agenda - network protocols...

Agenda

v  A hint or two on the Programming Assignment v  A very little bit about wireless networks and WiFi

(CSMA/CA) v  Future Internet Architecture Research v  Teacher-Course Evaluation

Wireless, Mobile Networks 6-1

Common framing errors

v  Checking to see if the output message contains the delimiter. DON’T DO THIS! In general, preconditions are the responsibility of the caller!

v  Failing to pass through a single ‘\n’ in the message! v  [Java] Using a List<Byte> or List<Integer> to store the

message in the framer


0 1

‘\n’ - ‘\n’

return buf[]

not ‘\n’ add ‘\n’ to buf; add input to buf[]

not ‘\n’ add input to buf[]

Receiving method State Machine:

Chapter 6 Wireless and Mobile Networks

Computer Networking: A Top Down Approach 6th edition Jim Kurose, Keith Ross Addison-Wesley March 2012

A note on the use of these ppt slides: We’re making these slides freely available to all (faculty, students, readers). They’re in PowerPoint form so you see the animations; and can add, modify, and delete slides (including this one) and slide content to suit your needs. They obviously represent a lot of work on our part. In return for use, we only ask the following: v  If you use these slides (e.g., in a class) that you mention their source

(after all, we’d like people to use our book!) v  If you post any slides on a www site, that you note that they are adapted

from (or perhaps identical to) our slides, and note our copyright of this material.

Thanks and enjoy! JFK/KWR All material copyright 1996-2012 J.F Kurose and K.W. Ross, All Rights Reserved



Ch. 6: Wireless and Mobile Networks Background: v  # wireless (mobile) phone subscribers now exceeds #

wired phone subscribers (at least 5-to-1)! v  # wireless Internet-connected devices equals #

wireline Internet-connected devices §  laptops, Internet-enabled phones promise anytime untethered

Internet access

v  two important (but different) challenges §  wireless: communication over wireless link §  mobility: handling the mobile user who changes point of

attachment to network


Elements of a wireless network

network infrastructure


wireless hosts v  laptop, smartphone v  run applications v  may be stationary (non-

mobile) or mobile §  wireless does not always

mean mobility




base station v  typically connected to

wired network v  relay - responsible for

sending packets between wired network and wireless host(s) in its “area” §  e.g., cell towers,

802.11 access points




wireless link v  typically used to connect

mobile(s) to base station v  also used as backbone

link v  multiple access protocol

coordinates link access v  various data rates,

transmission distance




Characteristics of selected wireless links

Indoor 10-30m

Outdoor 50-200m

Mid-range outdoor

200m – 4 Km

Long-range outdoor

5Km – 20 Km

.056

.384

1

4

5-11

54

2G: IS-95, CDMA, GSM

2.5G: UMTS/WCDMA, CDMA2000

802.15

802.11b

802.11a,g

3G: UMTS/WCDMA-HSPDA, CDMA2000-1xEVDO

4G: LTWE WIMAX

802.11a,g point-to-point

200 802.11n

Dat

a ra

te (M

bps)


infrastructure mode v  base station connects

mobiles into wired network

v  handoff: mobile changes base station providing connection into wired network




Wireless network taxonomy

single hop multiple hops

infrastructure (e.g., APs)

no infrastructure

host connects to base station (WiFi, WiMAX, cellular) which connects to

larger Internet

no base station, no connection to larger Internet (Bluetooth,

ad hoc nets)

host may have to relay through several

wireless nodes to connect to larger Internet: mesh net

no base station, no connection to larger Internet. May have to relay to reach other a given wireless node

MANET, VANET


IEEE 802.11 Wireless LAN 802.11b v  2.4-5 GHz unlicensed spectrum v  up to 11 Mbps v  direct sequence spread spectrum

(DSSS) in physical layer §  all hosts use same chipping

code

802.11a §  5-6 GHz range §  up to 54 Mbps

802.11g §  2.4-5 GHz range §  up to 54 Mbps

802.11n: multiple antennae §  2.4-5 GHz range §  up to 200 Mbps

v  all use CSMA/CA for multiple access v  all have base-station and ad-hoc network versions


