Slides adapted from Romit Roy Choudhury (Duke) Wireless & Mobile Networking CS 752/852 - Spring 2011 Lec #7: Medium Access Control – V Data Rate Control Tamer Nadeem Dept. of Computer Science Slides adapted from Romit Roy Choudhury (Duke)

What is Data Rate ? Number of bits that you transmit per unit time under a fixed energy budget Too many bits/s: Each bit has little energy -> Hi BER Too few bits/s: Less BER but lower throughput

How do we choose the rate of modulation Some Basics Friss’ Equation: Received Power Bit error (p) for BPSK and QPSK : Channel Bandwidth How do we choose the rate of modulation SNR Varying with time and space Data Rate Floor Noise

802.11b – Transmission rates

Static Rates SINR time # Estimate a value of SINR # Then choose a corresponding rate that would transmit packets correctly most of the times # Failure in some cases of fading  live with it

Adaptive Rate-Control SINR time # Observe the current value of SINR # Believe that current value is indicator of near-future value # Choose corresponding rate of modulation # Observe next value # Control rate if channel conditions have changed

Is there a tradeoff ? Rate = 10 B C E A D

What about length of routes due to smaller range ? Is there a tradeoff ? Rate = 10 B C E A D Rate = 20 What about length of routes due to smaller range ?

Will carrier sense range vary with rate Any other tradeoff ? Will carrier sense range vary with rate

Carrier sensing estimates energy in the channel. Total interference Rate = 10 B C E A D Rate = 20 Carrier sensing estimates energy in the channel. Does not vary with transmission rate

Bigger Picture Rate control has variety of implications Any single MAC protocol solves part of the puzzle Important to understand e2e implications Does routing protocols get affected? Does TCP get affected? … Good to make a start at the MAC layer RBAR OAR Opportunistic Rate Control

A Rate-Adaptive MAC Protocol for Multi-Hop Wireless Networks Gavin Holland HRL Labs Nitin Vaidya Paramvir Bahl UIUC Microsoft Research MOBICOM’01 Rome, Italy © 2001. Gavin Holland

Background Current WLAN hardware supports multiple data rates 802.11b – 1 to 11 Mbps 802.11a – 6 to 54 Mbps Data rate determined by the modulation scheme

Problem Modulation schemes have different error characteristics 8 Mbps 1 Mbps BER But, SINR itself varies With Space and Time SNR (dB)

Large-scale variation with distance (Path loss) Mean Throughput (Kbps) Impact Large-scale variation with distance (Path loss) 8 Mbps Path Loss SNR (dB) Mean Throughput (Kbps) 1 Mbps Distance (m) Distance (m)

Small-scale variation with time (Fading) Impact Small-scale variation with time (Fading) Rayleigh Fading SNR (dB) 2.4 GHz 2 m/s LOS Time (ms)

Question Which modulation scheme to choose? SNR (dB) SNR (dB) 2.4 GHz 2 m/s LOS Distance (m) Time (ms)

Answer  Rate Adaptation Dynamically choose the best modulation scheme for the channel conditions Desired Result Mean Throughput (Kbps) Distance (m)

How frequently must rate adaptation occur? Design Issues How frequently must rate adaptation occur? Signal can vary rapidly depending on: carrier frequency node speed interference etc. For conventional hardware at pedestrian speeds, rate adaptation is feasible on a per-packet basis Coherence time of channel higher than transmission time

Adaptation  At Which Layer ? Cellular networks Adaptation at the physical layer Impractical for 802.11 in WLANs For WLANs, rate adaptation best handled at MAC Why? RTS/CTS requires that the rate be known in advance D C B A CTS: 8 RTS: 10 10 8 Sender Receiver

Who should select the data rate? Collision is at the receiver Channel conditions are only known at the receiver SS, interference, noise, BER, etc. The receiver is best positioned to select data rate A B

Previous Work PRNet Pursley and Wilkins Periodic broadcasts of link quality tables Pursley and Wilkins RTS/CTS feedback for power adaptation ACK/NACK feedback for rate adaptation Lucent WaveLAN “Autorate Fallback” (ARF) Uses lost ACKs as link quality indicator

Lucent WaveLAN “Autorate Fallback” (ARF) DATA 2 Mbps Effective Range 1 Mbps Sender decreases rate after N consecutive ACKS are lost Sender increases rate after Y consecutive ACKS are received or T secs have elapsed since last attempt

