Networking and protocols for real-time signal transmissions

Page 1 Hans Peter SchwefelSIPCom9-3: RT Networking Lecture 1, Fall05

Networking and protocols for real-time signal transmissionsby Hans-Peter Schwefel

• Mm1 Introduction & simple performance models

• Mm2 Real-time Support in Wireless Technologies

• Mm3 Transport Layer Aspects and Header Compression

• Mm4 IP Quality of Service: Advanced Concepts

• Mm5 Session Signalling and Application Layer aspects

[email protected]

http://www.kom.auc.dk/~hps


Content1. Motivation & Background

• Real-Time applications• Parameters

2. Layering Models Revisited• Layers and their real-time relevant functionalities• Packet-switched (IP) vs. circuit switched transmission• ’compromise’: ATM

3. L2 QoS Provisioning: Ethernet• Traffic classes: ATM, UMTS, 802.1q• Queueing/scheduling

4. Introduction to Performance Modeling• Continuous-time Markov chains• Birth-Death Processes, M/M/1 type Queues• Circuit-Switched case: Erlang formula• Packet-based traffic models

5. Summary and Outlook


Real-time requirements: Parameters• User Plane QoS/Network Performance

– End-2-End Packet Delay (in particular interactive applications)

– Delay Jitter – Packet Loss– Throughput/Goodput

• Application Level QoS– e.g. Video/Voice Quality (depending on codecs)

• Signalling Plane– Call Setup Delays– Fraction of blocked Calls

• Reliability Aspects– Failure probablities of entities– Downtime distribution

• Behavior at Handover– Dropped Calls– Delayed / Lost packets

MM1-4

See other lectures (Wireless Networks II and III)

MM5


Extended layered communication model• Ultimate goal of RT service

provisioning: user satisfaction• Focus in this course:

network and application aspects, i.e. L2-5 and application layer

Relevant functionalities:• PHY Layer

– Bit/Symbol transmission Throughput

– Symbol error probabilities (channel conditions, interference)

– Propagation delays

L3: Network Layer: IP

L2: MAC/LLC

L4: Transport: TCP, UDP, RTP/UDP

Application

(L5) Session Control, e.g. SIP

Middleware

User Interface

User

L1: PHYS

User Environment

Netw

ork QoS

Application Q

oS

User perceived Q

oS


Relevant Functionalities (cntd.)• Link Layer (L2)

– Medium Access Delays– Collisions/unsuccessful transmissions– Fragmentation– Forward error correction (FEC) and error detection (CRC)– Link-layer Retransmission Mechanisms (ARQ)– L2 scheduling, switching, buffering

• Network Layer (L3)– Path selection (routing)– Processing delays (e.g. for routing table lookup)– L3 buffering, scheduling, buffer management (RED)– [L3 fragmentation]


Relevant Functionalities (cntd.)• Transport Layer (L4)

– Multiplexing/de-multiplexing (UDP/TCP)– Error detection/checksums– In-order delivery, sequence numbers (TCP)– Acknowledgements and Retransmissions (TCP)– Flow/Congestion Control (TCP)

• Application Layer/Codecs– FEC/CRC– Application Layer Retransmissions– Application Layer sequence numbers

All Layers– Increased volume due to headers


Internet Protocol (IP)Internet Protocol IP, IPv4:• Layer 3 Protocol (Network Layer): implemented in hosts and routers• Packet (IP datagram) transmission between two hosts

(variable packet size up to 65535 bytes, often restricted by Layer 2 protocols)• Routing using 32 bit adresses (v4)

– Normally based on destination address only!

