Special Topics on Wireless Ad-hoc Networks

1
Special Topics on Wireless Ad-hoc Networks
Lecture 3 Wireless LANs

• University of Tehran
• Dept. of EE and Computer Engineering
• By
• Dr. Nasser Yazdani

2
Covered topic

• How to build a small wireless network?
• considerations
• Media access issues
• References
• Chapter 3 of the book
• Wireless Medium Access control protocols a
  survey
• MACAW A Media Access Protocol for Wireless
  LANs
• Design alternative for Wireless local area
  networks,
• Bluetooth
• SSCH Slotted Seeded Channel Hopping for Capacity
  
• ECHOS Enhanced Capacity 802.11 Hotspots
• A backoff Algorithm for Improving Saturation
  Throughput in IEEE 802.11 DCF
• A wireless MAC protocol Using Implicit Pipelining

3
Outlines

• Why wireless LANs? Applications?
• Wireless LANs issues
• 802.11 standard
• Mac protocols.
• Bluetooth
• ZigBee

4
What is special on wireless?

• Channel characteristics
• Half-Duplex
• Location dependency
• Very noisy channel, fading effects, etc.,
• Resource limitation
• Bandwidth
• Frequency
• Battery, power.
• Wireless problems are usually optimization
  problems.

5
Why wireless networks?

• Mobility to support mobile applications
• Costs reductions in infrastructure and operating
  costs no cabling or cable replacement
• Special situations No cabling is possible or it
  is very expensive.
• Reduce downtime Moisture or hazards may cut
  connections.

6
Applications ?

• Pervasive computing or nomadic access.
• Ad hoc networking Where it is difficult or
  impossible to set infrastructure.
• LAN extensions Robots or industrial equipment
  communicate each others. Sensor network where
  elements are two many and they can not be wired!.

7
Ideal Wireless LAN?

• Wish List
• High speed
• Low cost
• No use/minimal use of the mobile equipment
  battery
• Can work in the presence of other WLAN
• Easy to install and use
• Etc

8
Wireless LAN Design Alternatives

• Wireless LAN Design Goals
• Portable product Different countries have
  different regulations concerning RF band usage.
• Low power consumption
• License free operation
• Multiple networks should co-exist
• Design Choices
• Physical Layer IR or RF?
• Radio Technology Direct-Sequence or
  Frequency-Hopping?
• Which frequency range to use?
• Which MAC protocol to use.
• Peer-Peer architecture or Base-Station approach?

9
Physical Layer Alternatives

• IR
• Simple circuitry, cost-effective, no regulatory
  constraints, no Rayleigh fading (waves are
  small), also nice for micro-cellular networks...
  (multiple cells can exist in a room providing
  more bandwidth)
• RF
• more complicated circuitry, regulatory
  constraints (ISM bands) in the U.S.

10
Physical Layer Alternatives
IR RF
Cost lt10 lt20
Regulation None No license on ISM bands
Interference Ambient Light Radiators
coverage Spot Wide Area
Performance Moderate Depends on Bandwidth
Multiple networks Limited Possible
11
Radio Technology

• Spread Spectrum Technologies
• Frequency Hopping The sender keeps changing the
  carrier wave frequency at which its sending its
  data. Receiver must be in synch with transmitter,
  and know the ordering of frequencies.
• Direct-Sequence The receiver listens to a set of
  frequencies at the same time. The subset of
  frequencies that actually contain data from the
  sender is determined by spreading code, which
  both the sender and receiver must know. This
  subset of frequencies changes during
  transmission.
• Non-Spread Spectrum requires licensing

12
Frequency Hopping versus Direct Sequence

• DS advantages
• Lower cost
• FH advantages
• Higher capacity
• Interference avoidance capability If some
  frequency has interference on it, simply don't
  hop there.
• Multiple networks can co-exist Just use a
  different frequency hopping pattern.

13
LAN Industry

• WANs are offered as service
• Cost of infrastructure
• Coverage area
• LANs are sold as end products
• You own, no service charge
• Analogy with PSTN/PBX
• WLAN vs. WAN Cellular Networks
• Data rate (2 Mbps vs. 54 Mbps)
• Frequency band regulation (Licensing)
• Method of data delivery (Service vs. own)

14
Growth of Home wireless
15
LAN standard

• IEEE 802 Standards
• 802.3, 802.4, 802.5 are wired LANs
• 802.9 ISO Ethernet
• 802.6 MAN
• 802.11, 802.15, 802.16 Wireless local net
• 802.14 Cable modem
• 802.10 Security management

16
LAN standard
17
Early Experiences

• IBM Switzerland,Late 1970
• Factories and manufacturing floors
• Diffused IR technology
• Could not get 1 Mbps
• HP Labs, Palo Alto, 1980
• 100 Kbps DSSS around 900 Mhz
• CSMA as MAC
• Experimental licensing from FCC
• Frequency administration was problematic, thus
  abandoned
• Motorola, 1985
• 1.73 GHz
• Abandoned after FCC difficulties

18
Architectures

• Distributed wireless Networks also called Ad-hoc
  networks
• Centralized wireless Networks also called last
  hop networks. They are extension to wired
  networks.

19
Base-Station Approach Advantages over Peer-Peer

• No hidden terminal base station hears all mobile
  terminals, are relays their information to ever
  mobile terminal in cell.
• Higher transmission range
• Easy expansion
• Better approach to security
• Problem?
• Point of failure,
• Feasibility?

20
Wireless LAN Architecture
Ad Hoc
Laptop
Laptop
Server
DS
Pager
Laptop
PDA
Laptop
21
Access Point Functions

• Access point has three components
• Wireless LAN interface to communicate with nodes
  in its service area
• Wireline interface card to connect to the
  backbone network
• MAC layer bridge to filter traffic between
  sub-networks. This function is essential to use
  the radio links efficiently

22
Medium Access Control

• Wireless channel is a shared medium
• Need access control mechanism to avoid
  interference
• MAC protocol design has been an active area of
  research for many years. See Survey.

