►
From YouTube: Capstones & System Projects - Public WiFi
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
A
Thank
everybody
for
coming
out
today.
My
name
is
Daniel
Robinson
I'm,
a
network
analyst
34
department,
innovation
of
performance
and
I
had
the
pleasure
of
working
with
this
fine
group
here
from
the
Carnegie
Mellon
Heinz
College.
The
project
that
they're
working
on
was
pittsburgh
public
Wi-Fi
projects.
A
What
that
project
did
was
explore
the
feasibility
of
having
free,
citywide
Wi-Fi
for
the
different
neighborhoods
within
the
city
we
have
today
who
spearheaded
it
was
Professor,
secure
yourself
sitting
right
there
we
have
Lindsay
parham,
we
have
Terry
Gibbs
and
we
have
Katya
Nalla
Balu.
So
without
further
ado,
I'm
going
to
have
Lindy
Lindsay
come
up
and
she'll
explain
some
dance
to.
B
Thank
you
so
much
for
for
coming
to
view
our
project
on
the
hybrid
Wi-Fi
framework.
We
thank
you
again
to
Dana
Robinson.
He
was
invaluable
throughout
this
entire
semester.
On
helping
us
prepare
this
this
presentation.
As
you
know,
we
were
working
with
the
city
of
Pittsburgh's
Department
of
innovation
in
performance
and
they
came
to
us
at
the
beginning
of
the
semester
saying
we
really
want
to
look
into
the
feasibility
of
municipal
wide
Wi-Fi.
B
So
the
agenda
for
today
we're
going
to
talk
a
little
bit
about
problems
with
the
municipal
Wi-Fi
model,
then
we're
going
to
discuss
our
solution
to
those
problems
are
hybrid
model
framework
I'm,
going
to
touch
a
little
bit
on
each
of
the
frameworks
that
comprise
the
entire
hybrid
model
solution,
which
is
cost
analysis,
asset
mapping,
network
design
and
security.
Then
we're
going
to
go
into
some
of
the
potential
issues
with
implementing
this
particular
model
and
then
some
next
steps
in
solution
of
next
steps
and
recommendations
for
the
city.
B
So
we
spent
the
initial
part
of
the
semester
discovering
problems
with
municipal
Wi-Fi
models,
so
these
red
triangles
represent
all
the
failed
municipal,
large-scale
municipal
Wi-Fi
implementations
over
the
past
15
years.
What
we
found
is
by
and
large,
these
models,
these
Wi-Fi
implementations
fail
and
they
fail
because
they're
usually
supported
by
one
or
two
initial
stakeholders
the
demand
for
that
is
very,
very
low.
They
anticipate
the
initial
demand
for
a
municipal
or
free
Wi-Fi
network
to
be
about
thirty
to
fifty
percent
of
the
population,
so
they
anticipated
thirty
to
fifty
percent
of
the
population.
B
Would
log
on
to
this
free
Wi-Fi
in
actuality?
We
discovered
that
about
two
to
three
percent
of
the
population
will
actually
log
on
to
the
Wi-Fi,
but
the
cost
of
this
municipal
Wi-Fi
is
still
the
same
really
if,
regardless
of
whether
you
have
100,000
users
or
100
users,
because
the
equipment
costs
will
be
the
same,
so
the
investors
backed
out
of
these
municipal
Wi-Fi
implementations,
because
that
the
demand
just
was
not
there
going
to
the
next
slide.
But
what
we
did
notice
was
small
community
and
neighborhood.
B
Wi-Fi
models
have
been
successful
all
over
the
country,
and
so
we
we
did
case
studies
of
why
they
were
successful
and
we
came
up
with
three
main
traits
why
three
main
traits
of
successful
community
implementations.
So
all
successful
community
implementations
have
a
clear
objective.
They
have
appropriate
demand,
they've
already
calculated
that
the
demand
will
be
there
to
support
the
network
and
they
start
really
small
at
particularly
in
a
pilot
area.
So
a
clear
objective.
B
A
lot
of
the
objectives
example
objectives
would
be
economic
development
or
social
social
good,
and
then
they
determine
the
demand
by
youth
by
a
needs
assessment.