• Real-time affecting properties– Packet duplications– Packet reordering– Packet loss– fragmentation

• Real-time relevant functionalities– Scheduling in routers– Buffer management– Route selection

Application

TCP/UDP

IPLink-Layer

L5-7

L4

L3

L2


IP Version 6 (IPv6)IPv6• Basic Header 40 Bytes• 128-bit Network Addresses• Flow label (QoS)• No fragmentation in the network• ‘Built-in’ Security• Neighbor Discovery• Extension Headers:

Routing, Fragmentation, Authentication, Encryption

IPv4• Basic Header 20 Bytes• 32-bit Network Addresses• Type of Service field• Router may fragment packets• IPsec as an enhancement• ARP (Address Resolution Protocol)• Options

Version Priority Flow LabelPayload Length Type of Next Hdr. Hop Limit

Source Address

Destination Address

Offset0812162024283236

4

40

0 1 2 3

Next Header


IPv6 (cntd.)• Large number of IP addresses • Stateless autoconfiguration (can replace DHCP)• Extension headers (selection)

– 43 Routing Header– 44 Fragment Header– 51 Authentication Header– 50 Encrypted Security Payload– 60 Destination Options Header– 0 Hop-by-hop header


Packet-Based Transport• Advantages of Packet-Based Transport (as opposed to circuit switched)

– Flexibility– Optimal Use of Link Capacities, Multiplex-Gain for bursty traffic

• Drawbacks– Buffering/Queueing at routers can be necessary– Delay / Jitter / Packet Loss can occur – Overhead from Headers (20 Byte IPv4, 20 Byte TCP)

... and it makes performance modeling harder!!

Main motivation for Performance Modeling:• Network Planning• Evaluation/optimization of protocols/architectures/etc.

queueing

ν+


’Compromise’: Asynchronous Transfer Mode (ATM)• Constant size cells (48 Byte payload)• Short header (5 Byte)• Connection oriented (unidirectional): Switched Virtual Circuit (SVC), PVCs (Permanent)

allows– Call Admission Control (CAC)– Traffic Policing/Shaping

• Switching using VPI/VCI• Routing at connection setup:

– source routing – Hierarchical approach (peer groups)– PNNI

• Differentiation of Traffic Classes• Cell Format:


ATM Switching

Source: Advanced Networks lecture

Source: CNTK









ATM Traffic Classes Source: Advanced Networks lecture


Cellular Networks: UMTS Traffic Classes

Source: 3GPP TS23.107, V5.2.0

Four basic classes:


UMTS: Range of Traffic/QoS Parameters

(Source TS23.107, V5.2.0)


Link Layer: IEEE 802 standards

Selected IEEE 802 standards• 802.1 High-level interface• 802.2 Logical Link Control (LLC)• 802.3 CSMA/CD (Ethernet!)• 802.5 Token Ring• 802.11 Wireless Lan (PHY and MAC), WiFi• 802.16 Fixed Wireless Access, WiMax• 802.20 Mobile Wireless Access


Traffic Classesin 802

Defined in 802.1D (MAC layer bridge)

• 8 classes (3 bit)• Up to 8 queues for

each outbound port

Source: CNTK


Ethernet frames: Q-tagging802.1q: • extension of

MAC frame• 2 Byte Tag (TCI)• Contains 3 bit

user priority• And 12 bit VLAN

identifier

Source: CNTK


Provisioning of Ethernet QoS: Scheduling• Strict priority scheduling: high priority queues drained first• Number of queues allowed to be < 8• Mapping of queues described in 802.1d:

• Mapping of user priorities for internal use within bridges can be defined

Prio 0Q5

Q2

M1

M2

Forwardingengine

Prio 5

Prio 2

Prio 0

Prio 0

A

B

C

Source: CNTK









Challenges in IP networks:• Multiplexing of packets at nodes (L3)• Burstiness of IP traffic (L3-7)• Impact of Dynamic Routing (L3)• Performance impact of transport layer, in particular TCP (L4)• Wide range of applications different traffic & QoS requirements (L5-7)• Feedback: performance traffic model, e.g. for TCP traffic, adaptive applications

Challenges in Wireless Networks:• Wireless link models (channel models)• MAC & LLC modeling• RRM procedures• Mobility models• Cross layer optimization

Analysis frequently with ‚stochastic‘ models

Challenges in Packet Switched SettingHTTP

TCP

IPLink-Layer

L5-7

L4

L3

L2


Basic concepts• Probabilities

– ’Random experiment’ with set of possible results Ω– Axiomatic definition on event set V(Ω)

• 0≤Pr(A)≤1; Pr(∅)=0; Pr(A∪B)=Pr(A)+Pr(B) if A∩B=∅ [ A,B∈℘(Ω) ]– Conditional probabilities: Pr(A|B)=Pr(A∩B) / Pr(B)