23
MAC A Simple Classification
Wireless MAC
Centralized
Distributed
On Demand MACs, Polling
Guaranteed or controlled access
Random access
Our focus
SDMA, FDMA, TDMA, Polling
24
Wireless MAC issues

• Half duplex operations difficult to receive data
  while sending
• Time varying channel Multipath propagation,
  fading
• Burst Channel error BER is as high as 10-3. We
  need a better strategy to overcome noises.
• Location dependant carrier sensing signal decays
  with path length.
• Hidden nodes
• Exposed nodes
• Capture when a receiver can cleanly receive data
  from two sources simultaneously, the farther one
  sounds a noise.

25
Performance Metrics

• Delay ave time on the MAC queue
• Throughput fraction used for data transmission.
• Fairness Not preference any node
• Stability handle instantaneous loads greater
  than its max. capacity.
• Robust against channel fading
• Power consumption or power saving
• Support for multimedia

26
Wireless LAN Architecture, Cont
Logical Link Control Layer
MAC Layer Consist of two sub layer, physical
Layer and physical convergence layer

• Physical convergence layer, shields LLC from the
  specifics of the physical medium. Together with
  LLC it constitutes equivalent of Link Layer of OSI

27
Multi-Channel MAC A simple approach

• Divide bandwidth into multiple channels
• Choose any one of the idle channels
• Use a single-channel protocol on the chosen
  channel
• ALOHA
• MACA

28
Multiple Channels

• Multiple channels in ad hoc networks typically
  defined by a particular code (CDMA) or frequency
  band (FDMA)
• TDMA requires time synchronization among hosts in
  ad hoc network
• Difficult
• Many MAC protocols have been proposed

29
MAC Network Topology

• CDMA Not beneficial under current regulations -
  difficult to get good spreading codes
• FDMA Inefficient spectrum utilization for bursty
  traffic
• CSMA Suitable for Peer-to Peer architecture
• TDMA favors Base-Station/Remote-Station
  architecture

30
CSMA versus TDMA

• CSMA Advantages
• Can be implemented on an Ethernet chipset
• TDMA advantages
• simple remote stations
• isochronous traffic supported (low-latency,
  consistent throughput for such things as voice)
• high power saving potential (only need to listen
  at certain times)

31
Integrated CSMA/TDMA MAC Protocol

• Supports guaranteed bandwidth traffic and random
  access traffic
• The bandwidth is divided into a random part and a
  reserved part.
• Random part is LBT, reserved part
• During high traffic all bandwidth can be used for
  reserved traffic (like wireless telephony)

H1
Reserved-1
H2
Reserved-2
H3
LBT
32
Reservation/Polling MAC Protocol

• Works only with AP
• Fair and slow. First-in-First-Out
• Wireless station send a request.
• All requests are queued.
• Wireless stations are polled in the same order
  that the requests have arrive.
• All data reception is acknowledged.

33
Power Management

• Battery life of mobile computers/PDAs are very
  short. Need to save
• The additional usage for wireless should be
  minimal
• Wireless stations have three states
• Sleep
• Awake
• Transmit

34
Power Management, Cont

• AP knows the power management of each node
• AP buffers packets to the sleeping nodes
• AP send Traffic Delivery Information Message
  (TDIM) that contains the list of nodes that will
  receive data in that frame, how much data and
  when?
• The node is awake only when it is sending data,
  receiving data or listening to TDIM.

35
IEEE 802.11 WLAN, History

• 1997 IEEE 802.11 working group developed standard
  for inter-working wireless LAN products for 1 and
  2 Mbps data rates in 2.4 GHz ISM (industrial,
  scientific, and medical) band (2400-2483 MHz)
• Required that mobile station should communicate
  with any wired or mobile station transparently
  (802.11 should appear like any other 802 LAN
  above MAC layer), so 802.11 MAC layer attempts to
  hide nature of wireless layer (eg, responsible
  for data retransmission)

36
802.11 WLAN History, Cont..

• 1999 IEEE 802.11a amendment for 5 GHz band
  operation and 802.11b amendment to support up to
  11 Mbps data rate at 24 GHz
• Different standards a, b, e, etc., differ in
  physical link properties, services, etc.
• MAC sub layer uses CSMA/CA (carrier sense
  multiple access with collision avoidance)

37
802.11 Features

• Power management NICs to switch to lower-power
  standby modes periodically when not transmitting,
  reducing the drain on the battery. Put to sleep,
  etc.
• Bandwidth To compress data
• Security
• Addressing destination address does not always
  correspond to location.

38
IEEE 802.11 Topology

• Independent basic service set (IBSS) networks
  (Ad-hoc)
• Basic service set (BSS), associated node with an
  AP
• Extended service set (ESS) BSS networks
• Distribution system (DS) as an element that
  interconnects BSSs within the ESS via APs.

39
Starting an IBSS

• One station is configured to be initiating
  station, and is given a service set ID (SSID)
• Starter sends beacons
• Other stations in the IBSS will search the medium
  for a service set with SSID that matches their
  desired SSID and act on the beacons and obtain
  the information needed to communicate
• There can be more stations configured as
  starter.

40
ESS topology

• connectivity between multiple BSSs, They use a
  common DS

41
Starting an ESS

• The infrastructure network is identified by its
  extended service set ID (ESSID)
• All APs will have been set according to this
  ESSID
• On power up, stations will issue probe requests
  and will locate the AP that they will associate
  with.

42
802.11 Logical Architecture

• PLCP Physical Layer Convergence Procedure
• PMD Physical Medium Dependent
• MAC provides asynchronous, connectionless service
• Single MAC and one of multiple PHYs like DSSS,
  OFDM, IR
• and FHSS.

