chuong 4 day du

8/10/2019 chuong 4 day du

1/77

-Teacher:

Masters NguynThanh c-Group member:

Trn Vn Sng 11232081

Bi Cng Tun 11261741Nguyn Vn Tin 11251111Nguyn Tt Thng 11043491

Trn V Thuyt 11061491Nguyn Cng 11041971Nguyn Vn Tn 11265281

PCM(Pulse code

modulation)


2/77

Topics to be Covered

-Analog to digital: PCM (pulse code modulation)

Getting sample Quantization

Encoding

-Line coding

-PCM systems and digital time division multiplexing (TDM)-Circuit switch

-Introdution


3/77

Digital and Analog Signals Some signals (like speech and video) are inherently

analog; some (like computer data) are inherentlydigital.

However both analog and digital signals can be

represented and transmitted digitally.Advantages of digital: Reduced sensitivity to line noise, temp. drift, etc. Low cost for switching and transmission.

Lower maintenance costs than analog. Uniformity in carrying voice, data, video, fax, etc. (a

bit is a bit) Better encryption.


4/77

Power Spectral Density Power spectrum(power spectral density)

describes how the average power isdistributed with respect to frequency.

Deterministic signals Fourier transform

Random signals Power spectral density A statistical representation for all

random signals in a particular application


5/77

Bandwidth For random signals, bandwidth is determined from

the power spectral density.

Bandwidth is determined only from the +vefrequencies.

There are different bandwidth definitions Absolute bandwidth

95% bandwidth Null-to-null bandwidth


6/77

Analog SignalsAnalog (continuous-time, continuous-amplitude)

signals (like speech) have a certain bandwidth.Their power spectrum(power spectral density)describes how their average power is distributedwith respect to frequency.

Powerspectraldensity(watts/Hz)

0 1 2 3 4 5 6 7....

High-fidelity speech

Telephone speech(limited by filtering)Bandwidth


7/77

Pulse Code Modulation (PCM) Key points PCM signal is developed by three steps:

sampling, quantizingand encoding.

Quantizing noise is reduced by using

variable sized steps. It is independent of linelength.

s(t) s(n)

Sampleat t=n Quantize Encode

011010001...

Filter


8/77


9/77

Getting Sample


10/77

Signal Sampling: The process variable signal

continuous signal into discrete samples over time.

facility-based sampling theorem kachenhihcop

Sampling Conditions: Signal (t) with unlimited

bandwidth fmax Sample rate: fs 2fmax.

Nyquist frequency fs/ 2, also known as the cutoff

frequency

Approximately Nyquist [- fs/ 2, fs/ 2]


11/77

Sampling ideal


12/77

where: Ts: sampling period [sec]

fs= 1 / Ts: the sampling frequency[Hz] or sampling rate [samples / sec]

Logical choice is the value of fsimportant issues:

fsmust be large enough to represent

the full nature of the signal.fstoo big to higher hardware

requirements, memory usage, etc. ...


13/77

Description of the

sampling process Described in the time domain

Stakeer Goverent


14/77

Described in the frequency

domain


15/77

Description of the sampling

process The relationship between input - output of the sample:

In the time domain:

In the frequency domain:


16/77

Comment:The process of creating a broad-spectrum

sampling infinite but cyclic fs cycle.

That is, the spectrum of xs(t) is thespectrum of x (t) and repeated at a

frequency fs, 2fs, etc ...


17/77

Nyquist Sampling TheoremUsing a low-pass filter

to restore the signal.

To restore true then:

where: fs: Nyquist rate

fs/ 2 Nyquist frequency[-fs/ 2; fs/ 2]: about Nyquist.


18/77


19/77

As such, in order from the sample can be properly

restored original signal, when taken

sample to select the sampling rate is greater than or at

least equal to two times the frequency components highest number of analog signals.

Nyquist Theorem define the lower limit of fs.


20/77

Nyquist Sampling Theorem :

The upper limit of fs :

Suppose p: time to process each data sample (depending

on hardware).

fp= 1 / Tp: speed of processing each sample.

For the sample values do not overlap, then

In summary, the range of fs:


21/77


22/77

Quantization:The process of approximation of discrete sample values

transforming a set of discrete sample values into a very large

number value less.

Position of blocks quantization in the system:

Double style quantization:

Style uniform.

Style nouniform.

Quantizationsampling encoding


23/77

Quantization:

Property of the reflected quantized input relations - output.

Example: uniform quantizer 3 bit.

Bipolar format Monopole format


24/77

Quantization


25/77

Quantization

Input signal Input signal

Output signalOutput signal

The more steps (levels) the less quantization noise. Nonuniform quantization

(e.g. -law) allows a larger dynamic range (important for speech).

