the phase of the correlation results.

[0026] In the embodiment of FIG. 2, the motion module 218 generates receiver motion information that is used by the motion compensation processor 216 to generate phasor sequences that are used to motion compensate the correlation results. The phasor sequences comprise a sequence of phase offsets to be made over time, e.g., across a received signal, to compensate for phase changes that occur over time due to movement of the receiver. In one embodiment, the motion module 218 uses the GNSS signal processor 214 and / or an IMU 222 to generate motion information for the motion compensation processor 216. Motion information can comprise a prediction of receiver velocity and heading. Alternative or in addition, the motion module 218 can comprise an optional IMU 222 to provide vehicle orientation information for the motion model. In the embodiment of FIG. 2, the motion compensation processor 216 provides motion estimation correction information along path 220 to the motion module 218. In this manner, the motion compensation processor 216 can provide corrective feedback to the motion module 218.

[0027] In the embodiment of FIG. 2, the rapid initialization module 224 is coupled to the GNSS processor 214 and the motion compensation processor 216. As is described in detail below, the rapid initialization module relies upon a trained neural network 226 that has been trained with GNSS signals, street maps and 3D buildingmodels (i.e., cityscapes). The input from the GNSS processor 214 includes a general location of the receiver (e.g., within a couple of kilometers of the true position). From this input the neural network 226 determines, for example, four hypotheses of the direction of travel for the receiver 104. These hypotheses are coupled to the motion compensation processor 216 to be used as initial estimates of receiver motion for SUPERCORRELATION™ processing. The motion compensation processor quickly determines which of the four hypotheses represents the correct direction of travel. That is, the correct hypothesis produces the highest correlation result for the GNSS signals.

[0028] FIG. 3 depicts a block diagram of a computing device programmed to function as a rapid initialization module 224 of the receiver 104 of FIGs. 1 and 2 in accordance with at least one embodiment of the invention. In some embodiments, the rapid initialization module 224 or its function can be performed remotely on a server. In such an embodiment, the receiver 104 communicates wirelessly with the server and motion direction hypotheses are remotely computed. In other embodiments, the rapid initialization module 224 is a component within the receiver 104. In the embodiment of FIG. 3, the rapid initialization module 224 comprises at least one processor 300, support circuits 302 and memory 304. The at least one processor 300 can be any form of processor or combination of processors including, but not limited to, central processing units, microprocessors, microcontrollers, field programmable gate arrays, graphics processing units, digital signal processors, and the like. The support circuits 302 can comprise well-known circuits and devices facilitating functionality of the processor(s). The support circuits 302 can further comprise one or more of, or a combination of, power supplies, clock circuits, analog to digital converters, communications circuits, cache, displays, and / or the like. The support circuits 302 form an interface between the processor 300 and both the motion compensation processor 216 and GNSS signal processor 214.

[0029] In the embodiment of FIG. 3, the memory 304 comprises one or more forms of non-transitory computer readable media including one or more of, or any combination of, read-only memory or random-access memory. The memory 304 stores software and data including, for example, neural network software 306 and data 308. In some embodiments, the data 308 comprises a receiver position estimate310, motion hypotheses 312, and various additional data used to perform the rapid initialization processing. The data 308 can include neural network training data 314 including GNSS signals 316, GNSS satellite positions 318, building maps 320, street maps 322, time 224 and the like. The neural network software 316, when executed by the one or more processors 300, becomes a trained neural network 226 of FIG. 2 that is capable of using the receiver position estimate to produce a plurality of hypotheses of receiver direction of motion. The operation of the neural network software 306 functions as the neural network 226 of the rapid initialization module 224 of FIG. 2. Training of the neural network is described with reference to FIG. 4 below and use of the trained neural network is described with reference to FIG. 5 below.

[0030] FIG. 4 depicts a flow diagram of a method 400 of training the neural network 226 in accordance with at least one embodiment of the present principles. In some embodiments, the neural network can be installed into the mobile device (receiver 104) and trained in the field. Alternatively or in addition, the training can be performed on a server and the trained neural network installed onto the mobile device. In other embodiments, the trained neural network can be located on a server and remotely accessed from the mobile device via wireless communications (e.g., WIFI, cellular, etc.).