43
802.11 MAC Frame Format
Bytes

342346
32
6
Preamble PLCP header MPDU
6
2
6
6
4
2
2
6
Bytes
Encrypted to WEP
Bits
2
1
2
4
1
1
44
802.11 MAC Frame Format

• Address Fields contains
• Source address
• Destination address
• AP address
• Transmitting station address
• DS Distribution System
• User Data, up to 2304 bytes long

45
IEEE 802.11 LLC Layer

• Provides three type of service for exchanging
  data between (mobile) devices connected to the
  same LAN
• Acknowledged connectionless
• Un-acknowledged connectionless, useful for
  broadcasting or multicasting.
• Connection oriented
• Higher layers expect error free transmission

46
IEEE 802.11 LLC Layer, Cont..

• Each SAP (Service Access Point) address is 7
  bits. One bit is added to it to indicate whether
  it is order or response.
• Control has three values
• Information, carry user data
• Supervisory, for error control and flow control
• Unnumbered, other type of control packet

47
IEEE 802.11 LLC lt-gt MAC Primitives

• Four types of primitives are exchanged between
  LLC and MAC Layer
• Request order to perform a function
• Confirm response to Request
• Indication inform an event
• Response inform completion of process began by
  Indication

48
Reception of packets

• AP Buffer traffic to sleeping nodes
• Sleeping nodes wake up to listen to TIM (Traffic
  Indication Map) in the Beacon
• AP send a DTIM (Delivery TIM) followed by the
  data for that station.
• Beacon contains, time stamp, beacon interval,
  DTIM period, DTIM count, sync info, TIM broadcast
  indicator

49
Frame type and subtypes

• Three type of frames
• Management
• Control
• Asynchronous data
• Each type has subtypes
• Control
• RTS
• CTS
• ACK

50
Frame type and subtypes, Cont..

• Management
• Association request/ response
• Re-association request/ response transfer from
  AP to another.
• Probe request/ response
• privacy request/ response encrypting content
• Authentication to establish identity
• Beacon (Time stamp, beacon interval, channels
  sync info, etc.)

51
Frame type and subtypes, Cont..

• Management
• TIM (Traffic Indication Map) indicates traffic to
  a dozing node
• dissociation

52
802.11 Management Operations

• Scanning
• Association/Reassociation
• Time synchronization
• Power management

53
Scanning in 802.11

• Goal find networks in the area
• Passive scanning
• Not require transmission
• Move to each channel, and listen for Beacon
  frames
• Active scanning
• Require transmission
• Move to each channel, and send Probe Request
  frames to solicit Probe Responses from a network

54
Association in 802.11
1 Association request
2 Association response
AP
3 Data traffic
Client
55
Reassociation in 802.11
1 Reassociation request
New AP
3 Reassociation response
5 Send buffered frames
2 verifypreviousassociation
Client
6 Data traffic
Old AP
4 send buffered frames
56
Time Synchronization in 802.11

• Timing synchronization function (TSF)
• AP controls timing in infrastructure networks
• All stations maintain a local timer
• TSF keeps timer from all stations in sync
• Periodic Beacons convey timing
• Beacons are sent at well known intervals
• Timestamp from Beacons used to calibrate local
  clocks
• Local TSF timer mitigates loss of Beacons

57
Power Management in 802.11

• A station is in one of the three states
• Transmitter on
• Receiver on
• Both transmitter and receiver off (dozing)
• AP buffers packets for dozing stations
• AP announces which stations have frames buffered
  in its Beacon frames
• Dozing stations wake up to listen to the beacons
• If there is data buffered for it, it sends a poll
  frame to get the buffered data

58
Authentication

• Three levels of authentication
• Open AP does not challenge the identity of the
  node.
• Password upon association, the AP demands a
  password from the node.
• Public Key Each node has a public key. Upon
  association, the AP sends an encrypted message
  using the nodes public key. The node needs to
  respond correctly using it private key.

59
IEEE 802.11 Wireless MAC

• Distributed and centralized MAC components
• Distributed Coordination Function (DCF)
• Point Coordination Function (PCF)
• DCF suitable for multi-hop and ad hoc networking
• DCF is a Carrier Sense Multiple Access/Collision
  Avoidance (CSMA/CA) protocol

60
IEEE 802.11 DCF

• Uses RTS-CTS exchange to avoid hidden terminal
  problem
• Any node overhearing a CTS cannot transmit for
  the duration of the transfer
• Uses ACK to achieve reliability
• Any node receiving the RTS cannot transmit for
  the duration of the transfer
• To prevent collision with ACK when it arrives at
  the sender
• When B is sending data to C, node A will keep
  quite

61
Hidden Terminal Problem Tobagi75

• Node B can communicate with A and C both
• A and C cannot hear each other
• When A transmits to B, C cannot detect the
  transmission using the carrier sense mechanism
• If C transmits, collision will occur at node B

62
MACA Solution for Hidden Terminal Problem Karn90

• When node A wants to send a packet to node B,
  node A first sends a Request-to-Send (RTS) to A
• On receiving RTS, node A responds by sending
  Clear-to-Send (CTS), provided node A is able to
  receive the packet
• When a node (such as C) overhears a CTS, it keeps
  quiet for the duration of the transfer
• Transfer duration is included in RTS and CTS both

63
IEEE 802.11
RTS Request-to-Send
RTS
C
F
A
B
E
D
64
IEEE 802.11
RTS Request-to-Send
RTS
C
F
A
B
E
D
NAV 10
NAV remaining duration to keep quiet
65
IEEE 802.11
CTS Clear-to-Send
CTS
C
F
A
B
E
D
66
IEEE 802.11

• DATA packet follows CTS. Successful data
  reception acknowledged using ACK.