The
demand
usually
in
these
small
community
models,
the
demand
drives
the
network,
so
the
demand
or
the
problem
is
there
and
then
the
network
acts
as
the
solution
and
then,
additionally,
they
don't
start
on
a
large
scale.
B
Municipal
Wi-Fi
networks
fail
because
they
try
to
cover
the
entire
city,
the
entire
city,
and
that's
just
that's
too
expensive.
It
doesn't
give
them
it
doesn't
allow
for
time
to
test
or
for
time
to
do
to
improve
the
network
plan.
So
community
community
networks
are
very
small
and
they
start
on
a
small
scale.
B
Okay,
so
we
came
up
with
a
hybrid
model,
which
is
a
combination
of
both
the
community
networks
and
the
municipal
Wi-Fi
network.
It's
a
brand
new
idea
that
we
that
we
we
thought
of
to
both
combined
this
community
network,
but
then
also
the
needs
of
the
city,
because
the
city
does
have
an
invested
interest
in
getting
Wi-Fi
access
across
the
entire
across
the
entire
district.
So
a
community
model,
as
I
said
before,
is
a
community-owned
and
funded
Wi-Fi
implementation
that
the
city
provides
resources
to
so
the
city.
B
Whatever
resources
the
city
has
available
to
them,
they
make
available
to
the
community
networks.
So
it's
a
series
of
small
scale
and
neighborhood
implementations
that
are
combined
together
cover
the
majority
of
the
city
of
Pittsburgh.
So
we
decided
that
these
these
implementations
should
use
mesh
network
technology
and
we
decided
on
match
after
looking
carefully
at
multiple
other
other
implementations.
Mesh
is
great
because
it's
affordable,
it's
easy
to
use
and
maintain,
and
most
importantly,
it's
scalable.
B
Our
hybrid
model
relies
on
both
bandwidth
sharing
and
the
traditional
purchase.
Isp
bandwidth
sharing
is
look,
for
example,
you
have
a
large
company
and
they
have
X
amount
of
bandwidth
that
they
either
don't
use
every
month
or
they're
willing
to
donate
to
the
key
to
the
network
that
alleviates
a
good
deal
of
the
cost
off
the
network.
Both
start
the
initial
cost
and
the
maintaining
costs.
B
Okay.
So
why
does
model
is
great
for
the
city.
First,
I
should
talk
about
a
little
bit
of
the
constraints
and
requirements
that
the
city
city
had
initially.
First
of
all,
the
city
is
constrained
by
a
very
limited
budget.
Additionally,
the
city
would
prefer
to
outsource
the
maintenance
and
the
maintenance
of
this
network
to
a
third
party.
The
city
was
also
very
clear
that
it
had
to
be
scalable.
Even
if
we
start
in
a
small
area,
we
need
to
be
able
to
scale
this
network
out
to
reach
all
the
all
the
residents.
B
Security
is
also
a
very
important
concern
in
order
for
the
city
to
put
its
name
on
something
it
has
to
have
a
minimum
security
standard.
Also
legal,
they
didn't
want
to
do
anything
illegal.
How
we
needed
to
look
into
the
legality
of
implementing
a
Wi-Fi
system
and-
and
it's
funny
to
think
about,
but
actually
a
good
portion
of
our
present
of
our
hour
of
our
research
was
focused
on
this
particular
law
that
that
was
unclear.
B
Whether
or
not
the
city
was
even
legally
allowed
to
open
up
a
municipal,
Wi-Fi
implementation,
a
lot
of
other
cities
get
sued
and
there
they
invest
tens
of
millions
of
dollars
into
a
network
just
to
have
incumbent.
Isp
providers
take
them
to
court,
so
it
was
a
huge
consideration
so
why
our
model
fits
all
those
requirements
and
constraints?
Is
the
community
funds
this
model,
the
community?
The
owners
of
the
community
network
will
be.
B
The
primary
fund
would
provide
the
primary
funds
and
finances
for
this
model,
but
the
city
will
help
them
identify
grants
and
identify
grants
that
meet
their
objective.
To
offset
some
of
the
costs,
the
maintenance
of
the
model
would
also
be
on
the
network
owners.
Thus
the
scalability
it
really.
We've
really
created
a
flexible
framework
that
that
can
be
applied
to
not
just
one
area
but
the
entirety
of
of
Pittsburgh's
individual
neighborhoods.
