计算机代考程序代写 Erlang assembly algorithm Real-time Conversational Applications

Real-time Conversational Applications

Advanced Network Technologies
4G LTE
School of Computer Science

Dr. | Lecturer

1

4G/5G cellular networks
the solution for wide-area mobile Internet
widespread deployment/use:
more mobile-broadband-connected devices than fixed-broadband-connected devices (2019)!
4G availability: 97% of time in Korea, 90% in US
transmission rates up to 100’s Mbps
technical standards: 3rd Generation Partnership Project (3GPP)
wwww.3gpp.org
4G: Long-Term Evolution (LTE) standard

2

4G/5G cellular networks
similarities to wired Internet
edge/core distinction, but both below to same carrier
global cellular network: a network of networks
widespread use of protocols we’ve studied: HTTP, DNS, TCP, UDP, IP, etc.
separation of data/control planes, SDN, tunneling
interconnected to wired Internet

differences from wired Internet
different wireless link layer
mobility
user “identity” (via SIM card)
subscriber identification module
business model: users subscribe to a cellular provider
“home network” versus roaming on visited nets
global access, with authentication infrastructure, and inter-carrier settlement

3

Mobile device
(UE)
Base station
(eNode-B)

Elements of 4G LTE architecture
radio access
network
all-IP Enhanced Packet Core (EPC)

Mobility Management Entity (MME)
Serving Gateway (S-GW)
Home Subscriber Service (HSS)
PDN gateway (P-GW)
to
Internet

Mobile device:
smartphone, tablet, laptop, IoT, … with 4G LTE radio
64-bit International Mobile Subscriber Identity (IMSI), stored on SIM (Subscriber Identity Module) card
LTE jargon: User Equipment (UE)

PDN: Packet Data Network

4

Elements of 4G LTE architecture

Mobility Management Entity (MME)
Serving Gateway (S-GW)
Home Subscriber Service (HSS)
PDN gateway (P-GW)

to
Internet
Mobile device
(UE)

Base station
(eNode-B)

Base station:
at “edge” of carrier’s network
manages wireless radio resources, mobile devices in its coverage area (“cell”)
coordinates device authentication with other elements
similar to WiFi AP but:
active role in user mobility
coordinates with nearly base stations to optimize radio use
LTE jargon: eNode-B

5

Elements of 4G LTE architecture

Mobility Management Entity (MME)
Serving Gateway (S-GW)
Home Subscriber Service (HSS)
PDN gateway (P-GW)

to
Internet
Mobile device
(UE)

Base station
(eNode-B)

Home Subscriber Service
stores info about mobile devices for which the HSS’s network is their “home network”
works with MME in device authentication

6

Elements of 4G LTE architecture

Mobility Management Entity (MME)
Serving Gateway (S-GW)
Home Subscriber Service (HSS)
PDN gateway (P-GW)

to
Internet
Mobile device
(UE)

Base station
(eNode-B)

Serving Gateway (S-GW), PDN Gateway (P-GW)
lie on data path from mobile to/from Internet
P-GW
gateway to mobile cellular network
Looks like other internet gateway router
provides NAT services
other routers:
extensive use of tunneling

7

Elements of 4G LTE architecture

Mobility Management Entity (MME)
Serving Gateway (S-GW)
Home Subscriber Service (HSS)
PDN gateway (P-GW)

to
Internet
Mobile device
(UE)
Base station
(eNode-B)

Mobility Management Entity
device authentication (device-to-network, network-to-device) coordinated with mobile home network HSS

mobile device management:
device handover between cells
tracking/paging device location
path (tunneling) setup from mobile device to P-GW

8

LTE: data plane control plane separation

control plane
new protocols for mobility management , security, authentication
MME
HSS
base station
P-GW
S-GW

base station
P-GW
S-GW

data plane
new protocols at link, physical layers
IP tunnels
extensive use of tunneling to facilitate mobility

9

IP
Packet Data
Convergence
Radio Link
Medium Access
Physical
LTE data plane protocol stack: first hop

data
plane
P-GW
S-GW

Application
Transport
IP
Packet Data
Convergence
Radio Link
Medium Access
Physical
Link

base station
LTE link layer protocols:
Packet Data Convergence: header compression, encryption
Radio Link Control (RLC) Protocol: fragmentation/reassembly, reliable data transfer
Medium Access: requesting, use of radio transmission slots