• Random Variables (RV)– Definition: X: Ω→ú; Pr(X=x)=Pr(X-1(x))– Probability density function f(x), cumulative distribution function F(x)=Pr(X≤x),

reliability function (complementary distr. Function) R(x)=1-F(x)=Pr(X>x)

– Expected value, moments: E(Xn)=∫ xn f(x) dx– Relevant Examples, e.g.:

• number of packets that arrive at the access router in the next hour (discrete)• Buffer occupancy (#packets) in switch x at time y (discrete)• Number of downloads (’mouse clicks’) in the next web session (discrete)• Time until arrival of the next IP packet at a base station (continuous)


Basic concepts: Exponential Distributions

Important Case: Exponentially distributed RV• Single parameter: rate λ• Density function f(x)= λ exp(- λx), x>0• Cdf: F(x)=1-exp(- λx), Reliability function: R(x)=exp(- λx)• Moments: EX=1/ λ; VarX=1/ λ2, C2 = VarX / [EX]2 = 1

Important properties:• Memory-less: Pr(X>x+y | X>x) = exp(- λy) • Properties of two independent exponential RV: X with rate λ, Y with rate µ

– Distribution of min(X,Y): exponential with rate (λ+µ)– Pr(X<Y)= λ/(λ+µ)


Basic concepts III: Stochastic Processes• Definition of process (Xi) (discrete) or (Nt) (continuous)

– Simplest type: Xi independent and identically distributed (iid)• Relevant Examples:

– Inter-arrival time process: Xi– Counting Process:

N(t) = maxn | Σi=1 Xi ≤t , alternatively Ni(∆) = N(i∆) – N([i-1]∆)

Important Example: Poisson Process• Assume i.i.d. exponential packet inter-arrival times (rate λ): Xi:=Ti-Ti-1• Counting Process: Number of packets Nt until time t

– Pr(Nt=n)= (λt)n exp(- λt) / n!• Properties:

– Merging: arrivals from two independent Poisson processes with rate λ1 and λ2 Poisson process with rate (λ1+ λ2)

– Thinning: arrivals from a Poisson process of rate λ are discarded independently with probability p Poisson process with rate (1-p) λ

– Central Limit Theorem: superposition of n independent processes results in the limit n→∞in a Poisson process (under some conditions on the processes)

n-1


• Defined by– State-Space: finite or countable infinite, w/o.l.g. E=0,1,2,...,K (K=∞ also allowed)– Transition rates: µjk

• Holding time in state j: exponential with rate Σk≠j µjk =: µj

• Transition probability from state j to k: µjk / Σl≠j µjl = µjk / µj

• Xt = RV indicating the current state at time t; πi(t):=Pr(Xt=i)• ’Markov Property’: transitions do not depend on history but only on current state

t0<t1<...<tn , ∀i0,i1,...in ,j ∈E

• Computation of steady-state probabilities– Chapman Kolmogorov Equations: dπi(t)/dt = - µi πi(t) + Σj≠i µji πj(t) – Flow-balance equations, steady-state (t ∞): µi πi = Σj≠i µji πj

• Here: restriction to irreducible, homogeneous processes

Continuous Time Markov Processes


Queueing Models: Kendall NotationX/Y/C[/B] Queues (example: M/M/1, GI/M/2/10, M/M/10/10, ...)• X: Specifies Arrival Process

– M=Markovian Poisson– GI=General Independent iid

• Y: Specifies Service Process (M,G(I),...)• C: Number of Servers• B: size of finite waiting room (buffer)

[also counting the packet in service]– If not specified: B=∞

• Often also specified: service discipline– FIFO: First-In-First-Out (default)– Processor Sharing: PS– Last-in-first-out LIFO (preemptive or non-preemptive)– Earliest Deadline First (EDF), etc.