802.11 LAN architecture v  wireless host

communicates with base station §  base station = access point

(AP)

v  Basic Service Set (BSS) (aka “cell”) in infrastructure mode contains: §  wireless hosts §  access point (AP): base

station §  ad hoc mode: hosts only

BSS 1

BSS 2

Internet

hub, switch or router


802.11: Channels, association

v  802.11b: 2.4GHz-2.485GHz spectrum divided into 11 channels at different frequencies §  AP admin chooses frequency for AP §  interference possible: channel can be same as that

chosen by neighboring AP!

v  host: must associate with an AP §  scans channels, listening for beacon frames containing

AP’s name (SSID) and MAC address §  selects AP to associate with §  may perform authentication [Chapter 8] §  will typically run DHCP to get IP address in AP’s

subnet


802.11: passive/active scanning

AP 2 AP 1

H1

BBS 2 BBS 1

1 2 3

1

passive scanning: (1)  beacon frames sent from APs (2)  association Request frame sent: H1 to

selected AP (3)  association Response frame sent from

selected AP to H1

AP 2 AP 1

H1

BBS 2 BBS 1

1 2 2

3 4

active scanning: (1) Probe Request frame broadcast

from H1 (2) Probe Response frames sent

from APs (3) Association Request frame sent:

H1 to selected AP (4) Association Response frame sent

from selected AP to H1


IEEE 802.11: multiple access v  avoid collisions: 2+ nodes transmitting at same time v  802.11: CSMA - sense before transmitting

§  don’t collide with ongoing transmission by other node

v  802.11: no collision detection! §  difficult to receive (sense collisions) when transmitting due to weak

received signals (fading) §  can’t sense all collisions in any case: hidden terminal, fading §  goal: avoid collisions: CSMA/C(ollision)A(voidance)

space

A B

C A B C

A’s signal strength

C’s signal strength


IEEE 802.11 MAC Protocol: CSMA/CA 802.11 sender 1 if sense channel idle for DIFS then

transmit entire frame (no CD) 2 if sense channel busy then

start random backoff time timer counts down while channel idle transmit when timer expires if no ACK, increase random backoff interval,

repeat 2

802.11 receiver - if frame received OK return ACK after SIFS (ACK needed due to

hidden terminal problem)

sender receiver

DIFS

data

SIFS

ACK


Avoiding collisions (more) idea: allow sender to “reserve” channel rather than random

access of data frames: avoid collisions of long data frames v  sender first transmits small request-to-send (RTS) packets

to BS using CSMA §  RTSs may still collide with each other (but they’re short)

v  BS broadcasts clear-to-send CTS in response to RTS v  CTS heard by all nodes

§  sender transmits data frame §  other stations defer transmissions

avoid data frame collisions completely using small reservation packets!


Collision Avoidance: RTS-CTS exchange

AP A B

time

RTS(A) RTS(B)

RTS(A)

CTS(A) CTS(A)

DATA (A)

ACK(A) ACK(A)

reservation collision

defer


frame control duration address

1 address

2 address

4 address

3 payload CRC

2 2 6 6 6 2 6 0 - 2312 4 seq

control

802.11 frame: addressing

Address 2: MAC address of wireless host or AP transmitting this frame

Address 1: MAC address of wireless host or AP to receive this frame

Address 3: MAC address of router interface to which AP is attached

Address 4: used only in ad hoc mode


Internet router H1 R1

AP MAC addr H1 MAC addr R1 MAC addr address 1 address 2 address 3

802.11 frame

R1 MAC addr H1 MAC addr dest. address source address

802.3 frame

802.11 frame: addressing

¡  Internet Protocol (IP) as the “waist of the hourglass” §  Basic interoperability: globally-

unique, routable addresses ¡  Transport Protocols above

§  TCP: reliable byte stream §  UDP: best-effort datagram

¡  Link/Network below §  “IP over everything”

¡  Applications on top

4/26/13 22

IP

TCP UDP

DNS

Application Protocols: HTTP, SMTP, etc.

Ether WiFi PPP ...

4/26/13 23

End systems (“hosts”)

Intermediate Systems (“routers”)

Routing Domains (AS’s)

Channels

4/26/13 24

Access Providers

Regional Providers

Tier 1 (Clique)

$

¥

€

Customers

When TCP/IP was born, the world was rather different than it is today...