10

IP
Packet Data
Convergence
Radio Link
Medium Access
Physical
LTE data plane protocol stack: first hop

Application
Transport
IP
Packet Data
Convergence
Radio Link
Medium Access
Physical
Link

base station
LTE radio access network:
 downstream channel: FDM, TDM within frequency channel (OFDM – orthogonal frequency division multiplexing)
“orthogonal”: minimal interference between channels
upstream: FDM, TDM similar to OFDM
each active mobile device allocated two or more 0.5 ms time slots over 12 frequencies
scheduling algorithm not standardized – up to operator
100’s Mbps per device possible

11

IP
Packet Data
Convergence
Radio Link
Medium Access
Physical

GTP-U
UDP
IP
link
Physical

LTE data plane protocol stack: packet core

GTP-U
UDP
IP
link
Physical

GTP-U
UDP
IP
link
Physical

P-GW
S-GW

base station
\

tunneling:
mobile datagram encapsulated using GPRS Tunneling Protocol (GTP), sent inside UDP datagram to S-GW
S-GW re-tunnels datagrams to P-GW
supporting mobility: only tunneling endpoints change when mobile user moves

GTP-U: user data
GTP-C: control

12

LTE data plane: associating with a BS

data
plane
P-GW
S-GW

base station
BS broadcasts primary synch signal every 5 ms on all frequencies
BSs from multiple carriers may be broadcasting synch signals
1

mobile finds a primary synch signal, then locates 2nd synch signal on this freq.
mobile then finds info broadcast by BS: channel bandwidth, configurations; BS’s cellular carrier info
mobile may get info from multiple base stations, multiple cellular networks
2

mobile selects which BS to associate with (e.g., preference for home carrier)
3

more steps still needed to authenticate, establish state, set up data plane
4

1

2

3

13

LTE mobiles: sleep modes

data
plane

ZZZZ…
as in WiFi, Bluetooth: LTE mobile may put radio to “sleep” to conserve battery:
light sleep: after 100’s msec of inactivity
wake up periodically (100’s msec) to check for downstream transmissions
deep sleep: after 5-10 secs of inactivity
mobile may change cells while deep sleeping – need to re-establish association

14

Global cellular network: a network of IP networks

visited mobile
carrier network

public Internet
and
inter-carrier IPX

P-GW

P-GW
home mobile
carrier network
Home
Subscriber
Server

in home network

roaming in visited network

SIM card: global identify info in home network
all IP:
carriers interconnect with each other, and public internet at exchange points
legacy 2G, 3G: not all IP, handled otherwise

home network HSS:
identify & services info, while in home network and roaming

15

Mobility manager

Mobility in 4G networks: major mobility tasks
Visited network

P-GW
Streaming server

Home
network
Internet
base station
MME

1

2

3

4

Home
Subscriber
Server

S-GW
P-GW

base station association:
covered earlier
mobile provides IMSI – identifying itself, home network

1
control-plane configuration:
MME, home HSS establish control-plane state – mobile is in visited network

2
data-plane configuration:
MME configures forwarding tunnels for mobile
visited, home network establish tunnels from home P-GW to mobile

3
mobile handover:
mobile device changes its point of attachment to visited network

4

16

Mobility manager

Mobile communicates with local MME via BS control-plane channel
MME uses mobile’s IMSI info to contact mobile’s home HSS
retrieve authentication, encryption, network service information
home HHS knows mobile now resident in visited network
BS, mobile select parameters for BS-mobile data-plane radio channel

 Configuring LTE control-plane elements
Visited network

P-GW
Home
network
base station
MME

2

Home
Subscriber
Server

S-GW
P-GW

17

Mobility manager

 Configuring data-plane tunnels for mobile

P-GW
Streaming server

Internet
base station
MME

Home
Subscriber
Server

S-GW
P-GW

Visited network
Home
network
S-GW to BS tunnel: when mobile changes base stations, simply change endpoint IP address of tunnel
S-GW to home P-GW tunnel: implementation of indirect routing

tunneling via GTP (GPRS tunneling protocol): mobile’s datagram to streaming server encapsulated using GTP inside UDP, inside datagram

18

Handover between BSs in same cellular network

P-GW

S-GW
MME

source BS

1

3

2

data path before handover

data path after
handover

1
current (source) BS selects target BS, sends Handover Request message to target BS