CTS Clear-to-Send
CTS
C
F
A
B
E
D
NAV 8
67
IEEE 802.11
DATA
C
F
A
B
E
D
68
IEEE 802.11
Reserved area
ACK
C
F
A
B
E
D
69
IEEE 802.11
DATA
C
F
A
B
E
D
70
IEEE 802.11
ACK
C
F
A
B
E
D
71
CSMA/CA

• Carrier sense in 802.11
• Physical carrier sense
• Virtual carrier sense using Network Allocation
  Vector (NAV)
• NAV is updated based on overheard
  RTS/CTS/DATA/ACK packets, each of which specified
  duration of a pending transmission
• Collision avoidance
• Nodes stay silent when carrier sensed
  (physical/virtual)
• Backoff intervals used to reduce collision
  probability

72
Backoff Interval

• When transmitting a packet, choose a backoff
  interval in the range 0,cw
• cw is contention window
• Count down the backoff interval when medium is
  idle
• Count-down is suspended if medium becomes busy
• When backoff interval reaches 0, transmit RTS

73
DCF Example
B1 and B2 are backoff intervals at nodes 1 and 2
cw 31
74
Backoff Interval

• The time spent counting down backoff intervals is
  a part of MAC overhead
• Choosing a large cw leads to large backoff
  intervals and can result in larger overhead
• Choosing a small cw leads to a larger number of
  collisions (when two nodes count down to 0
  simultaneously)

75
Backoff Interval (cont)

• Since the number of nodes attempting to transmit
  simultaneously may change with time, some
  mechanism to manage contention is needed
• IEEE 802.11 DCF contention window cw is chosen
  dynamically depending on collision occurrence

76
Binary Exponential Backoff in DCF

• When a node fails to receive CTS in response to
  its RTS, it increases the contention window
• cw is doubled (up to an upper bound)
• When a node successfully completes a data
  transfer, it restores cw to Cwmin
• cw follows a sawtooth curve
• 802.11 has large room for improvement

Random backoff
Data Transmission/ACK
RTS/CTS
77
Inter Frame Spacing

• SIFS Short inter frame space dependent on PHY
• PIFS point coordination function (PCF) inter
  frame space SIFS slot time
• DIFS distributed coordination function (DCF)
  inter frame space PIFS slot time
• The back-off timer is expressed in terms of
  number of time slots.

78
802.11 Frame Priorities

• Short interframe space (SIFS)
• For highest priority frames (e.g., RTS/CTS, ACK)
• PCF interframe space (PIFS)
• Used by PCF during contention free operation
• DCF interframe space (DIFS)
• Minimum medium idle time for contention-based
  services

DIFS
PIFS
contentwindow
Frame transmission
Busy
SIFS
Time
79
SIFS/DIFS

• SIFS makes RTS/CTS/Data/ACK atomic

RTS
Data
Time
Sender1
CTS
ACK
SIFS
SIFS
SIFS
DIFS
Time
Receiver1
RTS
DIFS
Time
Sender2
80
MACA protocol

• Key observation in CSMA/CA any node hearing RTS
  or CTS differ communication.
• This is to avoid collision with ACKs.
• We can leave ACKs, Reliability to upper layer.
• Any node hearing RTS, not CTS, only need to
  differ the RTS sender to receive CTS.
• Then, that node can start communication. No
  exposed node.

81
MACAW

• Based on MACA
• Design based on 4 key observations
• Contention is at receiver, not the sender
• Congestion is location dependent
• To allocate media fairly, learning about
  congestion levels should be a collective
  enterprise
• Media access protocol should propagate
  synchronization information about contention
  periods, so that all devices can contend
  effectively

82
MILD Algorithm in MACAW

• When a node successfully completes a transfer,
  reduces cw by 1
• In 802.11 cw is restored to cwmin
• In 802.11, cw reduces much faster than it
  increases
• MACAW cw reduces slower than it increases
• Exponential Increase Linear Decrease
• MACAW can avoid wild oscillations of cw when
  large number of nodes contend for the channel

83
Receive-Initiated Mechanism

• In most protocols, sender initiates a transfer
• Alternatively, a receiver may send a
• Ready-To-Receive (RTR) message to a sender
  requesting it to being a packet transfer
• Sender node on receiving the RTR transmits data
• How does a receiver determine when to poll a
  sender with RTR?
• Based on history, and prediction of traffic from
  the sender

84
Reliability

• Wireless links are prone to errors. High packet
  loss rate detrimental to transport-layer
  performance.
• Mechanisms needed to reduce packet loss rate
  experienced by upper layers
• When node B receives a data packet from node A,
  node B sends an Acknowledgement (Ack). This
  approach adopted in many protocols
• If node A fails to receive an Ack, it will
  retransmit the packet

85
Fairness Issue

• Assume that initially, A and B both choose a
  backoff interval in range 0,31 but their RTSs
  collide
• Nodes A and B then choose from range 0,63
• Node A chooses 4 slots and B choose 60 slots
• After A transmits a packet, it next chooses from
  range 0,31
• It is possible that A may transmit several
  packets before B transmits its first packet

A
B
Two flows
C
D
86
MACAW Solution for Fairness

• When a node transmits a packet, it appends the cw
  value to the packet, all nodes hearing that cw
  value use it for their future transmission
  attempts
• Since cw is an indication of the level of
  congestion in the vicinity of a specific receiver
  node, MACAW proposes maintaining cw independently
  for each receiver
• Using per-receiver cw is particularly useful in
  multi-hop environments, since congestion level at
  different receivers can be very different

87
Another MACAW Proposal

• For the scenario below, when node A sends an RTS
  to B, while node C is receiving from D, node B
  cannot reply with a CTS, since B knows that D is
  sending to C
• When the transfer from C to D is complete, node B
  can send a Request-to-send-RTS to node A.
• Node A may then immediately send RTS to node B
• This approach, however, does not work in the
  scenario below
• Node B may not receive the RTS from A at all, due
  to interference with transmission from C

D
C
B
A
88
Priorities in 802.11

• CTS and ACK have priority over RTS
• After channel becomes idle
• If a node wants to send CTS/ACK, it transmits
  SIFS duration after channel goes idle
• If a node wants to send RTS, it waits for DIFS gt
  SIFS

