# US 2020/0005036 A1 · Time-space de-noising for distributed sensors

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

- Number: US 2020/0005036 A1
- Document ID: US20200005036A1
- Status: Published application
- Kind: A1
- Application number: 16/455,009
- Filed: 2019-06-27
- Published: 2020-01-02
- Earliest claimed: 2018-06-28
- Assignees: NEC Laboratories America, Inc.
- Applicants: NEC Laboratories America, Inc.
- Inventors: Philip Nan JI; Shuji MURAKAMI; Giovanni MILIONE; Ting WANG
- Web page: https://patentgrep.com/patent/US20200005036A1

## CPC

- G01D 5/35358: {using backscattering to detect the measured quantity}
- G01D 3/02: with provision for altering or correcting the law of variation
- G01N 23/203: Measuring back scattering
- G06T 5/70: Denoising; Smoothing
- G06F 2218/06: by applying a scale-space analysis, e.g. using wavelet analysis
- G06F 2218/08: Feature extraction
- G01D: MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Aspects of the present disclosure describe systems, methods and structures employing time-space de-noising for distributed sensor.

## Claims

### Claim 1

An improved, computer-implemented method for time-space de-noising of distributed sensor data configured as part of a distributed sensor system, the improved method CHARACTERIZED BY:
simultaneously de-noising both time domain and spatial domain data.

### Claim 2 (depends on 1)

The improved method of claim 1 FURTHER CHARACTERIZED BY:
the simultaneous de-noising is performed in real-time concurrently with sensing operation.

### Claim 3 (depends on 1)

The improved method of claim 1 FURTHER CHARACTERIZED BY:
waterfall data is represented as a 2-dimensional array W(xi,tj), where xi=x1, x2, . . . , xI and tj=t1, t2, . . . tJ. xI and tJ are the number of elements in the columns and rows of the waterfall, respectively.

### Claim 4 (depends on 3)

The improved method of claim 3 FURTHER CHARACTERIZED BY:
for each location at each time instant (one point on the waterfall plot called “point-of-interest” or POI), say (xko, tlo), a virtual time-space window is selected, with Δx points on the space domain (x-axis), and Δt points on the time domain (t-axis) and the POI (xko, tlo) is at the center.

### Claim 5 (depends on 4)

The improved method of claim 4 FURTHER CHARACTERIZED BY:
the time-space window is be represented as W(xk, tl), where
x
k
=
x
k
o
-
Δ


x
2
,

.
.
.

,
x
k
o
+
Δ


x
2


and


t
l
=
t
l
o
-
Δ


t
2
,

.
.
.

,
t
l
o
+
Δ


t
2
.

### Claim 6 (depends on 5)

The improved method of claim 5 FURTHER CHARACTERIZED BY:
within the time-space window, the amount of pixels with sensor data value (the z-axis value) exceeding a threshold (Wth) is counted, say p(xko,tlo), and the minimum value of p is 0, so that none of the pixels within the window exceeds the threshold, and the maximum value of p is Δx×Δt, indicating that all pixels exceed the threshold; and
the ratio of p and the window size is then calculated to obtain the percentage ratio r for this POI expressed as:
r

(
x
k
o
,
t
l
o
)
=
∑
x
k

∑
t
l

f

(
W

(
x
k
,
t
l
)
)
Δ


x


Δ


t
×
100

%


where
f

(
W

(
x
k
,
t
l
)
)
=
{
1
,
W

(
x
k
,
t
l
)
≥
W
th
0
,
W

(
x
k
,
t
l
)
<
W
th
and, the ratio r is output as the de-noised value for a particular location at a particular time instant.