2
target BS pre-allocates radio time slots, responds with HR ACK with info for mobile

3
source BS informs mobile of new BS
mobile can now send via new BS – handover looks complete to mobile

4
source BS stops sending datagrams to mobile, instead forwards to new BS (who forwards to mobile over radio channel)

target BS

4

19

Handover between BSs in same cellular network

1

6

3

5

5

2

5
target BS informs MME that it is new BS for mobile
MME instructs S-GW to change tunnel endpoint to be (new) target BS

6
target BS ACKs back to source BS: handover complete, source BS can release resources
P-GW
S-GW
MME
source BS
target BS

4

7

7
mobile’s datagrams now flow through new tunnel from target BS to S-GW

20

On to 5G!
goal: 10x increase in peak bitrate, 10x decrease in latency, 100x increase in traffic capacity over 4G
5G NR (new radio):
two frequency bands: FR1 (450 MHz–6 GHz) and FR2 (24 GHz–52 GHz): millimeter wave frequencies
not backwards-compatible with 4G
MIMO: multiple directional antennae
millimeter wave frequencies: much higher data rates, but over shorter distances
pico-cells: cells diameters: 10-100 m
massive, dense deployment of new base stations required

21

Advanced Network Technologies
Mobile Network Analysis, Erlang B, Erlang C
School of Computer Science
Dr. | Lecturer

22

λ
λ
λ
λ
λ


(S −1)μ Sμ

λ λ
λ λ

… …
M/M/1
M/M/m/m


μ
μ
μ
μ
μ
(m=S)
Review of M/M/1 and M/M/m/m Queue

23

M/M/m/m Queue Model
The cell can serve S users (S channels)
User arrival follows a Poisson process (arrival rate λ)
User will use the channel for t, where t follows exponential distribution (mean 1/μ)
t is called “channel holding time”
New arrivals will be dropped if all channels are occupied.

λ
λ
λ
λ
λ


(S −1)μ Sμ

M/M/m/m


μ
(m=S)
M/M/m/m Queue Model
Q: What is the probability if a new arrival is dropped?

PS

Poisson arrival sees time average.

25


S n=0
(λ / μ)n / n!
(λ / μ)S / S!
Erlang B Formula
Erlang B formula
PS =
Erlang B formula, or Erlang loss formula, the formula for the blocking probability
PS is also called grade of service (GOS).

Erlang B formula for system design
Q1 (Performance Evaluation): Given traffic load (λ / μ), number of channels (S), calculate blocking probability PS
A: Directly apply Erlang B formula

Q2 (Traffic Shaping) : Given blocking probability, number of channels, calculate max traffic load
A: Solve the value of λ / μ by Erlang B formula; using Erlang table/chart.

Q3 (Channel Reservation) : Given blocking probability, traffic load, calculate the number of channels needed
A: Solve the value of S by Erlang B formula; using Erlang table/chart.

Erlang: a dimensionless unit used as a measure of offered load

The average number of concurrent calls measured over a time unit.

https://www.callcentrehelper.com/what-are-erlangs-154459.htm
Erlang

28

Erlang: In our model

λ: average number of arrivals per unit time.

1/μ: average channel holding time.

Traffic load is λ/μ Erlang.

Erlang

29

Traffice Intensity in Erlangs
Number of Channels
Erlang B chart
log-log chart

Assume there are 10 channels.
Assume each user uses the channel for 6 minutes (0.1 hour).
What is the arrival rate can be supported for 0.5% blocking probability?

Example A

Traffice Intensity in Erlangs
Number of Channels
log-log chart

λ/μ=4
Example A

λ/μ=4
and 1/μ = 0.1 hour
λ=40 units/hour

Example A

Consider a base station
An average call lasts two minutes
1500 calls per hour on average
the probability of blocking is to be no more than 1%.
How many channels do we need?

λ=1500 units/hour
1/μ = 1/30 hour
λ /μ = 50 Erlang

Example B

Traffice Intensity in Erlangs
Number of Channels
Example B

Between 60 channels and 70 channels;
60 is not enough; 70 is the number of channels needed.
(More accurate calculation is 64)

Queue Model with Handover
The cell can serve S users (S channels)
Intra-cell new arrival follows a Poisson process (arrival rate λn)
Inter-cell handover arrival follows a Poisson process (arrival rate λh)
User will use the channel for t, where t follows exponential distribution (mean 1/μ)
S channels.