[0031] The method 400 can begin at 402 and proceed to 404 during which the neural network software is executed to form a neural network. At 406, training data is applied to the neural network. The training data can include, but is not limited to GNSS signals, GNSS satellite positions, building maps, street maps, time of day, and the like. The GNSS signals are typically the signal strengths for satellite transmissions of GNSS satellites that are above the horizon at a particular location and time of day on Earth. Building maps are three-dimensional maps of the cityscape environment at a particular location and street maps are typically two-dimensional maps of roadways proximate a particular location. The neural network can be trained with regional data representing buildings, streets and visible satellite signal information covering a region of Earth (e.g., a particular city) or the data can represent the entire world. The result of training the neural network is a network that contains a GNSS signal environment signature related to any location or time. The network is able to determine a limitednumber of direction of travel hypotheses based upon currently measured satellite signals, a receiver position estimate, and a time of day estimate.

[0032] To generate the GNSS signal signature data, a reference receiver can be used to log visible satellites and their transmission signal strength at various locations and times of day, e.g., every few meters. Such data forms a signal signature for each location and time. Training the neural network with this information as well as street and building maps produces a neural network with the knowledge of the signal environment at any location and time. Such knowledge, enables the neural network to use currently produced signal information that is generated by the receiver to rapidly determine the location and direction of motion hypotheses. In essence, the neural network is performing pattern matching between its learned signal environment patterns and the current pattern being seen by the receiver. These hypotheses are constrained by the street map information, building map information and GNSS signal signature information. As the receiver moves, these constraints enable the neural network to identify a limited number (e.g., four) of possible directions of motion for the receiver - as described in further detail below with reference to FIG. 5.

[0033] Once an initial amount of training data is applied to the neural network, the method 400 queries at 408 whether the training is to continue. If the query is affirmatively answered, the method 400 proceeds to 406 to continue applying additional training data. If the query is negatively answered, the method 400 proceeds to 410 and ends. The neural network is then trained and ready for use.

[0034] FIG. 5 depicts a flow diagram of a method 500 of operation of a receiver in accordance with at least one embodiment of the present principles. The method 500 of FIG. 5 can be implemented in software, hardware or a combination of both (e.g., using the GNSS processor 214, the motion compensation processor 216 and the rapid initialization module 224 of FIG. 2).

[0035] The method 500 can begin at 502 and proceed to 504 during which signals are received at a receiver from at least one remote source (e.g., transmitters such as the GNSS satellites 112 of FIG. 1 ) in a manner as described with respect to FIG. 1. Each received signal comprises a synchronization or acquisition code, e.g., a Gold code, extracted from the radio frequency (RF) signal received at the antenna. Theprocess of downconverting the RF signal and sampling the digital code is well known in the art. The method 500 can proceed to 506.

[0036] At 506, the method 500 queries whether a rapid initialization procedure is needed to process the received signals. If the query is affirmatively answered, the rapid initialization module (at 508, 510, 512, and 514) can define a search space in which to process the received signals. The search space defines a quantity of phasor sequence hypotheses to be tested to adjust the correlation result phase until a preferred hypothesis is found. Upon initialization (e.g., receiver is activated or exits an area where the satellite signals were unavailable), the search space is defined by the rapid initialization module to determine the direction of motion of the receiver. The method 500 can proceed to 508.

[0037] At 508, the receiver position and time estimates are accessed and, at 510, the GNSS signal information (e.g., signal strength) is accessed. The method 500 can proceed to 512.

[0038] At 512, the position estimate and signal information are applied to the trained neural network. The neural network searches for a satellite signal signature that matches the current signal information. The method 500 can proceed to 514. At 514, the neural network outputs a limited number (e.g., typically four) of direction of motion hypotheses to be used in the SUPERCORRELATION™ process. The method 500 can proceed to 516.

[0039] At 516, as the hypotheses are tested in the SUPERCORRELATION™ process and the correlation results converge on an optimal value, the search space can be narrowed for subsequent correlation result processing. For example, a search space is defined having, for example, four phasor sequence hypotheses related to the direction of travel. Once a preferred hypothesis is determined, the next received signal is tested with hypotheses that are near to or identical to the prior preferred hypothesis. That is, at 516, the method 500 performs SUPERCORRELATION™ using the motion hypotheses and the received signals. During this process, the SUPERCORRELATION™ procedure generates a plurality of phasor sequence hypotheses related to the motion information. These hypotheses comprise a plurality of local signals representing code phase estimates. Each phasor sequencehypothesis comprises a phase estimate that varies with motion parameters of the receiver. The signal processing correlates a local code encoded in a local signal with a code encoded in the received RF signal. The phasor sequence hypotheses are used to adjust, at a sub-wavelength accuracy, the carrier phase of the local code. Such adjustment or compensation may be performed by adjusting a local oscillator signal, the received signal(s), or the correlation result. The signals and / or correlation results comprise complex signal samples having in-phase (I) and quadrature phase (Q) components. The method applies each phase offset in the phasor sequence to a corresponding complex sample in the signals and / or correlation results. For each received signal, the process correlates the received signals with a set (plurality) of phasor sequence hypotheses containing estimates of a phase offset necessary to accurately correlate the received signals.