Slow to adapt to channel conditions Performance of ARF SNR (dB) Time (s) Dropped Packets Rate (Mbps) Time (s) Failed to Increase Rate After Fade Attempted to Increase Rate During Fade Slow to adapt to channel conditions Choice of N, Y, T may not be best for all situations

Move the rate adaptation mechanism to the receiver RBAR Approach Move the rate adaptation mechanism to the receiver Better channel quality information = better rate selection Utilize the RTS/CTS exchange to: Provide the receiver with a signal to sample (RTS) Carry feedback (data rate) to the sender (CTS)

Receiver-Based Autorate (RBAR) Protocol CTS (1) RTS (2) 2 Mbps 1 Mbps D DATA (1) RTS carries sender’s estimate of best rate CTS carries receiver’s selection of the best rate Nodes that hear RTS/CTS calculate reservation If rates differ, special subheader in DATA packet updates nodes that overheard RTS

Performance of RBAR SNR (dB) Time (s) Rate (Mbps) Time (s) Rate (Mbps) ARF Time (s)

There are two types of fading Question to the class There are two types of fading Short term fading Long term fading Under which fading is RBAR better than ARF ? Under which fading is RBAR comparable to ARF ? Think of some case when RBAR may be worse than ARF

Implementation into 802.11

Implementation into 802.11 PLCP Header

Reservation Subheader (RSH) Implementation into 802.11 Frame Control Sequence Control Duration DA SA FCS BSSID Body FCS Reservation Subheader (RSH) Encode data rate and packet length in duration field of frames Rate can be changed by receiver Length can be used to select rate Reservations are calculated using encoded rate and length New DATA frame type with Reservation Subheader (RSH) Reservation fields protected by additional frame check sequence RSH is sent at same rate as RTS/CTS New frame is only needed when receiver suggests rate change WHY

Performance Analysis 1E-5 Ns-2 with mobile ad hoc networking extensions Rayleigh fading Scenarios: single-hop, multi-hop Protocols: RBAR and ARF RBAR Channel quality prediction: SNR sample of RTS Rate selection: Threshold-based Sender estimated rate: Static (1 Mbps) BER 1E-5 2 Mbps Threshold SNR (dB) 8 Mbps Threshold

Performance Results Single-Hop Network

Mean Throughput (Kbps) Single-Hop Scenario Mean Throughput (Kbps) Distance (m) A B

Varying Node Speed UDP Performance RBAR ARF Mean Throughput (Kbps) CBR source Packet Size = 1460 Mean Node Speed (m/s) RBAR performs best Declining improvement with increase in speed Adaptation schemes over fixed RBAR over ARF Some higher fixed rates perform worse than lower fixed rates WHY?

Varying Node Speed TCP Performance RBAR ARF Mean Throughput (Kbps) FTP source Packet size = 1460 Mean Node Speed (m/s) RBAR again performs best Overall lower throughput and sharper decline than with UDP Caused by TCP’s sensitivity to packet loss More higher fixed rates perform worse than lower fixed rates

No Mobility UDP Performance RBAR CBR source Packet size = 1460 ARF Mean Throughput (Kbps) Mean Throughput (Kbps) Distance (m) Distance (m) WHY? RSH overhead seen at high data rates Can be reduced using some initial rate estimation algorithm Limitations of simple threshold-based rate selection seen Generally, still better than ARF

No Mobility UDP Performance RBAR-P CBR source Packet size = 1460 Mean Throughput (Kbps) Distance (m) RBAR-P – RBAR using a simple initial rate estimation algorithm Previous rate used as estimated rate in RTS Better high-rate performance Other initial rate estimation and rate selection algorithms are a topic of future work Why useful ?

RBAR Summary Modulation schemes have different error characteristics Significant performance improvement may be achieved by MAC-level adaptive modulation Receiver-based schemes may perform best Proposed Receiver-Based Auto-Rate (RBAR) protocol Implementation into 802.11 Future work RBAR without use of RTS/CTS RBAR based on the size of packets Routing protocols for networks with variable rate links

Questions?