89
SIFS and DIFS
DATA1
ACK1
backoff
RTS
DIFS
SIFS
SIFS
90
Energy Conservation

• Since many mobile hosts are operated by
  batteries, MAC protocols which conserve energy
  are of interest
• Two approaches to reduce energy consumption
• Power save Turn off wireless interface when
  desirable
• Power control Reduce transmit power

91
Power Control with 802.11

• Transmit RTS/CTS/DATA/ACK at least power level
  needed to communicate with the receiver
• A/B do not receive RTS/CTS from C/D. Also do not
  sense Ds data transmission
• Bs transmission to A at high power interferes
  with reception of ACK at C

B
C
D
A
92
A Plausible Solution

• RTS/CTS at highest power, and DATA/ACK at
  smallest necessary power level
• A cannot sense Cs data transmission, and may
  transmit DATA to some other host
• This DATA will interfere at C
• This situation unlikely if DATA transmitted at
  highest power level
• Interference range sensing range

Data sensed
B
C
D
A
Data
RTS
Ack
Interference range
93

• Transmitting RTS at the highest power level also
  reduces spatial reuse
• Nodes receiving RTS/CTS have to defer
  transmissions

94
Bridge Functions

• Speed conversion between different devices,
  results in buffering.
• Frame format adaptation between different
  incompatible LANs
• Adding or deleting fields in the frame to convert
  between different LAN standards

95
Wireless Capacity

• Wireless channel is inefficient due to
• MAC backoff procedure
• RTS/CTS mechanism
• Frequency interference.
• Possible solutions
• Use better backoff mechanisms.
• Exploit more physical resources more spectrum
  Cell mechanism
• Exploit diversity, use different frequencies.
• Parallel control with data

96
Improve Spatial ReusePower/Rate Control
A
B
C
D
97
Exploit Infrastructure

• Infrastructure provides a tunnel to forward
  packets

infrastructure
BS1
BS2
B
C
D
E
A
Z
Ad hoc connectivity
X
98
Exploit Antennas

• Diversity antenna
• Steered beam directional antenna

99
Path Diversity

• Multiple paths to a destination
• ? Multiple next-hops to a destination

100
Inefficiency of IEEE 802.11

• Backoff interval should be chosen appropriately
  for efficiency
• Backoff interval with 802.11 far from optimum
• Ms. Khalaj thesis

101
Proposed Method

• The current method used in DCF seems to lead to
  high jitter and wasted bandwidth
• When CW is reset to its minimum after a large
  value, the next packet delays will be too low in
  compare with delays before CW size reduction
• Collision gt Network busy
• Transmission gt low load.
• These rapid changes in CW size cause high
  variations in delay or jitter
• In real conditions these assumptions are not
  always true. A packet being transmitted
  successfully does not necessarily mean the
  network is not congested
• Unfair access to the medium.
• Hence resetting CW to CWmin may cause more
  collisions and lead to wasting bandwidth

102
Proposed Method (Cont.)

• We attempt to know how much reduction in CW will
  give better performance
• Scheme 1 is the method used in DCF, resetting CW
  to its minimum size
• After a successful transmission
• In scheme 2, CW will be set to CWmin (CWcurrent
  - CWmin) / 4
• In scheme 3, CW will be set to CWmin (CWcurrent
  - CWmin) / 2
• In scheme 4, CW will be set to CWmin
  3(CWcurrent - CWmin) / 4
• By comparing the results of these schemes we can
  see how reduction of CW size will influence the
  performance

103
Simulation Model

• We used NS-2 (Network Simulator-2) for simulation
  
• Three types of traffics were generated in our
  simulation audio, video and data
• Audio traffics have the highest priority and data
  traffics the lowest
• All of the stations are in direct access range of
  each other
• All stations send their flows to a common
  receiver
• We have considered throughput, delay, and jitter
  to evaluate the performance of different schemes

104
The parameters of our simulation
Audio Video Data
CWmin 7 15 31
CWmax 255 511 1024
IFS 50us 70us 90us
Packet Size 160 bytes 1280 bytes 1500 bytes
Packet Interval 20 ms 10 ms 12.5 ms
Flow Rate 8 KBps 128 KBps 120 KBps
105
Results

• Throughput, delay and jitter of audio traffic

106
Results (Cont.)

• Throughput, delay and jitter of video traffic

107
Results (Cont.)

• Throughput and delay of data traffic

108
Results (Cont.)

• Throughput of three classes in scheme 1

109
Results (Cont.)

• Throughput of three classes in scheme 2

110
Results (Cont.)

• Throughput of three classes in scheme 3

111
Results (Cont.)

• Throughput of three classes in scheme 4

112
Observation

• Backoff and RTS/CTS handshake are unproductive
• Do not contribute to goodput

Unproductive
Random backoff
Data Transmission/ACK
RTS/CTS
113
Pipelining

• Two stage pipeline
• Random backoff and RTS/CTS handshake
• Data transmission and ACK
• Total pipelining Resolve contention completely
  in stage 1

114
How to pipeline?

• Use two channels
• Control Channel Random backoff and RTS/CTS
  handshake
• Data Channel Data transmission and ACK

Random backoff
RTS/CTS
Random backoff
RTS/CTS
RTS/CTS
Random backoff
Data Transmission/ACK
Data Transmission/ACK
115
Pipelining works well only if two stages are
balanced!
Control Channel
Data Transmission/ACK
Data Transmission/ACK
Data Channel
116
Difficult to keep the two stages balanced

• Length of stage 1 depends on
• Control channel bandwidth
• The random backoff duration
• The number of collisions occurred
• Length of stage 2 depends on
• Data channel bandwidth
• The data packet size

117
How much bandwidth does control channel require?

• If small, then
• RTS/CTS takes very long time.
• Collision detection is slow
• If large, then
• The portion of channel bandwidth used for
  productive data packet transmission is reduced
• Total bandwidth is fixed!