B
The
security
we've
set
a
standard
baseline
security
model
that
all
networks
would
have
to
comply
to
if
they
were
trying
to
utilize
city
resources
and
we
looked
into
the
legality
of
it
and
we
are
fairly
cut.
We're
really
confident
that
we're
not
overstepping
any
federal
or
state
laws
and
also
since
the
community
would
own
the
community
network
owners
would
own
the
model.
The
city
would
not
be
held
liable.
B
Ok,
so
we're
community
models
are
already
working,
so
these
are
not
examples
of
hybrid
models,
but
there
are
examples
of
public
access
Wi-Fi
throughout
Pittsburgh.
So
it's
not
a
big
stretch
to
think
that
more
communities
would
be
interested
in
implementing
their
own
Wi-Fi
implementations.
We
have
wireless
shadyside
olive
Walnut.
Street
is
connected
in
shadyside,
with
free
Wi-Fi,
hosted
by
a
hotel
on
the
downtown
Pittsburgh.
Partnership,
as
you
know,
provides
limited
access,
Wi-Fi
and
then
wireless
waterways
provides
what
access
to
access
at
the
waterfront.
B
Ok.
So
what
did
we
deliver?
We
we
delivered
what
I
call
up
a
hybrid
Wi-Fi
framework
and
within
that
big
framework,
are
smaller
templates
and
frameworks
that
give
give
you
all
the
tools
you
need
to
create
and
implement
a
Wi-Fi
model,
this
particular
wife
Wi-Fi
model.
Those
frameworks
include
a
needs
assessment,
asset
mapping,
cost
analysis,
network
planning,
security
and
operation
standards
yeah.
So
those
are
the
individual
frameworks
and
during
this
presentation,
I'm
going
to
talk
particularly
about
cost
analysis
and
asset
mapping.
B
Okay,
so,
like
I,
said
before
a
lot
of
a
lot
of
municipal
Wi-Fi
networks
fail
because
they
start
two
large.
We
applied
our
framework
to
a
small
pilot
area
of
east
liberty.
Just
to
show
that
this
this
is
a
feasible,
a
feasible
model
that
can
actually
be
applied
to
neighborhoods
in
in
Pittsburgh.
So
we
chose
these
Liberty
as
our
pilot
area
for
because
it
has
a
lot
of
economic
development
opportunities,
particularly
I,
think
of
the
target
project.
What
I
think
of
economic
development
opportunities?
The
city
is
already
invested
in
this
area.
B
So
having
that
that,
on
that
strong
pole
is
really
great
for
the
sustainability
of
a
network
there's
also
it's
rich
in
resources
and
businesses.
We
looked
at
the
city
of
commerce
website
and
there
are
hundreds
of
businesses
located
within
East
Liberty.
Those
businesses
can
act
as
access
points
for
internet,
so
it
was
important
that
there
are
businesses,
it's
a
manageable
size,
it's
about
a
half
a
square
mile
and
it
has
6,000
6,000
residents
as
well
as
people
coming
in
and
out
to
take
advantage
of
retail
and
shopping.
B
And
then
there
are
a
lot
of
nonprofits
and
social
social
groups
there.
So
it's
a
it's
rich
in
its
rich
in
assets
and
that's
why
we
chose
it
as
our
pilot
area
so
asset
map,
the
idea
of
asset
mapping
is
identifying
the
large
business
stakeholders
in
the
area,
both
large
businesses,
small
businesses,
nonprofits
and
public
services.
So
these
green
dots
represent
the
businesses
we
took.
B
We
used
the
ArcGIS
and
some
of
the
GIS
data
from
the
city
of
Pittsburgh
website,
but
then
we
had
to
add
additional
additional
entities
onto
this
map
upload
it
and
we
have
identified
we've
identified
them
here.
So
this
is
just
a
cut
out
of
East
Liberty.
Oh,
go
ahead,
all
right,
so
this
is.
We
took
the
access
points.
B
We
identified
these
businesses
and
these
are
the
access
points
where
we
would
set
up
mesh
network
mesh
network
equipment
and
the
circles
represent
the
buffer
zone,
and
so
these
yellow
dots
here
would
be
where
we
would
have
our
internet
gateway.
