# US 2019/0246305 A1 · Large scale radio frequency signal information processing and analysis system

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2019/0246305 A1
- Document ID: US20190246305A1
- Status: Published application
- Kind: A1
- Application number: 16/384,621
- Filed: 2019-04-15
- Published: 2019-08-08
- Earliest claimed: 2016-03-08
- Assignees: Aurora Insight Inc.
- Applicants: Aurora Insight Inc.
- Inventors: Brian Thomas Mengwasser; Jennifer Alvarez; Benjamin Kempke; Michael J. Skaug; Dan Lisogurski; Wyatt J. Tyree
- Web page: https://patentgrep.com/patent/US20190246305A1

## CPC

- H04W 24/02: Arrangements for optimising operational condition
- H04W 24/08: Testing, {supervising or monitoring} using real traffic
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/09: Supervised learning
- G06N 20/00: Machine learning
- H04B 17/346: {Noise values (signal-to-interference ratio [SIR] or carrier-to-interference ratio [CIR] H04B17/336)}
- H04W 64/003: {locating network equipment}
- G06N 3/045: Combinations of networks
- G06N 3/08: Learning methods
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- H04B: TRANSMISSION

## Abstract

A large-scale radio frequency signal information processing and analysis system that provides advanced signal analysis for telecommunication applications, including band capacity and geographical density determinations and detection, classification, identification, and geolocation of signals across a wide range of frequencies and across broad geographical areas. The system may utilize a range of novel algorithms for bin-wise processing, Rayleigh distribution analysis, telecommunication signal classification, receiver anomaly detection, transmitter density estimation, transmitter detection and location, geolocation analysis, telecommunication activity estimation, telecommunication utilization estimation, frequency utilization estimation, and data interpolation.

## Claims

### Claim 1

A large-scale radio frequency signal information processing and analysis system, comprising:
a computing device comprising at least a memory and one or more processors;
a signal analyzer, comprising at least a first plurality of programming instructions stored in the memory of, and operating on at least one processor of, the computing device;
wherein the first plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information comprising at least one detected or confirmed signal; and
analyze the signal information using at least a first algorithm designed to provide information about the radio frequency signal landscape;
a signal confirmer, comprising at least a second plurality of programming instructions stored in the memory of, and operating on at least one processor of, the computing device;
wherein the second plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information comprising at least one suspected signal;
confirm the existence of suspected signals within the signal information using at least a second algorithm;
for portions of the data in which the existence of a signal is confirmed, forward that portion of the data to the signal analyzer; and
for portions of the data in which the existence of a signal is not confirmed, store or discard that portion of the data; and
a signal detector, comprising at least a third plurality of programming instructions stored in the memory of, and operating on at least one processor of, the computing device;
wherein the third plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information;
detect radio frequency signals within the signal information using at least a third algorithm;
for portions of the data in which a signal is detected, forward that portion of the data to the signal analyzer;
for portions of the data in which no signal is detected, store or discard that portion of the data; and
for portions of the data in which a signal is suspected, forward that portion of the data to the signal confirmer.

### Claim 2 (depends on 1)

The system of claim 1, wherein one of the algorithms used to detect, confirm, or analyze signals is a novel implementation of a bin-wise processor, comprising at least an additional plurality of programming instructions stored in the at least one memory of, and operating on at least one processor of, the computing device, wherein the additional plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information;
average or sum the values in a given frequency bin over a number of frequency domain samples to reduce the variance in the signals and noise levels and thereby allow for low or negative signal-to-noise ratio (SNR) radio frequency signal detection;
find the median value in each frequency bin over a number of frequency domain samples to overcome the effects of a signal that is significantly stronger or weaker than the average;
find the maximum value, or minimum value, or both, in each frequency bin over a number of frequency domain samples and use to resolve signals that are continuously present and signals that are intermittent in time;
find the mean, median, maximum, minimum, variance, skew, and kurtosis values plus the time component for a bin or set of bins in a geographical area, and compare each statistic to the same statistic computed for that bin or set of bins over a larger area to identify deviations and identify anomalies;
perform bin-by-bin processing of the average, median, maximum, minimum, variance, skew and kurtosis values over a number of frequency domain samples to reduce the quantity of data in results and expedite processing;
adjust the number of frequency domain samples used in the average, median, minimum, maximum, variance, skew, and kurtosis to increase or decrease time resolution as needed and thereby aid in signal detection;
combine frequency bins as needed, using average, median, minimum, maximum, variance, skew, and kurtosis value to reduce quantity while also detecting the presence of signals;
combine the frequency domain samples over time to reduce noise and improve signal detection.

### Claim 3 (depends on 1)

The system of claim 1, wherein one of the algorithms used to detect, confirm, or analyze signals is a novel implementation of a Rayleigh distribution analyzer comprising at least an additional plurality of programming instructions stored in the at least one memory of, and operating on at least one processor of, the computing device, wherein the additional plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information;
find the Rayleigh distribution of in-phase and quadrature-phase samples from the signal information;
compare the Rayleigh distribution of signals over a frequency band of interest against noise distribution; and
apply thresholds to isolate signals.