118
Difficulty with Total Pipelining

• The optimum division of channel bandwidth varies
  with contention level and data packet size
• Performance with inappropriate bandwidth division
  could be even worse than 802.11 DCF
• How to get around the issue of bandwidth division
  ?

119
Partial Pipelining

• Only partially resolve channel contention in
  stage 1
• Since no need to completely resolve contention,
  the length of stage 1 can be elastic to match the
  length of stage 2

120
Modified Two Stage Pipeline

• Stage 1 Random backoff phase 1
• Stage 2 Random backoff phase 2, RTS/CTS
  handshake and Data/ACK transmission

Backoff phase 1
Data/ACK
RTS/CTS
Backoff phase 2
Stage1
Stage2
121
How to pipeline?

• Still use two channels
• Narrow Band Busy Tone Channel
• Random backoff phase 1
• Data Channel Random backoff phase 2, RTS/CTS
  handshake and Data/ACK

Random backoff phase 1
Random backoff phase 1
Random backoff phase 1
Data/ACK
RTS/CTS
Backoff phase 2
Data/ACK
RTS/CTS
Backoff phase 2
122
Random Backoff Phase 1

• Each Station maintains a counter for random
  backoff phase 1
• The stations, which count to zero first, send a
  busy tone to claim win in stage 1
• Multiple winners are possible
• Other stations know they lost on sensing a busy
  tone

123
Gain over total pipelining?

• No packets transmitted on busy tone channel
• bandwidth can be small
• the difficulty of deciding optimum bandwidth
  division in total pipelining is avoided
• Length of stage 1 is elastic so the two stages
  can be kept balanced

124
Benefits of Partial Pipeline

• Only winners of stage 1 can contend channel in
  stage 2
• reduces the data channel contention
• reduces collision probability on the data channel

Stage 1
Stage 2
125
Sounds like HIPERLAN/1?
HIPERLAN / 1 (no pipelining)
Elimination Stage
Data Transmission
Yield Stage
Partial Pipelining
Random backoff phase 1
Random backoff phase 1
Random backoff phase 1
Data/ACK
RTS/CTS
Backoff phase 2
Data/ACK
RTS/CTS
Backoff phase 2
126
Benefits of Partial Pipeline
Because of pipelining, stages 1 and 2 proceedin
parallel. Stage 1 costs little except for a
narrow band busy tone channel
Partial Pipelining
Random backoff phase 1
Random backoff phase 1
Random backoff phase 1
Data/ACK
RTS/CTS
Backoff phase 2
Data/ACK
RTS/CTS
Backoff phase 2
127
Benefits of Partial Pipeline

• By migrating most of the backoff to busy tone
  channel,
• bandwidth cost of random backoff is reduced
• Cost of backoff Channel bandwidth backoff
  duration

Using IEEE 802.11 DSSS, the backoff duration
could be several milliseconds
Data Channel Bandwidth
Area cost of backoff
Busy Tone Channel Bandwidth
Backoff Duration
128
Results of Partial Pipelining

• Improved throughput and stability over 802.11 DCF

Partial Pipelining
802.11 DCF
129
Can we avoid usingbusy tone channel?
130
Observation

• Busy tone may not always be sensed
• Narrow-band channel for busy tone

131
Observation

• Taking this into account did not make the
  performance much worse
• Sensing probability 0 as well !
• Suggests the implicit pipelining scheme

132
Implicit Pipeline

• Stage 1 Random backoff phase 1
• Stage 2 Random backoff phase 2, RTS/CTS
  handshake and Data/ACK transmission

Backoff phase 1
Data/ACK
RTS/CTS
Backoff phase 2
Stage1
Stage2
133
Still two stages, but with single channel

• Similar to busy tone probability 0

Random backoff phase 1
Random backoff phase 1
Random backoff phase 1
Data/ACK
RTS/CTS
Backoff phase 2
Data/ACK
RTS/CTS
Backoff phase 2
134

• Stations do not know when a station counts to 0
• Effectively, all stations may count down till the
  end of phase 1 (as marked by end of pipelined
  data transmission)

Random backoff phase 1
Random backoff phase 1
Random backoff phase 1
Implicit stage 1
Data/ACK
RTS/CTS
Backoff phase 2
Data/ACK
RTS/CTS
Backoff phase 2
Channel usage
135
Backoff Phase 1

• During the random backoff phase 1, the stations
  counting down the backoff counter to zero win
  stage 1. Only the winners of stage 1 contend
  channel in stage 2
• Difference from partial pipelining
• With busy tone, only stations counting down to 0
  first win stage 1. Multiple winners are possible
  only if they count down to 0 together
• Without busy tone sensing, no way for a station
  to claim channel explicitly
• more stations can win stage 1

136
Backoff Phase 1

• Nodes can count down number of slots duration
  of on-going data transmission
• Generalize
• Ignore data packet size
• Each node reduces backoff interval by an
  arbitrary (reasonably chosen) amount at the end
  of current busy channel period

137
Implicit Pipeline(Dual-Stage)

• Choose backoff such that number of winners from
  stage 1 (entering stage 2) is non-zero but small
  at the end of each busy period
• Backoff increased aggressively (on failure to win
  phase 2, not just on collision)
• Backoff decreased faster for nodes that have been
  waiting longer

138
Implicit Pipeline

• Two stages as in Hiperlan/1, but no need to use
  busy tone

139
Average number of stationsin stage 2
140
Implicit Pipelining

• Inherites benefits of partial pipelining
• Reduces channel contention by reducing the number
  of contending stations.
• Backoff phase 1 proceeds in parallel with other
  channel activities

141
Contention Window 1
142
Implicit Pipelining

• Advantages compared with partial pipelining
• No busy tone channel is needed
• Can be applied to multi-hop ad hoc networks
• Disadvantage compared with partial pipelining
• More stations may win stage 1, which leads to
  degraded stability in large networks

143
Simulation results for Implicit Pipelining

• Obtained via modified ns-2 simulator
• Constant Bit Rate (CBR) traffic
• Channel bit rate 11 Mbps
• Active stations are always backlogged
• Various packet sizes
• Simulated both in wireless LANs and multi-hop ad
  hoc networks