NonuniformUniform


26/77

=

2

Quantization

With the quantization have full-scale range R, performances

B bit 2b level quantization.. With the quantization have a

global distance

Width quantization:

Quantization error :

or:

Quantization error and quantization noise random variable with evenlydistributed, error utility profile.

e = xsQ xs

e(t ) = xsQ (t ) xs (t )

= =

12


27/77

Quantization

The ratio SNR of the quantized:

(law 6 dB on bit)

Remarks :

The ADC increase by 1 bit, the ratio SNR 6 dBincrease.

The higher the number of bits as small quantum noise.

The ratio SNR not dependent on the signal amplitude.

Example: Telephone system: = 8 ; performances 8bit/sample; R=10.

Solution: Utility quantum error :

SN R = 6 B [dB ]

=

12=

2

12=

10

2 12= 11.3 ()

Rate bit

. = 8( ) 8 ( sec) = 64


28/77

Encryption process is

performed each

discrete value xq (n)

by a sequence

obinary bits b.

Encryption


29/77


30/77

1. Another method is standardized by

CCITT encoding rules used in

paragraph 13 telecommunication

networks.

Law 13 coding segments (13-segment

coding law) the conversion of the

signal amplitude of the signal x as

follows:


31/77


32/77


33/77

LINE CODE

Line coding is changed signal so that it is consistent withthe waveform characteristics revenue channels anddevices.

ClassificationBase band coding and modulation codingBase band coding : transform the source data into a squarewaveform voltage.

Modulation coding : source signal is transmitted over a longdistance modulated by carrier.Diagram of base band:


34/77

LINE CODE

UNIPOLAR POLAR BIPOLAR

NRZ RZ BIPHASE AMI B8ZS HDB3

NRZ-L NRZ-I MANCHESTER DIFFERENTIALMANCHESTER

A


35/77

Unipolar coding

Simple, low cost. DC high averagelevels, ability to nag at the very least bit0 or 1 chain stretching.

A voltage level perform for bits 0 and a voltage levelperform for bit 1. typically ,Bit 1is performed at highvoltage.


36/77

Polar coding : NRZ (NRZ- L

and NRZ- I), RZ , BIPHASE

NRZ-L

DC component decreased compared with unipolarcodes, hard sync when multiple bit 0 or 1 in a row.


37/77

NRZ-I

Synchronization problem was solved when havingconsecutive sequence bit 1.


38/77

RZ

Synchronization problem has been resolved. But broadband andthere are three voltage levels. This is considered the mosteffective method.


39/77

Biphase coding

Manchester

Existence of positive and negative voltage in 1 bit. 0 DC componentsby better synchronization method.

Differential Manchester


40/77

Bipolar coding: AMI, B8ZS and HDB3.

AMI

Eliminate the DC component, bit synchronizationsequence 1, not 0 synchronous bit stream.

Bit 0 = 0 V.Bit 1: perform alternating voltagepositive or negative.


41/77

- B8ZS

If encryption is 0 consecutive group of8 bits is encoded:

+ 00000000 -> + 000+-0-+-00000000 -> - 000 - + 0 + -


42/77

HDB3

If you do not have to group 4 consecutive bits are coded 0 v 0. if bit 0of 4 in a row it will calculate the total number of pulses.

Odd numbers : + 0000 -> +000+even numbers: + 0000 -> + - 00-

-0000->-+00+


43/77

LINE CODE Averag

e DC

The ability to sync bandwi

dth

Unipolar large Less when transmitted bit

sequence of 0 or 1

LOW

NRZ-I LOW Less when transmitted bitsequence of 0

LOW

RZ LOW GOOD HIGH

MANCHESTER 0 GOOD HIGH

AMI 0 Less when transmitted bitsequence of 0

LOW

B8ZS 0 GOOD LOW

HDB3 0 GOOD LOW

Summary of Line odes


44/77

44


45/77

MODULATION CODINGConsists of analog modulation and digital

modulation:

Analog modulation

AM

FM

amplitude carrier signal changesaccording to the news, do not change

the frequency and phase


46/77

FM

PM


47/77

Digital modulation

ASK amplitude shift key modulation: the course of the bits 1 and 0by changing the amplitude of the carrier signal (frequency andphase do not change). ASK usually sensitive to noise amplitude.


48/77

FSK frequency shift key modulation: a method that changes thecarrier frequency to represent the bit 1 and 0 (amplitude andphase angle do not change).


49/77

- PSK phase shift key modulation: carrier phase change torepresent the bit 1 and 0 (amplitude and frequency did notchange)


50/77


51/77

51

PCM systems and digital time division multiplexing

(TDM)

In digital multiplexing several messages are transmitted viasame physical channel. For multiplexing 64 kbit/s channels in

digital exchanges following three methods are popular:

PDH (plesio-synchronous digital hierarchy) (the dominant

method today) (50-60, G.702)

SONET (synchronous optical network) (85) SDH (synchronous digital hierarchy) (CCITT 88)


52/77

PDH

Digital transmission systems (T-carrier, E carrier)

combine lower order multiplexing stream to get higher

bit rate

Different streams have small differences in clocksignals.