Finite buffer (size B)

µλ

Scope here: Point-process models as opposed to fluid-flow queues


M/M/1 queue• Poisson arrival of packets (first ’M’ Markovian) with rate λ• Exponentially distributed service times of rate µ (second ’M’)• Single Server (1)• FIFO service discipline

Qt = Number of packets in system is continuous-time Markov Process

’Derived’ Parameter:• Utilization, ρ= λ/ µ : if ρ≥1, instable case (no steady-state q.l.d)

Performance Parameters• Queue-length distribution: π(t) , steady-state limit: π=lim π(t) (if ρ<1)• Queue-length that an arriving customer sees• Waiting/System time distribution• Buffer-Overflow Probability for level B = Pr(arriving customers sees buffer occupancy B or

higher)

Infinite buffer

µλ


M/M/1 queue: Performance

• Birth-Death Process– Probability of i packets in queue [using flow-balance equations]

πi := Pr(Q=i) = (1-ρ)* ρi , where ρ= λ/ µ <1– Probability of idle server: π0 = (1-ρ)– Average Queue-length: EQ= ρ/(1-ρ)– Average Delay (System Time): ES= EQ/ λ = 1/(µ-λ)– Buffer Overflow Probabilities (PASTA principle)

Pr(Q(a)≥B)= Pr(Q≥B) = ρB

λ λ λ

µµ µ


General Birth-Death Processes

• Steady-State Probabilities (from balance equations):πi := Pr(Q=i) = π0 ∏k=0

i-1 λk / ∏ k=1i µk

• Models in this class, e.g.– M/M/1/B– M/M/C, M/M/C/C– Load-dependent services, discouraged arrivals

λ0 λ1 λ2

µ2µ1 µ3


Packet-based link model: M/M/1/K queue• Assumptions

– Poisson arrival of packets with rate λ– Exponentially distributed service times of rate µ– Single Server– Finite waiting room (buffer) for K packets

• Suitable e.g. for modeling ’bottleneck’ link in packet-based wireless networks

– [Full network models: see traffic analysis lecture]

Finite buffer (size K)

µλ


M/M/1/K queue: Performance

• From Birth-Death Process Theory:– Probability of i packets in queue

πi := Pr(Q=i) = (1-ρ)/(1- ρK+1) * ρi , where ρ= λ/ µ ≠1, i=0,…,K

– Probability of packet loss:p(loss) = πK = (1-ρ)/(1- ρK+1) * ρK

– Average Delay:

Ď = 1/[λ (1-pK)] * ρ/(1- ρK+1) * [(1- ρK)/ (1-ρ) – K ρK ]

λ λ λ λ

µµ µ µ

K


Extension: Models for packet traffic

• Poisson assumption for packet arrivals may be applicable for highly aggregated traffic (core networks), but otherwise traffic tends to be bursty– High data rates in ftp download but less activity between downloads– http: activities after mouse-clicks– Video streaming: high data rates in frame transmissions– Interactive Voice: talk and silent periods

• Model Modifications:– Bulk Arrival processes– ON/OFF models– Hierarchical models


The circuit switched scenario

• K channels• Users allocate one channel per call for certain call duration• If all channels are allocated additional starting calls are blocked• How many channels are necessary to achieve a call certain maximal

blocking probability?

Common Model Assumptions:• Calls are arriving according to a Poission Process (justified for large user

population, limit theorems for stochastic processes) with rate λ• Call durations are exponentially distributed with mean T (okay for voice

calls)


Computation of blocking probabilities:M/M/K/K model, Erlang-B formula

• Finite Birth-Death Process:– Probability of i calls active

πi := Pr(n=i) = π0 (λT)i /i! , i=1,…,K

where π0 = 1/[Σ(λT)i /i!] (sum taken over i=0 to K)– Probability of blocked call:

p(Blocking) = πK = π0 (λT)K /K![also known as Erlang-B formula]

λ λ λ λ

2/T1/T 3/T K/T

K









Outlook: Bulk Arrival Models

• Queue-length at arrival instances increases not only by 1, but by a Random Variable B, the bulk-size

• Parameter set of model– Bulk arrival process, e.g. exponential with rate λ– Bulk-Size distribution: pi (e.g. geometric)– Service rate (single packet)

• Steady-state solution for mean system time [Chaudhry & Templeton 83]:ES = [ EB+EB2 ] / [2 EB µ (1-ρ) ]