144
Wireles LANs with RTS/CTS Handshake packet size
256 bytes
Normalized throughput
Implicit Pipelining
53
improvement
802.11 DCF
145
Wireless LANs with RTS/CTS Handshake packet
size 512 bytes
Normalized throughput
Implicit Pipelining
46
improvement
802.11 DCF
146
Wireless LANs with RTS/CTS Handshake packet
size 2048 bytes
Implicit Pipelining
Normalized throughput
26
improvement
802.11 DCF
147
Wireless LANs NO RTS/CTS Handshake packet size
512 bytes
Normalized throughput
Implicit Pipelining
87
improvement
802.11 DCF
148
Fairness Comparable to 802.11
Fairness Index
802.11 DCF
Implicit Pipelining
149
Fairness Comparable to 802.11
Max/Min Throughput Ratio
Implicit Pipelining
802.11 DCF
150
Simulation results for multi-hop Ad hoc networks
Throughput Ratio of implicit pipelining over
802.11
Simulated in 30 1000m1000m random networks
with 80 active stations
151
Simulation results for multi-hop Ad hoc networks
Number of collisions
802.11 DCF
Implicit Pipelining
Simulated in 30 1000m1000m random networks
with 80 active stations
152
SSCH Slotted Seeded Channel Hopping Overview

• A dynamic assignment algorithm
• divides the time into equal sized slots (e.g. 10
  ms) and switches each radio across multiple
  orthogonal channels on the boundary of slots in a
  distributed manner
• Main aspect of SSCH
• channel scheduling
• self-computation of tentative schedule
• communication of schedules
• synchronization with other nodes

153
SSCH Desired Properties

• No Logical Partition Ensure all nodes come into
  contact occasionally so that they can communicate
  their tentative schedule
• Synchronization Allow nodes that need to
  communicate to synchronize
• De-synchronization Infrequently overlap between
  nodes with no communication

154
Channel Scheduling -Self-Computation

• Each node use (channel, seed) pairs to represent
  its tentative schedule for the next slot
• Seed 1 , number of channels -1 Initialized
  randomly
• Focus on the simple case of using one pair
• Update rule
• new channel (old channel seed)
  mod (number of channels)

1
0
2
1
0
2
1
0
A Seed 2
0
1
2
0
1
2
0
1
B Seed 1
Example 3 channels, 2 seeds
155
Channel Scheduling Logical Partition

• Are nodes guaranteed to overlap?
• same init channel, same seed (always overlap)
• same init channel, different seeds (overlap
  occasionally)
• different init channels, different seeds (overlap
  occasionally)
• Special case Nodes may never overlap if they
  have the same seed but different channels

156
Channel Scheduling Solution to Logical Partition

• Parity slot
• every (number of channels) slots, add a parity
  slot
• in parity slot, the channel number is the seed
• do not allow the seed to change until the parity
  slot

Parity Slot
Parity Slot
157
Channel Scheduling -Communication of Schedules

• Each node broadcasts its tentative schedule
  (represented by the pair) once per slot

158
Channel Scheduling - Synchronization

• If node B needs to send data to node A, it
  adjusts its (channel, seed) pair to be the same
  as A.

Seed
A
Sync starts upon the parity slot
Flow starts
B
Seed
159
Channel Scheduling Channel Congestion

• It is likely various nodes will converge to the
  same (channel, seed) pair and communicate
  infrequently after that.

(1,2)
(1,2)
(1,2)
(1,2)
(1,2)
160
Channel Scheduling Solution to Channel
Congestion

• De-synchronization
• To identify channel congestion compare the
  number of the synchronized nodes and the number
  of the nodes sending data. De-synchronize when
  the ratio gt 2
• To de-synchronize, simply choose a new (channel,
  seed) pair for each synchronized and non-sending
  nodes

161
Channel Scheduling Synchronizing with Multiple
Nodes

• Examples
• a sender with multiple receivers
• a forwarding node in a multi-hop network
• Solution Use multiple seeds per node
• use one seed to synchronize with one node
• add a parity slot every cycle ( number of
  channels number of seeds) the channel number
  of the parity slot is the first seed.

Green slots are generated by seed 1 Yellow
slots are generated by seed 2
2
2
1
0
1
1
0
2
2
1
0
0
1
162
Channel Scheduling Partial Synchronization
Seed
A
Flow starts
B
Seed
Partial Sync Sync the second seed only
163
Evaluations of SSCH

• Simulate in QualNet
• 802.11a, 54Mbps, (used) 13 orthogonal channels
• Slot switch time 80 µs
• 4 seeds per node, slot duration 10 ms
• UDP flows CBR flows of 512 bytes sent every 50
  µs (enough to saturate the channel)

164
Evaluation Throughput (UDP)
165
Evaluation Multi-hop Mobile Networks
166
The Problem ?

• Situation
• The total number of hotspot users around the
  world is expected to to 30 million by the end of
  2004 according to researcher Gartner.
• Given the explosive growth in hotspot wireless
  usage, enhancing capacity of 802.11-based hotspot
  wireless networks is an important problem.

167
The ECHOS Solution ?

• AP CST algorithm
• Dynamically adjusts the CST in order to allow
  more flows to co-exist in the same channel in
  current 802.11 architectures.
• RNC SC algorithm
• Allows each cell or AP access to all available
  channels.
• RNC algorithm executes in a centralized radio
  network controller
• Uses one channel as primary the other two as
  secondary channels
• Allows to improve Hotspot performance beyond
  AP-CST.

168
Abilities of the Algorithms

• Dynamically allocate channels to stations
• Flexibly adopts parameters such as CST and/or
  transmit power
• THE CLAIM !
• Performance of 802.11-based hotspots can be
  improved by both these algorithms by up to 195
  per-cell and 70 overall.