How to block/drop calls?

Block intra-cell new arrival. Penalty
Drop Inter-cell handover arrival. Higher penalty
Guard channel approach!
S-L guarded channels
[0,L-1] active users, new arrivals and handover arrivals are accepted.
[L, S-1] active users, new arrivals are blocked, handover arrivals are accepted.
S or more active users. new arrivals are blocked, handover arrivals are dropped.

Guard channel approach

λn + λh


(S −1)μ Sμ

λn + λh
…λn + λh
…λh
λh

S
μ
All arrivals are accepted
Only handover arrivals are accepted
None is accepted
New arrival blocking probability Pb=PL +PL+1 +…+PS
Handover dropping probability Pd= PS


L n=0
((λn + λh ) / μ)i / i!
Pi =
((λn + λh ) / μ)n / n!
+∑
((λn + λh ) / μ)L (λh / μ)n-L / n!
S n=L+1
0≤ i ≤ L
Calculate the L< i ≤ S case by yourself 38 System design problem Q1 (Performance Evaluation): Given traffic load (λn , λh , and μ), number of channels (L and S), calculate probabilities (Pb and Pd ) Q2 (Traffic Shaping) : Given required probabilities (Pb and Pd ) number of channels (L and S), calculate the traffic load Q3 (Channel Reservation) : Given required probabilities (Pb and Pd ) , traffic load, calculate the number of channels needed S is also given, calculate the optimal L 40 Erlang C An arriving unit will need to wait if all servers are busy (it is not blocked) Examples: Call centers M/M/m queue λ λ λ λ λ 2μ kμ (S −1)μ Sμ 3μ M/M/m … … μ (m=S) S+1 S+2 … Sμ Sμ λ λ λ Sμ 41 Erlang C Pw: The probability that a new arrival has to wait (cannot be served immediately). λ λ λ λ λ 2μ kμ (S −1)μ Sμ 3μ M/M/m … … μ (m=S) S+1 S+2 … Sμ Sμ λ λ λ Sμ Pw = ∑ ∞ n=S Pn 42 Erlang C Pw: The probability that a new arrival has to wait (cannot be served immediately). Erlang C formula Pw = ∑ ∞ n=S Pn Pw = ∑ S-1 n=0 (λ / μ)n (λ / μ)S S! S S- λ / μ n! (λ / μ)S S! S S- λ / μ + Pw is grade of service (GOS). 43 Erlang C Chart Advanced Network Technologies Mobile Network Analysis, Bit Error Detection and Correction School of Computer Science Dr. | Lecturer 45 Bit Error Detection and Recovery EDC= Error Detection and Correction bits (redundancy) D = Data protected by error checking, may include header fields Error detection not 100% reliable! protocol may miss some errors, but rarely larger EDC field yields better detection and correction otherwise f(D)=EDC f(D’)=EDC’? 46 Modulo 2 operation Binary, modulo 2 domain addition, subtraction 1+1=0, 1-1=0, 1+0=1, 1-0=1 0+1=1, 0-1=1 0+0=0, 0-0=0 “-”, “+”, are equivalent to, XOR,  11+11=00: no carry over Multiplication 11*100=11*22=1100 (left shift 2 bits) 11*11=11*10+11*1=110+11=101 47 Parity Checking Single Bit Parity: Detect single bit errors 110100 1 d data bits parity bit odd number of ‘1’s -> 1
even number of ‘1’s -> 0
total number of ‘1’s -> even

48

Bit Check in Matrix format
Parity Checking
dataword
generator matrix

codeword
identity submatrix
parity submatrix

49

Bit Check in Matrix format
Linear Block Code: Generalized Parity Check
dataword
generator matrix
codeword
identity submatrix
parity submatrix
dG=c
1*k vector
k*n matrix
1*n vector

50

Linear Block Code: Generalized Parity Check
k data bits
n-k parity bits
code rate: k/n

51

Linear Block Code: Decoding

paritycheck matrix

Transpose of parity matrix
Identity submatrix
H
GHT=0

52

Linear Block Code: Decoding

cHT=dGHT =0

Not 0? Error detected.

53

/docProps/thumbnail.jpeg