• Example: M(B)/M/1 queue with geometrically distributed B


Outlook: ON/OFF Models

Parameters:• N sources, each average rate κ

• During ON periods: peak-rate λp

• Mean duration of ON and OFF times

κ = λp ON/(ON+OFF)

’bursty’ traffic, when λp >> κ


Outlook: General hierarchical models• Frequently used: Several levels with increasing granularity

– E.g. 3 levels: sessions, connections, packets– Or: 5-level model:


Example: HTTP traffic model• ‘Main’ objects contain zero or more embedded objects that the browser retrieves

Correlated requests for embedded objects within retrieval of main object

HTTP Session (User A)

HTTP Session (User B)

HTTP Session (User C)...

Session Level

Download Phase 1 Download Phase 2 Dld. Phase 3 ...Idle timeRead time

’exit browser’’click’ ’click’

Dld. Phase K

’click’’start browser’

Get Main Object Get embedded Obj. 1 Get emb. Obj. 2 Get emb. Obj. N...

Connection/ Flow Level

Packet Level, TCP dynamics (not shown here)

• Statistics:– Session arrivals: Renewal process (Poisson)– Idle time: heavy-tail

– # embedded objects: geometric (measurements e.g. mean 5)

– Object size: heavy-tailed


Outlook: Feedback Models, e.g. for TCP

End-to-End flow/congestion control Feedback: network behavior (congestion) ↔ ingress trafficNo separation traffic model/network model possibleNew traffic/performance models required

Approaches:1. Connection level models2. Sender/receiver behavior & fixpoint iteration3. Integrated models

Traffic Model

Network Model Performance

Values(Delay, Loss, etc.)

feedback


Exercises:1. Analytic Models: Traffic measurements in a GPRS radio cell result in the following traffic model: voice calls

arrive at Poisson rate 1call/min and have an average duration of 1.5 min. GPRS data sessions start at rate 1session/5min, have an average duration of 20min, and generate traffic with an averate rate of 10kb/sec using IP packets of 1500 byte size and CS-II.

a) How many time-slots would have to be reserved for GSM voice calls to keep the call blocking probability below 1e-6?

b) Compute the average RLC block delay, if 4 GPRS time-slots are used for the data traffic (as simplification: use an M/M/1 queue on RLC layer, RLC block size for CS-II is 247 bits, TDMA frame duration is 4.615ms; neglect header overhead as well as the overhead of TBF assignments).

2. Priority Queues: Assume a single Ethernet output port (rate mu) that supports two QoS classes. The two types of packets, High priority (HP) and low-priority (LP), arrive according to Poisson processes with rates lambda_l and lambda_h. Strict priority scheduling is used. Also, assume that an arrival of a high-priority packet pre-empts a currently ongoing service of a low-priority packet.– Draw the Markov chain for such priority model for buffer-sizes 2 packets for HP, 3 packets for LP.– (optional) Implement a MATALB function which allows to set-up the Q matrix of the Markov chain for

arbitrary lambda_I and mu, and for arbitrary buffer-size for the low-priority packets (the high priority buffer is fixed to size 2).

– (optional) Compute the steady-state probability vector and the average queue-lengths (the latter for varying lambda_h).


References• ’Quality of Service – State of the art survey’, Deliverable, Center for Network and

Service Convergence (CNTK), in progress.

Analytic Models• Cassandras, Lafortune, ’Introduction to Discrete Event Systems’,

Chapts. 7 and 8, Kluwer, 1999.Extended Material:• H. Gogl, ’Measurement and Characterization of Traffic Streams in High-Speed Wide Area

Networks’, VDI Verlag, 2001.• H.-P. Schwefel: ’Performance Analysis of Intermediate Systems Serving Aggregated ON/OFF

Traffic with Long-Range Dependent Properties’, Dissertation, TU Munich, 2000.

Related Lectures• ’Traffic analysis I and II’ (H Schiøler, HP Schwefel, 8/9th Sem DIRS/NPM)• Stochastic Models (B. Lindberg, HP Schwefel, B Fleury, 9th Sem SipCom)

Documents

Networking and protocols for real-time signal transmissions