[0040] In embodiments of the present principles, the motion estimates are typically hypotheses of the motion in a direction of interest such as in the direction of the satellite that transmitted the received signal, e.g., along the signal propagation path. At initialization, the direction of interest can be unknown or inaccurately estimated. Consequently, the rapid initialization module is used to identify a limited number of directions of interest. A comparison of correlation results over the direction hypotheses enables the method 500 to narrow the search space when processing subsequently received signals. Consequently, subsequent compensation is performed over a narrow search space.

[0041] In one embodiment, if a signal from a given satellite was received previously, the set of hypotheses for the newly received signal include a group of phasor sequence hypotheses using the expected Doppler and Doppler rate and / or last Doppler and last Doppler rate used in receiving the prior signal from that particular satellite. The hypotheses values can be centered around the last values used or the last values used additionally offset by a prediction of further offset based on the expected receiver motion. The method 500 correlates each received signal with that signal’s set of hypotheses. The hypotheses are used as parameters to form the phase-compensated phasors to phase compensate the correlation process. As such, the phase compensation can be applied to the received signals, the local frequency source (e.g., an oscillator), or the correlation result values. The result of the correlationprocess is a plurality of phase-compensated correlation results - one phase- compensated correlation result value for each hypothesis for each received signal.

[0042] The correlation results are processed to find a “best” or optimal result for each received signal. A joint correlation output is produced as a function (e.g., summation) of the plurality of correlation results resulting from all the hypotheses and received transmitter signals. The joint correlation output can be a single value or a plurality of values that represent the parameter hypotheses (preferred hypotheses) that provide an optimal or best correlation output. In general, a cost function is applied to each set of correlation values for each received signal to find the optimal correlation output corresponding to a preferred hypothesis or hypotheses.

[0043] For example, assuming all other receiver parameters are known except receiver motion direction, the method 500 tests hypotheses with various phasor sequences that compensate for phase changes due to each direction hypothesis. The correct phasor sequence hypothesis that represents the accurate direction estimate will produce the highest correlation result magnitude for a given received signal. By processing the received signals from different satellites, the correlation results will converge upon hypotheses representing the true receiver motion direction. The method 500 can then proceed to 518.

[0044] At 518, the compensated correlation results are used by a positioning solution (e.g., using range to each satellite transmitter) to generate the position of the receiver. The method 500 ends at 520.

[0045] If, at 506, the receiver is not being initialized, the query is negatively answered and the method 500 proceeds to 516 without using the rapid initialization module. Consequently, the SUPERCORRELATION™ procedure is used to motion compensate the correlation results as is known in the art. At 518, the method 500 uses the correlation results to generate the receiver position in a manner that is known in the art. The method 500 ends at 520.

[0046] Alternatively, in some embodiments, rather than using the largest magnitude correlation value, other test criteria can be used. For example, the progression of correlations can be monitored as hypotheses are tested and a costfunction can be applied that indicates the best hypotheses when the cost function reaches a minimum (e.g., a small hamming distance amongst peaks in the correlation plots). As such, the joint correlation output can be a joint correlation value or a group of values.

[0047] In some embodiments, a method for performing rapid initialization of a radio signal receiver includes receiving radio signals from a plurality of transmitters, applying an estimate of receiver position, time and the received radio signals to a neural network that is trained with signal environment signatures, generating, from the neural network, a plurality of receiver direction of motion hypotheses, generating a plurality of phasor sequences, where each phasor sequence forms a phasor hypothesis comprising a signal phase representing one of the direction of motion hypotheses, correlating at least one local signal with at least one received signal to produce at least one correlation result, compensating a phase of the at least one of the local signal, the at least one received signal or the at least one correlation result based on the plurality of phasor hypotheses to generate a plurality of phase- compensated correlation results, determining a preferred hypothesis in the plurality of hypotheses that optimizes the at least one correlation result of at least one of the phase-compensated correlation results, and improving signal reception of the radio receiver using at least the phase compensated correlation result optimized by the preferred hypothesis.