Multi-rate Anomaly [Heusse03] The question is, how severe the LA scheme on network performance. In order to answer this question, we ran a simple experiments with two nodes n1 & n2 with this simple network toplogy where n1 is fixed and connected to BS and n2 is mobile one and move from left to right We first allowed n1 & n2 to run with 11Mbs (Red and Green), then when n1&n2 run with 1Mbs (Blue and pink). These two middle lines define the expected boundaries when nodes all run 11 MBs and when al nodes run 1Mbs. Now, given the heterogeneous case where n1 runs 11Mbs and n2 runs 1Mbs. We get the curve for (light blue and Yellow). Comparing this curves with the expected boundaries for the homogeneous case, we can see that Fast user throughput (n1), light blue curve is severely degraded while the slow user got higher throughput than he is exepcted. This is a well know behavior of WiFi networks first identified by Huesse in InfoCom 2003 and called this behaviour the Multi-rate Anamoly Now the question is: How serious is it? The SIGCOMM picture Three scenarios: Both nodes use 11 Mb/s Both nodes use 1 Mb/s n1 uses 11 Mb/s, n2 uses 1 Mb/s Fast user’s throughput is severely degraded whereas Slow user achieves throughput even larger than it expects

OAR: An Opportunistic Auto-Rate Media Access Protocol for Ad Hoc Networks B. Sadeghi, V. Kanodia, A. Sabharwal, E. Knightly Rice University Slides adapted from Shawn Smith

Motivation Consider the situation below ARF? RBAR? A B C

Throughput Fairness vs Temporal Fairness Motivation What if A and B are both at 56Mbps, and C is often at 2Mbps? Slowest node gets the most absolute time on channel? A B C Timeshare A B C Throughput Fairness vs Temporal Fairness

Opportunistic Scheduling Goal Exploit short-time-scale channel quality variations to increase throughput. Issue Maintaining temporal fairness (time share) of each node. Challenge Channel info available only upon transmission

Opportunistic Auto-Rate (OAR) In multihop networks, there is intrinsic diversity Exploiting this diversity can offer benefits Transmit more when channel quality great Else, free the channel quickly RBAR does not exploit this diversity It optimizes per-link throughput

OAR Idea D C A B A C Basic Idea Data Data Data Data Data Data Data If bad channel, transmit minimum number of packets If good channel, transmit as much as possible D C A B Data Data Data Data A Data Data Data Data C

Why is OAR any better ? D C A B A C 802.11 alternates between transmitters A and C Why is that bad D C A B Data Data Data Data A Data Data Data Data C Is this diagram correct ?

Why is OAR any better ? D C A B A C Bad channel reduces SINR  increases transmit time Fewer packets can be delivered D C A B Data Data A Data Data Data C Data

OAR Protocol Steps Transmitter estimates current channel Can use estimation algorithms Can use RBAR, etc. If channel better than base rate (2 Mbps) Transmit proportionally more packets E.g., if channel can support 11 Mbps, transmit (11/2 ~ 5) pkts OAR upholds temporal fairness Each node gets same duration to transmit Sacrifices throughput fairness  the network gains !!

MAC Access Delay Simulation Back to back packets in OAR decrease the average access delay Increase variance in time to access channel Why?

OAR Protocol Rates in IEEE 802.11b: 2, 5.5, and 11 Mbps Number of packets transmitted by OAR ~

Simulations Three Simulation experiments Fully connected networks: all nodes in radio range of each other Number of Nodes, channel condition, mobility, node location Asymmetric topology Random topologies Implemented OAR and RBAR in ns-2 with extension of Ricean fading model [Punnoose et al ‘00]

Fully Connected Setup Every node can communicate with everyone Each node’s traffic is at a constant rate and continuously backlogged Channel quality is varied dynamically

Fully Connected Throughput Results OAR has 42% to 56% gain over RBAR Increase in gain as number of flows increases Note that both RBAR and OAR are significantly better than standard 802.11 (230% and 398% respectively)

Asymmetric Topology Results OAR maintains time shares of IEEE 802.11 Significant gain over RBAR

OAR thoughts OAR does not offer benefits when OAR may not be suitable for applications like With TCP how can OAR get affected ?

OAR thoughts OAR does not offer benefits when Neighboring nodes do not experience diverse channel conditions Coherence time is shorter than N packets With TCP can OAR get affected ? Back-to-back packets can increase TCP performance However, bottleneck bandwidth can get congested quick Also, variance of RTT can increase

Multi-user Data Rate Adjustment Algorithm for Enterprise Networks Nazif Tas, Tamer Nadeem, Ashok Agrawala INFOCOM 2011