169
Related Work

• Different techniques for parallelism in 802.11 to
  form ad-hoc networks
• Involve either modifications to the 802.11 MAC
  protocol or using out-of-band tones and thus,
  cannot be used to enhance the performance of the
  hundreds of millions of already deployed 802.11
  cards and access points.
• Very recently, they have discovered that varying
  CST can help boost performance.
• Use of multiple channels for throughput
  enhancement
• has been proposed for ad hoc multi-hop wireless
  networks,
• BUT
• these solutions rely on each node making
  decisions based on its locally perceived medium
  characteristics and there is little scope for
  centralized coordination.

170
Related Work Cont

• Centralized coordination of APs in PCF mode by
  allocating channels time slots to APs
• Through graph coloring centralized scheduling
• However these most of the work in this area
  assume that each AP is capable of using only a
  single channel at a given time.

171
Observations on Carrier Sensing in 802.11

• Qualnet simulator
• transmission at 2Mbps
• with a CST of -93dBm
• transmit power of 15dBm
• How to calculate the ranges?

172
Range Calculation

• Suppose T T are two transmitters at distance
  dt di from the receiver.
• T is the interferer to the transmission from T.
• Then,
• SNR at the receiver is assuming that
• both the transmitters transmit with the same
  power
• Strength of the received signal falls off as
• Where,
• K is a suitable constant
• is the transmission power
• d is the distance from the signal source
• For successful reception, the requirement is that
  the SNR be above a threshold
• This yields the requirement

Range
173
Observation 1
How to chose the optimum value of CST ? - Dynamic
174
Observation 2
The value of CST needed at T to sense the carrier
of any interfering source I at or within a
distance 2.78 d from T is given by
Where, Po transmit power
SST,R Received signal strength from T at R
Is the optimum carrier sense threshold required
at T
175
Architecture Algorithms

• Principle
• Dynamically identify flows that can coexist
• Allow them to coexist by setting optimum CST
  values (Observation 2)

Note All clients report load signal conditions
to AP in RNC-SC the AP reports to the RNC
176
Algorithm AP - CST
Basic idea If, by reducing the CST, we can allow
additional flows to operated without causing
interference beyond the available tolerance of
existing flows, we have improved the performance
for those flows.
Else CSTAPinfinity For each station s
CSTAP min(CSTAP,SSAP,s/alpha epsilon)

• The only issue is of distinguishing between
  inside and outside cell transmissions.
• This is solved by identifying the data frame RTS
  or CTS

177
Algorithm RNC - SC

• Algorithm RNC-SC has two main steps
• Determine if a cell is overloaded.
• Choose and switch a client to a secondary
  channel in overloaded cell, if possible.
• Measuring load and overload
• MAC service time (i.e., the time between the
  instant a frame is submitted to the MAC for
  transmission and the time instant the ACK is
  received) seen by a node as the measure of load.
• This value is smoothed using an exponential
  filter and averaged over all members of the cell.
• Where,
• is the threshold providing hystersis

178
Choosing Client Secondary Channel

• To create a secondary cell that has no impact on
  the primary cell, we need
• any AP/station of the secondary cell should not
  interfere with the primary
• the throughput in primary should not decrease
  because of this change.

179
Get Client Algorithm

• Has three main steps
• Compute maximum tolerated interference on each
  secondary channel k
• Reduce the transmit powers of secondary AP and
  clients on each
  secondary channel k
• Choose the client,channel pair such that the
  client observes minimum interference from outside
  the cell on that channel

180
Performance Evaluation

• Topology
• 1000 1000m divided into 4 cells.
• Each AP covers 250m approx. 11Mbps
• Each simulation run lasted 100sec results
  averaged over 10 runs
• Homogeneous user/load distribution
• Each cell has 15 clients, half the max no. of
  clients allowed in current practice
• Each client station has an HTTP
• client with a think time of 1sec
• Each cell has an FTP client
• Both upstream downstream
• traffic present

181
HIPERLAN

• 1995 ETSI technical group RES 10 (Radio Equipment
  and Systems) developed HIPERLAN/1 wireless LAN
  standards using 5 channels in 5.15-5.3 GHz
  frequency range
• Technical group BRAN (Broadband Radio Access
  Network) is standardizing HIPERLAN/2 for wireless
  ATM
• ETSI URL for Hiperlan information
  http//www.etsi.org/frameset/home.htm?
  /technicalactiv/Hiperlan/hiperlan2.htm

182
HIPERLAN Characteristics

• HIPERLANs with same radio frequencies might
  overlap
• Stations have unique node identifiers (NID)
• Stations belonging to same HIPERLAN share a
  common HIPERLAN identifier (HID)
• Stations of different HIPERLANs using same
  frequencies cause interference and reduce data
  transmission capacity of each HIPERLAN
• Packets with different HIDs are rejected to avoid
  confusion of data

183
HIPERLAN Protocol Layers

• Data link layer logical link control (LLC) sub
  layer MAC sub layer channel access control
  (CAC) sub layer

network
LLC
data link
MAC
physical
CAC
184
HIPERLAN Protocol Layers, Cont..

• MAC sub layer
• Keeps track of HIPERLAN addresses (HID NID) in
  overlapping HIPERLANs
• Provides lookup service between network names and
  HIDs
• Converts IEEE-style MAC addresses to HIPERLAN
  addresses
• Provides encryption of data for security

185
HIPERLAN Protocol Layers, Cont..

• MAC sub layer
• Provides multi hop routing certain stations
  can perform store-and-forwarding of frames
• Recognizes user priority indication (for
  time-sensitive frames)

186
HIPERLAN Protocol Layers, Cont..

• CAC sub layer
• Non-preemptive priority multiple access (NPMA)
  gives high priority traffic preference over low
  priority
• Stations gain access to channel through channel
  access cycles consisting of 3 phases

187
HIPERLAN CAC Protocol

• CAC sub layer

Cycle
Prioritization Phase
Transmission Phase
Contention Phase
1
2
3
4