[0048] In some embodiments, the method can further include updating a motion model of the receiver using a direction of motion estimate associated with the preferred hypothesis.

[0049] In some embodiments, the neural network is trained using street layouts, building models and GNSS signals that together form a signature of a signal environment for a given position of a receiver at a given time.

[0050] In some embodiments, the neural network matches a trained signal signature with a current signal signature sensed by the receiver to generate the plurality of receiver direction of motion hypotheses.

[0051] In some embodiments, the plurality of generated receiver direction of motion hypotheses comprise a reduced number of motion hypotheses as compared to receiver motion hypotheses generated without the neural network.

[0052] In some embodiments, the plurality of generated receiver direction of motion hypotheses comprise four receiver direction of motion hypotheses, two in a first direction and two in a second direction, orthogonal to the first direction.

[0053] In some embodiments, an apparatus for performing rapid initialization of a radio signal receiver includes at least one processor and at least one non-transient computer readable medium for storing instructions. When the instructions are executed by the at least one processor, the apparatus performs receiving radio signals from a plurality of transmitters, accessing an estimate of receiver position and time, accessing the received radio signals, applying the estimate of receiver position, time and the received radio signals to a neural network that is trained with a signal environment signatures, generating, from the neural network, a plurality of receiver direction of motion hypotheses, generating a plurality of phasor sequences, where each phasor sequence forms a phasor hypothesis comprising a signal phase representing one of the direction of motion hypotheses, correlating at least one local signal with at least one received signal to produce at least one correlation result, compensating a phase of the at least one of the local signal, the at least one received signal or the at least one correlation result based on the plurality of phasor hypotheses to generate a plurality of phase-compensated correlation results, determining a preferred hypothesis in the plurality of hypotheses that optimizes the at least one correlation result of at least one of the phase-compensated correlation results, and improving signal reception of the radio receiver using at least the phase compensated correlation result optimized by the preferred hypothesis.

[0054] In some embodiments, a system for performing rapid initialization of a radio signal receiver includes at least one source of radio signals and a radio signal receiver comprising an antenna, a neural network, at least one processor, and at least one memory for storing programs and instructions. When the instructions are executed by the at least one processor, the receiver performs receiving radio signals from the at least one source using the antenna, applying an estimate of receiver position, timeand the received radio signals to the neural network that is trained with signal environment signatures, generating, from the neural network, a plurality of receiver direction of motion hypotheses, generating a plurality of phasor sequences, where each phasor sequence forms a phasor hypothesis comprising a signal phase representing one of the direction of motion hypotheses, correlating at least one local signal with at least one received signal to produce at least one correlation result, compensating a phase of the at least one of the local signal, the at least one received signal or the at least one correlation result based on the plurality of phasor hypotheses to generate a plurality of phase-compensated correlation results, determining a preferred hypothesis in the plurality of hypotheses that optimizes the at least one correlation result of at least one of the phase-compensated correlation results, and improving signal reception of the radio receiver using at least the phase compensated correlation result optimized by the preferred hypothesis.

[0055] Those skilled in the art will appreciate that, while various items are illustrated as being stored in memory or on storage while being used, these items or portions of them can be transferred between memory and other storage devices for purposes of memory management and data integrity. Alternatively, in other embodiments some or all of the software components can execute in memory on another device and communicate with the illustrated computer system via intercomputer communication. Some or all of the system components or data structures can also be stored (e.g., as instructions or structured data) on a computer-accessible medium or a portable article to be read by an appropriate drive, various examples of which are described above. In some embodiments, instructions stored on a computer- accessible medium separate from a computing device can be transmitted to the computing device via transmission media or signals such as electrical, electromagnetic, or digital signals, conveyed via a communication medium such as a network and / or a wireless link. Various embodiments can further include receiving, sending or storing instructions and / or data implemented in accordance with the foregoing description upon a computer-accessible medium or via a communication medium. In general, a computer-accessible medium can include a storage medium or memory medium such as magnetic or optical media, e.g., disk or DVD / CD-ROM, volatile or non-volatile media such as RAM (e.g., SDRAM, DDR, RDRAM, SRAM, and the like), ROM, and the like.