Solve by adding justification bit

PDH = Plesiochronous Digital Hierachy

almost synchronous


53/77

Digital carrier standard T-carrier

North America, Japan

E-carrier

Europe, Mexico, South America

European PCM frame

32 time slots x 8 bits x 8000 Hz = 2048 kbit/s

frame synchronization slot

signaling or traffic

traffic

125 s

PDH E-1 frame


54/77

T1 carrier system

PCM PCM PCM

PCM PCM PCM

MUX

DEMUX

DS-1 frame


55/77

DS-1 frame

channel#1

channel#2 . . . . . . .

channel#24

frame

bit

193 bits

data rate = (24x8 +1 bit)/125 s = 1.544 Mbps

8 bit

each channel also refers as DS0

125 s


56/77

E1-frame

0 1 2 16 31

125 ms = 32 time slots = 2.048 Mbps

frame synchronization signaling channel

30 voice channel+2 controlchannel


57/77

E - carrier

E1 E1 E1 E1

E2E2 E2 E2

E3E3 E3 E3

E4E4 E4 E4

E5

Thirty 64 kbps channelsare multiplexed to createone 2.048 Mbps E1channel

Four E1 channels aremultiplexed into a single 8.448Mbps E2 channel

Four E2 channels aremultiplexed into a single34.368 Mbps E3 channel

Four E3 channelsare multiplexed

into a single139.264 Mbps E4channel

Four E4 channels aremultiplexed into a single565.148 Mbps E5channel


58/77

Digital carrier comparison

2.048 8.448 34.368 139.264 565.148

1.544 6.312 44.736 274.176

1.544 6.312 32.064 97.728 397.200

64

J1

E1

T1

J2

E2

T2

J3

E3

T3

J4

E4

T4

J5

E5x31

x24

x24

x 4

x 4

x 4

x 4

x 7

x 5

x 4

x 6

x 3

x 4

x 4

Europe

USA

Japan


59/77


60/77

Circuit SwitchCircuit Switch PCM:

A type of transplant switching activity onthe basis of time-multiplexing and pulsecode modulation.

A switched network consists centers(nodes) switches, the terminal andtransmission system.

circuitswcenteritching

communication

lines

communication

lines

lines from the PBX lines from the PBX

lines to subscribers lines to subscribers


61/77

Circuit Switch PCM

Picking swapbetween time slotexecuted by twomethods and can splitspecial or distributionset like following:

- Circuit Switch of time.

- Circuit Switch ofspace


62/77

Circuit Switch of time(T)

Circuit Switch of time about the basic thatperform change the information between thetime slot in the different same line PCM

Ci it S it h f ti (T)


63/77

Circuit Switch of time (T)

Methodology:Theoretically canbe done bytwo methods: general

principle :Onthe signal line,hysteresis unitshave a timesame as 1 oftime slot

The method uses the

hysteresis units


64/77


Using caching methods :Based on the voice samples recorded in thememory buffer and read out at the BM desired time.The address of the memory cells in the BM to write orread is provided by the memory controller CM

The method uses caching

Ci it S it h f ti (T)


65/77


Sequentialcontrol:Sequentialcontrollercontrols the read

(or write) on thememory cells ofthe memory BMa row.Use the time slotcounter counts

cycles R,sequentialcounter will addvalue to a slotafter a period oftime.

Sequentialcontrol:


66/77

Circuit Switch of time (T) Random

controls :Controls the read

(or write) thememory cells ofBM ondemand.

Use memorycontroller CM,CM memory

cells containingthe addressread (or write)of memory BM

Random controls

:


67/77


Improved switching T:

Multiplexer with bits in parallel:

Raising the possibility of switching the mode T-storey parallel transmission of 1 channel signal

through the floor of T


68/77

Circuit Switch of space(S)

- Principle

work principle of Circuit Switch of space baseddatabase Circuit Switch of space be useddiagonal . Circuit Switch of space execution failedto swap the information same the time slot but outtwo line PCM different

Circuit Switch ofspace(S)


69/77



70/77


S switching matrix 4 * 4


71/77


- Switching control of S:The determination of the switching

point can be done in two ways :

- Under the control inputs: Defineoutputs connect to the corresponding

input.

- Under the control output: Determine

what input will connect withcorresponding output



72/77


Control according

output

The number of binary bits required for input

n log2n. Total capacity of the memory CMis: CCM = R.log2n (with R is the number oftime slots in one frame).If S has m switching output, the memory of it

is a total CM: S CCM = mRlog2n


73/77


According to the input control


74/77

The type of multiplexing

Circuit Switch TS

Circuit Switch STS

Circuit Switch TST


75/77

Circuit Switch TS


76/77

Circuit Switch STS


77/77

Circuit Switch TST

chuong 4 day du

Documents