### Claim 4 (depends on 1)

The system of claim 1, wherein one of the algorithms used to detect, confirm, or analyze signals is a novel implementation of a telecommunication signal classifier comprising at least an additional plurality of programming instructions stored in the at least one memory of, and operating on at least one processor of, the computing device, wherein the additional plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information;
use image processing algorithms in multiple frequency and time resolutions to produce compact datasets; and
use the compact datasets to train machine learning algorithms to detect, identify, and classify telecommunication signals.

### Claim 5 (depends on 1)

The system of claim 1, wherein one of the algorithms used to detect, confirm, or analyze signals is a novel implementation of a receiver anomaly analyzer comprising at least an additional plurality of programming instructions stored in the at least one memory of, and operating on at least one processor of, the computing device, wherein the additional plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information;
calculate a Kurtosis value of the amplitude of a given signal from a given receiver;
compare the Kurtosis value to a subsequent signal from the same receiver; and
apply thresholds to determine whether the subsequent signal is a valid signal.

### Claim 6 (depends on 1)

The system of claim 1, wherein one of the algorithms used to analyze signals is a novel implementation of a transmitter density estimator comprising at least an additional plurality of programming instructions stored in the at least one memory of, and operating on at least one processor of, the computing device, wherein the additional plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information;
map signal strength data across a geographical area;
survey a portion of the geographical area;
count the number of transmitters in the survey;
correlate the number of transmitters with signal strength data to estimate transmitters per signal strength; and
apply estimate to the remaining geographical area.

### Claim 7 (depends on 1)

The system of claim 1, wherein one of the algorithms used to analyze signals is a novel implementation of a transmitter detector and locator comprising at least an additional plurality of programming instructions stored in the at least one memory of, and operating on at least one processor of, the computing device, wherein the additional plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information;
examine multiple signal samples taken from different times or look angles at a given frequency or frequencies;
examine signals across multiple frequencies or groups of frequencies; and
correlate signals in a given geographical area to determine whether a transmitter location has been identified.

### Claim 8 (depends on 1)

The system of claim 1, wherein one of the algorithms used to analyze signals is a novel implementation of a geolocation analyzer comprising at least an additional plurality of programming instructions stored in the at least one memory of, and operating on at least one processor of, the computing device, wherein the additional plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information;
if the transmitter time of transmission is known, for each signal:
calculate the time of flight;
translate the time of flight into a range value at each sensor or observation location; and
calculate multilateral location from intersection of range values; and
if the transmitter time of transmission is not known, for each signal:
use multiple signal observations to calculate the time of arrival, irrespective of the start of transmission being an unknown;
translate the time of arrival into a range value at each sensor; and
calculate multilateral location from intersection of range values.

### Claim 9 (depends on 8)

The system of claim 8, wherein the geolocation analyzer uses an alternate method of determining location comprising:
receiving signal information;
determining the best location estimate based on highest received signal power; and
examining detected signal power and the geographical extent over which the signal is observed to determine the error associated with this estimate of best location.

### Claim 10 (depends on 1)

The system of claim 1, wherein one of the algorithms used to analyze signals is a novel implementation of a telecommunication activity estimator comprising at least an additional plurality of programming instructions stored in the at least one memory of, and operating on at least one processor of, the computing device, wherein the additional plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information;
for each signal, estimate the time that the signal is above the noise floor;
for each signal, estimate the frequency of occurrences where the signal energy is above the noise floor;
produce a Rayleigh distribution of estimated values; and
correlate the distribution within the frequency band versus the capacity of the band.

### Claim 11 (depends on 1)

The system of claim 1, wherein one of the algorithms used to analyze signals is a novel implementation of a telecommunication utilization estimator comprising at least an additional plurality of programming instructions stored in the at least one memory of, and operating on at least one processor of, the computing device, wherein the additional plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information;
for each signal, detect the frequency edges of the signal;
for each signal, estimate the bandwidth compared to the overall frequency band; and
compile all signals within the band to determine the total band utilization.

### Claim 12 (depends on 1)

The system of claim 1, wherein one of the algorithms used to analyze signals is a novel implementation of a frequency utilization estimator comprising at least an additional plurality of programming instructions stored in the at least one memory of, and operating on at least one processor of, the computing device, wherein the additional plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information;
where a frequency band has been defined, use machine learning algorithms to identify clustering of frequencies;
where a frequency band has not been defined, use a convolutional filter tuned in time and frequency characteristics to a specific signal type, and convolve the filter over a large time-frequency space, using peaks of the result to identify a signal or signals of interest, from which utilization over the larger frequency can be estimated.

### Claim 13 (depends on 1)

The system of claim 1, wherein one of the algorithms used to analyze signals is a novel implementation of a data interpolator comprising at least an additional plurality of programming instructions stored in the at least one memory of, and operating on at least one processor of, the computing device, wherein the additional plurality of programming instructions, when operating on the at least one processor, cause the computing device to:
receive signal information;
combine samples in the time domain;
combine samples in the frequency domain;
fill in missing sample data by incorporating additional known information;
where applicable, apply a reverse propagation model to determine transmitter antenna patterns, and use measured data to improve propagation models.