[0056] The methods and processes described herein may be implemented in software, hardware, or a combination thereof, in different embodiments. In addition, the order of methods can be changed, and various elements can be added, reordered, combined, omitted or otherwise modified. All examples described herein are presented in a non-limiting manner. Various modifications and changes can be made as would be obvious to a person skilled in the art having benefit of this disclosure. Realizations in accordance with embodiments have been described in the context of particular embodiments. These embodiments are meant to be illustrative and not limiting. Many variations, modifications, additions, and improvements are possible. Accordingly, plural instances can be provided for components described herein as a single instance. Boundaries between various components, operations and data stores are somewhat arbitrary, and particular operations are illustrated in the context of specific illustrative configurations. Other allocations of functionality are envisioned and can fall within the scope of claims that follow. Structures and functionality presented as discrete components in the example configurations can be implemented as a combined structure or component. These and other variations, modifications, additions, and improvements can fall within the scope of embodiments as defined in the claims that follow.

[0057] In the foregoing description, numerous specific details, examples, and scenarios are set forth in order to provide a more thorough understanding of the present disclosure. It will be appreciated, however, that embodiments of the disclosure can be practiced without such specific details. Further, such examples and scenarios are provided for illustration, and are not intended to limit the disclosure in any way. Those of ordinary skill in the art, with the included descriptions, should be able to implement appropriate functionality without undue experimentation.

[0058] References in the specification to “an embodiment,” etc., indicate that the embodiment described can include a particular feature, structure, or characteristic, but every embodiment may not necessarily include the particular feature, structure, or characteristic. Such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is believed to be within the knowledge of one skilledin the art to affect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly indicated.

[0059] Embodiments in accordance with the disclosure can be implemented in hardware, firmware, software, or any combination thereof. Embodiments can also be implemented as instructions stored using one or more machine-readable media, which may be read and executed by one or more processors. A machine-readable medium can include any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computing device or a “virtual machine” running on one or more computing devices). For example, a machine-readable medium can include any suitable form of volatile or non-volatile memory.

[0060] In addition, the various operations, processes, and methods disclosed herein can be embodied in a machine-readable medium and / or a machine accessible medium / storage device compatible with a data processing system (e.g., a computer system), and can be performed in any order (e.g., including using means for achieving the various operations). Accordingly, the specification and drawings are to be regarded in an illustrative rather than a restrictive sense. In some embodiments, the machine-readable medium can be a non-transitory form of machine-readable medium / storage device.

[0061] Modules, data structures, and the like defined herein are defined as such for ease of discussion and are not intended to imply that any specific implementation details are required. For example, any of the described modules and / or data structures can be combined or divided into sub-modules, sub-processes or other units of computer code or data as can be required by a particular design or implementation.

[0062] In the drawings, specific arrangements or orderings of schematic elements can be shown for ease of description. However, the specific ordering or arrangement of such elements is not meant to imply that a particular order or sequence of processing, or separation of processes, is required in all embodiments. In general, schematic elements used to represent instruction blocks or modules can be implemented using any suitable form of machine-readable instruction, and each such instruction can be implemented using any suitable programming language, library, application-programming interface (API), and / or other software development tools orframeworks. Similarly, schematic elements used to represent data or information can be implemented using any suitable electronic arrangement or data structure. Further, some connections, relationships or associations between elements can be simplified or not shown in the drawings so as not to obscure the disclosure.

[0063] While the foregoing is directed to embodiments of the present principles, other and further embodiments of the invention can be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow.

Claims

Claims:

1. A method for performing rapid initialization of a radio signal receiver, comprising: receiving radio signals from a plurality of transmitters; applying an estimate of receiver position, time and the received radio signals to a neural network that is trained with signal environment signatures; generating, from the neural network, a plurality of receiver direction of motion hypotheses; generating a plurality of phasor sequences, where each phasor sequence forms a phasor hypothesis comprising a signal phase representing one of the direction of motion hypotheses; correlating at least one local signal with at least one received signal to produce at least one correlation result; compensating a phase of the at least one of the local signal, the at least one received signal or the at least one correlation result based on the plurality of phasor hypotheses to generate a plurality of phase-compensated correlation results; determining a preferred hypothesis in the plurality of hypotheses that optimizes the at least one correlation result of at least one of the phase-compensated correlation results; and improving signal reception of the radio receiver using at least the phase compensated correlation result optimized by the preferred hypothesis.

2. The method of claim 1 , further comprising updating a motion model of the receiver using a direction of motion estimate associated with the preferred hypothesis.

3. The method of claim 1 , wherein the neural network is trained using street layouts, building models and GNSS signals that together form a signature of a signal environment for a given position of a receiver at a given time.