This
is
where
you're
getting
the
ISP.
These
represent
large
businesses
that
would
most
likely
share
bandwidth
to
this
network.
Each
small
businesses
and
city
light
posts
will
host
nodes.
B
So
then
we
talked
about
funding
options.
Funding
is
obviously
a
huge
huge
component
of
of
a
municipal
Wi-Fi
network.
We
would
work
with
the
city
to
have.
We
worked
with
with
the
city
to
identify
grants
and
donations
and
crowd
funding
sources
for
this
network,
so
it
wouldn't
just
be
the
Community
Network
footing
the
bill.
They
could
rely
on
grants,
donations
from
the
businesses
in
the
area
and
then
a
crowdfunding.
It
is
a
new
popular
way
of
getting
of
getting
money
for
an
initiative.
We
ought
so.
B
We
took
an
example:
we
plugged
in
east
liberty
to
our
cost
analysis
framework,
and
we
decided
that
about
it
would
cost
113
thousand
dollars
to
create
a
mesh
network
in
East
Liberty,
90,000
of
which
would
be
dedicated
towards
labor.
So
if
the
city
was
somehow
able
to
offset
that
labor
cost,
it
would
be
much
more
feasible
to
start
this
network
at
a
lower
at
a
lower
price.
B
Okay,
so
some
potential
issues
and
then
the
next
step.
So
the
biggest
issues
with
this
particular
model
is
incentivizing
stakeholders
to
invest
so
part
of
the
majority
of
the
way
we
get
our
bandwidth
in
order
to
lower
the
cost
is
by
bandwidth
sharing,
particularly
through
hospitals,
universities,
big
retailers
like
Target.
They
would
have
to
share
their
bandwidth
to
offset
that
bandwidth.
Cost
incentivizing
them
to
do,
though,
do
so
could
be
an
issue
also.
We
picked
East
Liberty
because
it
was
really
rich
in
resources.
Some
areas
don't
have
as
many
resources.
B
B
So
the
city
of
Pittsburgh's
role,
I
talked
about
a
little
bit
about
what
the
city
of
Pittsburgh
could
do
throughout
this
throughout
this
initiative,
but,
most
of
but
most
importantly,
I
think
they
need
to
act
as
a
congregation
point
for
information.
There
are
so
many
rich
resources
out
there
on
the
web
that
aren't
that
could
help.
Each
of
these
networks
really
lay
out
a
framework
for
how
how
they
can
implement
and
at
what
cost
and
what's
worked
and
what
hasn't
worked.
B
But
it's
not
together
in
one
point,
so
if
the
city
could
perhaps
create
a
website
that
brings
all
those
resources
together.
That
would
be
the
most
important
thing
that
they
could
do
as
I
said
before.
Next
steps
create
a
website,
and
then
we've
also
we've
also
done
all
the
legwork
for
creating
a
municipal,
Wi-Fi
implementation
in
East
Liberty.
The
next
step
would
be
to
actually
create
it
by
reaching
out
to
stakeholders
and
gauging
their
interest
and
getting
them
involved.
I
want
to
thank
everyone.
B
C
Actually,
maybe
you
explain
this,
but
I
actually
not
clear
on
the
demand
side
for
this,
because
with
the
example
of
the
the
failed
municipal
ones
where
you
anticipated
was
25
to
40
percent
or
something
like
that,
and
it
actually
was
two
to
three
percent
and
I'm
wondering
whether
I
have
some
ideas
about
why
that
was,
but
but
I'm
wondering
whether
this
is
in
it.
Whether
there
actually
is
demand
say
any
celebrity.
If
you
looked
at
you
know
the
anticipated
users
and
so
forth,
part.
B
Of
our
framework
is
doing
a
needs
assessment,
which
is
one
of
the
very
first
things
that
you
should
do,
because
a
lot
of
municipal
Wi-Fi
networks
did
fail
because
of
that
demand
peace,
a
lot
of
the
community
networks
they're
built,
because
the
demand
already
exists.
A
lot
of
grassroots
movements
happen
be
organically
because
of
the
demand,
and
the
problem
is
already
there.
This
would
just
be
kind
of
a
different
way
of
helping
those
networks
along
and.