It is truly amazing that something designed for such a different environment has become such a crucial component of global society. But...

4/26/13 25

Characteristic 1981 2013

Backbone Channel Capacity 5 x 104 bps 4 x 1010 bps

Personal Computer Storage < 106 bytes > 1011 bytes

Telecomm Service Providers ~102 ~104 (?)

Users/Computer ~10 ~1-‐0.1

Computers Connected ~104 ~109

The Internet has some known shortcomings: ¡  Difficult to change ¡  Routing & forwarding are entangled ¡  Single-path service ¡  Money flow ¡  Trust

4/26/13 26

¡  It is very difficult to change the core protocols and architecture of the Internet

¡  Providers are very conservative/proprietary ¡  Core protocols are embedded in silicon ¡  IETF processes often take a long time to converge

Consider IPv6 (minimal change: bigger addresses) §  Process started ca. 1994 §  RFC 2460: December 1998 §  Still not available everywhere! Not to mention: multicast, integrated services, resource

reservation...

4/26/13 27

¡  Routing (path selection) and forwarding (path elaboration) are both completely distributed §  Both happen hop-by-hop

¡  Each hop (router) makes an independent decision where to send each packet

¡  Each end-to-end path is determined by the collection of policies

4/26/13 28

Routing Policy

¡  Routing protocols admit only a single path between hosts (no multipath)

¡  Can’t take advantage of path diversity to: §  Increase bandwidth §  Improve robustness §  Differentiate service

4/26/13 29

4/26/13 30

¡  Most users choose between ≤ 2 access providers §  Access providers pay transit providers

¡  Money only enters the system at edge ¡  “Vote with your wallet” is hard

§  Even for large domains: Topology-based addressing change provider -> renumber

¡  Consequences: §  No provider controls the whole E2E path §  Providers cannot really compete to offer new

services

Access

Regional

Tier-1

$ ¥

€

¡  The Internet was originally a research project §  Focus was on just getting it to work

§  Researchers (mostly) all trusted each other

§  Original security features atrophied/were never implemented

¡  Trust is outside the current Internet Architecture

4/26/13 31

Part of NSF FIND Program, 2006-2011

... plus students Idea: How might the architecture be designed “from scratch” today?

4/26/13 32

Bobby Bhattacharjee, Neil Spring

James P. G. Sterbenz Ken Calvert, Jim Griffioen

¡  Identify channels, not nodes §  Network layer goes all the way to the application

¡  Flat identifiers from a large space (~160 bits) §  ID = hash of public key §  Every channel ID has a self-certifying public/private key pair §  Auto-assignable, no central registry §  Locator-identifier separation

¡  Hierarchical loose source routing §  Packet carries a forwarding directive (FD) = sequence of channel

IDs §  Push “intra” FD when entering a lower-level domain

4/26/13 33

¡  Naming nodes

4/26/13 34

F E

D C B

H I

K J

L G

3 1

2

4 5

9 8

6 7 A

?

?

?

?

¡  Naming Channels

4/26/13 35

2

a b c

d e

f

g i j

k

8

p m n o

q

r s

t

b

h u

v w

x

y z

4/26/13 36

i

k

a

b

c

d

e f g

h j m

To send to destination k, source sends request to topology server for path (or path set)

path request

Route server

source

ami abci amde

paths to server

4/26/13 37

i

k

a

b

c

d

e f g

h j

ami abci amde

m

To send to destination k, source sends request to topology server for path, receives response + motivation via reverse path.

path response

source

paths to server Route server

4/26/13 38

i

k

a

b

c

d

e f g

h j m

Source constructs and sends packet

FD: amdjhk x0,x1,x2

x3,x4

At each hop, motivation is checked before packet is relayed to next channel

4/26/13 39

i

k

a

b

c

d

e f

g

h j m

Domains are indistinguishable from routers in the architecture. (Unlike IP, which has limited hierarchy “baked in”.)

¡  Challenge: scaling. Identifier does not encode location! §  Which domain contains a particular destination channel?