4. The method of claim 3, wherein the neural network matches a trained signal signature with a current signal signature sensed by the receiver to generate the plurality of receiver direction of motion hypotheses.

5. The method of claim 1 , wherein the plurality of generated receiver direction of motion hypotheses comprise a reduced number of motion hypotheses as compared to receiver motion hypotheses generated without the neural network.

6. The method of claim 1 , wherein the plurality of generated receiver direction of motion hypotheses comprise four receiver direction of motion hypotheses, two in a first direction and two in a second direction, orthogonal to the first direction.

7. An apparatus for performing rapid initialization of a radio signal receiver, comprising: at least one processor; and at least one non-transient computer readable medium for storing instructions that, when executed by the at least one processor, causes the apparatus to perform operations comprising: receiving radio signals from a plurality of transmitters; applying an estimate of receiver position, time and the received radio signals to a neural network that is trained with a signal environment signatures; generating, from the neural network, a plurality of receiver direction of motion hypotheses; generating a plurality of phasor sequences, where each phasor sequence forms a phasor hypothesis comprising a signal phase representing one of the direction of motion hypotheses; correlating at least one local signal with at least one received signal to produce at least one correlation result; compensating a phase of the at least one of the local signal, the at least one received signal or the at least one correlation result based on the plurality of phasor hypotheses to generate a plurality of phase-compensated correlation results; determining a preferred hypothesis in the plurality of hypotheses that optimizes the at least one correlation result of at least one of the phase-compensated correlation results; and improving signal reception of the radio receiver using at least the phase compensated correlation result optimized by the preferred hypothesis.

8. The apparatus of claim 7, wherein the apparatus further performs: updating a motion model of the receiver using a direction of motion estimate associated with the preferred hypothesis.

9. The apparatus of claim 7, wherein the neural network is trained using street layouts, building models and GNSS signals that together form a signature of a signal environment for a given position of a receiver at a given time.

10. The apparatus of claim 9, wherein the neural network matches a trained signal signature with a current signal signature sensed by the receiver to generate the plurality of receiver direction of motion hypotheses.11 . The apparatus of claim 7, wherein the plurality of generated receiver direction of motion hypotheses comprise a reduced number of motion hypotheses as compared to receiver motion hypotheses generated without the neural network.

12. The apparatus of claim 7, wherein the plurality of generated receiver direction of motion hypotheses comprise four receiver direction of motion hypotheses, two in a first direction and two in a second direction, orthogonal to the first direction.

13. A system for performing rapid initialization of a radio signal receiver, comprising: at least one source of radio signals; and a radio signal receiver comprising: an antenna; a neural network; at least one processor; and at least one memory for storing programs and instructions that, when executed by the at least one processor, causes the receiver to perform operations comprising: receiving radio signals from the at least one source using the antenna;applying an estimate of receiver position, time and the received radio signals to the neural network that is trained with signal environment signatures; generating, from the neural network, a plurality of receiver direction of motion hypotheses; generating a plurality of phasor sequences, where each phasor sequence forms a phasor hypothesis comprising a signal phase representing one of the direction of motion hypotheses; correlating at least one local signal with at least one received signal to produce at least one correlation result; compensating a phase of the at least one of the local signal, the at least one received signal or the at least one correlation result based on the plurality of phasor hypotheses to generate a plurality of phase-compensated correlation results; determining a preferred hypothesis in the plurality of hypotheses that optimizes the at least one correlation result of at least one of the phase- compensated correlation results; and improving signal reception of the radio receiver using at least the phase compensated correlation result optimized by the preferred hypothesis.

14. The system of claim 13, wherein the receiver further performs: updating a motion model of the receiver using a direction of motion estimate associated with the preferred hypothesis15. The system of claim 13, wherein the neural network is trained using street layouts, building models and GNSS signals that together form a signature of a signal environment for a given position of a receiver at a given time.

16. The system of claim 13, wherein the neural network matches a trained signal signature with a current signal signature sensed by the receiver to generate the plurality of receiver direction of motion hypotheses.

17. The system of claim 13, wherein the plurality of generated receiver direction of motion hypotheses comprise a reduced number of motion hypotheses as compared to receiver motion hypotheses generated without the neural network.

18. The system of claim 13, wherein the plurality of generated receiver direction of motion hypotheses comprise four receiver direction of motion hypotheses, two in a first direction and two in a second direction, orthogonal to the first direction.

Images