D
We
would
we
want
to
base
these
community
models
off
of
different
purposes,
so
the
demand
would
be
based
on
the
purpose.
So
if
we
in
Sadie's
Liberty,
the
purpose
was
economic
development,
we
would
assume
that
the
businesses
would
rather
share
their
bandwidth
and
reduce
their
costs
than
paying
for
their
own
internet.
And
if
the
purpose
was
education
in
a
certain
community,
then
the
schools
could
lower
their
costs.
By
doing
this,.
E
I,
like
your
approach
and
I,
think
it
has
great
merit.
Have
you
ever
thought
about
its
kind
of
switching
the
model
from
areas
that
have
a
relatively
high
demand,
like
you
know,
airs
that
are
developing
economically,
like
East
Liberty,
for
example,
where
users
there
may
our
be
high
demand
relatively
high
demand?
Have
you
thought
about
focusing
on
areas
where
there's
little
or
no
demand
yeah,
but
the
value
could
be
that
much
more
exponential
in
its
value,
yeah.
E
For
new
communities
like
or
new,
older
cunha,
sly
community,
like
larmer,
for
example,
that's
adjacent
Janice
Liberty
we're
used
to
have
16,000
families
there.
Now
you
have
1600,
maybe,
but
the
community
is
coming
back
and
developing.
Would
it
be
a
wonderful
kind
of
driver
and
an
amenity
to
have
Wi-Fi
for
the
community
even
before
its
kind
of
fully
formed?
So.
B
That
that's
actually
one
of
the
biggest
hurdles
from
our
presentation
when
we
were
just
trying
to
decide
wide
municipal
Wi-Fi
failed
was
failing
across
the
board
and
we
decided
it
was
because
low
demand.
We
had
to
look
at
why
why
the
demand
was
so
low
and
it
was
because
people
had
a
lot.
The
majority
of
people
have
ready
access
to
bandwidth
in
their
home
or
at
work,
and
they
also
have
cell
phone
data
plans,
and
so
we
said
we
decided
well.
How
do
you?
B
B
Those
errors
are
also
the
poorest
areas
within
Pittsburgh,
and
so,
if
you
were
to
put
Wi-Fi
in
communities
that
were
had
low
employment
rates,
high
high
school
dropout
rates,
single
single-mother
household,
you
would
get
the
highest
demand
because
they
just
don't
have
access
to
broadband
in
any
other
way,
with
the
model
that
working
under
the
constraints
of
it
has
to
be.
It
has
to
be
low
cost.
You
have
to
also
consider
okay
consider
who's
going
to
pay
for
it.
B
We
actually
came
up
with
a.
We
came
up
with
a
map
that
showed
these
target
areas
were
turned
out
to
be
some
of
the
poorest
areas
in
Pittsburgh,
but
then
they
had,
they
don't
have
the
cell
phone
to
connect
to
the
data
they
don't
her
to
the
internet.
They
don't
have
the
laptop
to
connect
to
the
internet
and
it
brought
up
a
whole
other
issue
where
you
would
have
to
get
them
the
device
to
connect
to
your
product.
E
No
I
think
it's
a
great
I
mean
your
approach,
I
think,
is
valid,
but
the
one
institution
that
has
kind
of
universal
appeal
and
is
geographically
distributed
throughout
the
cities.
Our
parks,
though,
and
I
know,
there's
no
community
their
lives
in
a
park,
but
the
about
ninety
percent
of
the
people
in
the
city
put
through
a
used
parks
over
the
course
of
a
year
or
three
know
a
few
months,
and
it
might
be
an
interesting
point
to
think
about
that
is
almost
a
public
utility
because
it's
becoming
that
way.
B
C
It
because
it
does
think
there
is
a
sweet
spot
for
need,
which
is
actually
communities
actually
have
the
portable
devices.
But
in
fact
you
know,
but
are
operating
them
on
the
type
of
plans
that
don't
have
a
cellular
data
plan
and
I.
Think
that
in
that
that's,
where
they're
sort
of
like
there's
no
opportunity
to
get
to
the
actual
to
the
internet
without
a
broadband.
Without
this
kind
of
Wi-Fi
thing
or
you
know,
or
the
or
business
that
has
it
and
so-
and
that
seems
like
where
you'd
have
a
really
intense
kind
of
demand.