¡  Solution: Locator = set of sequences of channels ¡  Requires channel ID-to-locator resolution service

§  Destination endpoint registers its attachment channel

§  Access provider extends path with ingress channel(s), recurses

4/26/13 40

e x y

p

q s

t

e → {xe, ye}

e → {pxe, qye}

e → {tpxe, sqye}

¡  Destination provider chooses ingress path(s) during locator construction

¡  Source provider chooses egress and top-level transit paths

¡  Missing channels supplied by pushing/popping FDs §  E.g., path through transit domain

4/26/13 41

FD: abcejpsvz

a b

c

d

e

f

g h

j

k

m n

p

q s v

u z

Locator: {psvz,qsuz}

¡  Goals: §  Stimulate research to explore, design and evaluate trustworthy

future Internet architectures §  Engage the community in long-range, transformative thinking §  Design and experiment with new ... networking concepts that

take into consideration the larger social, economic and legal issues

¡  Four projects initially funded: §  Nebula §  Mobility First §  eXtensible Internet Architecture (XIA) §  Named Data Networking (NDN)

¡  ChoiceNet added later

4/26/13 42

¡  J. Smith, U. Penn + Cornell, MIT, Princeton, Purdue, Stanford, Stevens Inst. of Tech, UC Berkeley, Delaware, Illinois, Texas, Washington

¡  Design Emphasis: Cloud-computing-centric architecture ¡  Key Components/principles:

§  Ultra-high-reliability secure backbone interconnects data centers §  Enable always-available computing & storage services §  “New trustworthy data, control and core networking approaches

to support the emerging cloud computing model”

4/26/13 43

¡  D. Raychaudhuri, Rutgers + UMass, MIT, Duke, Michigan, Wisconsin, UNC, Nebraska

¡  Design emphasis: “mobility as the norm, not exception” ¡  Key components/principles:

1.  Separation of naming & addressing (fast global resolution svc) 2.  Routing on flat, self-certifying addresses (pub key based) 3.  Generalized delay-tolerant routing w/in-net storage, computing 4.  Separate net management plane for enhanced visibility 5.  Privacy features for user and location data 6.  Integrated computing & storage layer at routers

4/26/13 44

¡  P. Steenkiste, Carnegie-Mellon + Boston U., Wisconsin ¡  Design emphasis: Support multiple “narrow waists” ¡  Key principles/components:

§  “Intrinsic security:” integrity and accountability properties based on self-certifying identifiers

§  Flexible identifier semantics: may refer to host, content, or service §  Forwarding based on packet-borne directed acyclic graph, with

fallback advice §  Design for network to evolve

4/26/13 45

¡  L. Zhang, UCLA + Colorado State, PARC, Arizona, Illinois, UC Irvine, Memphis, UC San Diego, Washington U/St. Louis, Yale

¡  Design emphasis: Content as first-class entity ¡  Key components/principles:

§  Focus on “what”, not “where” §  Names identify data, not nodes, interfaces, or channels §  Secure content, not channels §  Customers pay providers for delivering content

4/26/13 46

¡  T. Wolf, U. Mass + Kentucky, North Carolina State, RENCI

¡  Design emphasis: Encourage innovation by supporting multiple alternatives and user choice

¡  Key components/principles: §  Services as first-class entities §  “Economy Plane” to provide incentives and enable user choice @

fine timescales §  Verification mechanisms so users to “know what happened” §  Facilitate service creation by composition of other services

4/26/13 47

¡  Technology exists for network-based services §  Network components and paradigms ▪  Programmable routers ▪  Virtual networks

§  Abstractions to describe network services ▪  Protocols ▪  In-network services

¡  Remaining (big!) challenge: deployment in the Internet ¡  How to encourage deployment of innovative in-network

services? §  Recognize the (huge) role of economic incentives to encourage

provider participation

4/26/13 48

¡  Competition drives innovation §  Choices are exposed throughout protocol stack §  Users (or their applications) control choices

¡  “Encourage alternatives” §  Provide services with different

functionality, quality, and cost ¡  “Know what happened”

§  Evaluate service experience ¡  “Vote with your wallet”

§  Reward good services through continued use

“Encourage Alternatives”

“Vote With Your Wallet”

“Know What Happened”

Innovation Through Choice

4/26/13 49

agenda - network protocols...

Documents