Data Rate Adjustment Mechanisms Previous Work Data Rate Adjustment Mechanisms Agreement between sender and the transmitter on rate to use. Sender-based: statistical. Easy, resource efficient. ARF, AARF, CARA, AMRR… Receiver-based: receiver chooses data rate and notifies the sender. More accurate, unnecessary resource usage. RBAR Multi-rate Anomaly Solutions Infrastructure-based solutions AP selection, traffic scheduling, etc [Wang, Ji, Im04, …] MAC Layer modifications Packet bursting, minimum CW adjustment, etc [Mai09, Yah-hong07, …] Higher Layer adjustments TCP window adjustment, traffic slow down [Yoo08, Kashibuchi09, …] Requires holistic view of the system Requires modifications in the standards Time fairness? OAR, Cwmin adjustment, scheduling etc. But, they propose to change the standard or assume global view of the system NOT realistic. Can we do it without modification and holistic view. This requires a good analysis of Multi-rate anomaly Requires non-seamless cross layer support Can we lessen the effects of Multi-Rate anomaly efficiently in a distributed & seamless manner with no modifications in the standards?

Binary Exponential Backoff (BEB) IEEE 802.11 BEB Binary Exponential Backoff (BEB) IEEE 802.11 DCF CSMA/CA: carrier sensing Binary Exponential Backoff

Baseline Fairness Criteria : throughput seen by data rate d users in a network with ns number of slow users and nf number of fast users. [Tan04] Multi-rate Anomaly Why is multi-rate anomaly bad? Baseline Fairness Criteria Fast Slow

Bianchi Model with Retry Limit Can we use arbitrary limits to mitigate Multi-rate Anomaly? k: retry limit Our extension supports arbitrary retry limits per user groups For data rate di,

Multi-user Data Rate Adjustment Algorithm (MORAL) Observe Automatically adjust the retry limits of each user such that Improve fairness: Take the fair throughput share without hurting others. Distributed computation: No centralized decision making. Standard Coherence: No packet modification, restructuring or protocol alteration. LA MAC Decide Act Two step operation: Collect information about the overall network. Adjust the retry limit accordingly “The first thing to note is that [happiness] is a relative quality. We experience it differently according to our circumstances. What makes one person glad may be a source of suffering to another”

MORAL Details: Heuristic Principles After each transmission cycle, we have two pieces of information: ccurrent: number of transmissions this cycle for the current node Successful transmission (ccurrent =1) Failed transmission(ccurrent =0) ccalculated: Fair number of transmissions for the current node

Experiments I 60% 61.3% 1 Mb/s 11 Mb/s MORAL 11, 1 Mb/s – 20 Users Each 60% Default MORAL Balanced Static 1 Mb/s 61.3% 11 Mb/s

Experiments II 43.0% 31.8% 11, 5.5, 2, 1 Mb/s– 10 Users Each Default MORAL 43.0% Balanced Static 31.8%

Experiments III 48.0% 20.8% 93.8% 30 1 Mb/s, 10 11 Mb/s Default 1 Mb/s MORAL 11 Mb/s 20.8% 93.8% 1 Mb/s

Experiments IV 40.2% 23.9% 11, 1 Mb/s – 20 Users Each

Summary MORAL is an effective MAC layer link adaptation mechanism which lessens the effects of multi-rate anomaly and promises - Better fairness - Increased throughput MORAL is Fully compatible with the current standards Totally distributed Highly adaptable Easily deployable in any IEEE 802.11 compliant devices

Other Ideas (Briefly)

Exploiting Diversity in Rate Adaptation Yet another idea exploits multiple user diversity Among many intermediate nodes, who has best channel Use that node as forwarding node Forwarding node can change with time Due to channel fluctuations at different time and space Channel Conditions SNR WLAN AP TIME USERS

MAD using Packet Concatenation (PAC) The Protocol Overview MAD using Packet Concatenation (PAC) DIFS SIFS SIFS SIFS SIFS sender GRTS DATA 0 DATA 1 DATA 2 SF user 1 CTS 1 CTS k CTS 2 … user 2 ACK 0~2 user k Since at least one intermediate node is likely to have good channel condition, transmitter can transmit at a high data rate or concatenate Multiple packets Choosing subset of neighbor-group is important Coherence time of channel must be greater than packet chain Group needs to really have independent channel gain Correlated channel gains can lead to performance hit.

Routing based on rate-control What lies ahead ? Routing based on rate-control Choosing routes that contain high-rate links ETX metric proposed from MIT accomodates link character Opportunistic routing from MIT again – takes neighbor diversity into account (best paper Sigcomm 2005) Fertile area for a project … Dual of rate-control is power control One might be better than the other Decision often depends on the scenario  open problem Directional antennas for DD link for data/ack Rate control can be introduced  Not been studied yet … many many more

Questions ?