Sensor based mesh network for body fall detection

waralligatorMobile - Wireless

Nov 21, 2013 (3 years and 8 months ago)

66 views

Sensor based mesh network for body fall detection

Dr. Kenneth Hsu,
Sam Skalicky,
Aniket

Mhatre
,
Sajin

George, Daniel Cheung

RIT Computer Engineering

Abstract






Medical

Application

:

Fall

detection

can

be

monitored

through

the

watch

if

it

worn

by

the

patients

and

a

medical

supervisor
.

Also,

other

vital

sign

data

can

be

sent

through

wireless

this

wireless

sensor

network
.



Nursing

homes
:

monitoring

the

elderly

and

handicapped

to

assist

them

without

intruding

into

their

personal

lives
.



Fire

fighting

departments
:

The

fire

fighters

can

be

assisted

if

the

signal

is

conveyed

through

the

watch

on

the

wireless

mesh

networks
.


The

proposed

framework

using

Ti

EZ
430

Chronos

development

kits

were

acquired

from

Texas

Instruments
.

The

eZ
430
-
Chronos

from

Texas

Instruments

wirelessly

transmits

accelerometer

data

for

motion

sensitive

control

and

provides

a

personal

display

for

notifications
.

Using

these

we

can

construct

a

mesh

network

of

patient

monitoring

devices

with

monitoring

devices

for

medical

personnel
.

This

network

will

notify

the

medical

personnel

when

a

patient

has

fallen

with

no

background

network

hardware

required
.

It

allows

for

a

quick

and

cheap

implementation

in

existing

environments
.

In

addition,

this

network

is

mobile,

so

for

events

off

campus

can

easily

be

held

without

any

setup

or

modifications

to

the

network
.

Application

Methodology

A

sensor

mesh

network

is

built

using

Texas

Instruments

eZ
430

Chronos

development

Tool
.

eZ
430

consists

of

a

CC
430
F
6137

processor,

LCD

display

,

integrated

pressure

sensor,

3
-
axis

accelerometer

for

motion

sensitive

control

and

a

RF

tranreceiver
.

The

kit

is

runs

an

algorithm

which

detects

a

human

fall
.

The

algorithm

runs

an

actual

human

fall

and

not

a

drop
.

The

3
-
axis

accelerometer

data

is

used

to

detect

the

fall
.

On

detection

of

the

fall

the

watch

transmits

an

alert

signal

to

all

the

adjacent

watches
.

All

the

watches

in

the

range

receive

the

signal
.

The

watches

worn

by

the

medical

attendants

would

notify

on

reception

alerting

them

of

the

emergency

and

give

them

detailed

information

on

the

patient

that

is

in

need
.

If

the

medical

attendant

is

not

in

the

range

of

the

watch

it

would

relay

the

signal

through

any

other

patient

watches

in

between
.

The

medical

personnel

can

then

attend

the

patient
.