Slide thông tin di động đại học bách khoa hà nội chương 8 MẠNG THÔNG TIN DI ĐỘNG 3g tiếp

Multiple Access Schemes  Frequency Division Multiple Access FDMA, different frequencies for different users  example Nordic Mobile Terminal NMT systems  Time Division Multiple Access

Trang 1

Trường Đại học Bách Khoa Hà Nội

Khoa Điện tử Viễn thông

(Universal Mobile Telecommunications System)

ξ 2 Nguyên lý trải phổ và sử dụng mã trong UMTS

cuu duong than cong com

Trang 2

Multiple Access Schemes

 Frequency Division Multiple Access (FDMA), different frequencies for different

users

 example Nordic Mobile Terminal (NMT) systems

 Time Division Multiple Access (TDMA), same frequency but different timeslots

for different users,

 example Global System for Mobile Communication (GSM)

 GSM also uses FDMA

 Code Division Multiple Access (CDMA), same frequency and time but users are

separated from each other with orthogonal codes

Cod e

Frequenc y Tim

e

1 2 N

Trang 3

CDMA

• Multiple users occupying the same band simultaneously by

having different codes is known as Code Division Multiple

Access or CDMA.

• This leads to universal frequency reuse.

• CDMA is a Spread Spectrum technique.

• Direct Sequence - CDMA is used in mobile communication.

Data X Spread Data Modulation Carrier

Code

Spread Spectrum

information bandwidth i.e transmitted signal is spread to a wider

bandwidth

 Bandwidth is not dependent on the information signal

 More secure communication

 Reduces the impact of interference (and jamming) due

to processing gain

Trang 4

Spread Spectrum

 Where does spread spectrum come from

 First publications, late 40s

 First applications: Military from the 50s

 Rake receiver patent 1956

 Cellular applications proposed late 70s

 Investigations for cellular use 80s

sequence i.e with spreading/channelization code

 The bits of the channelization code are called chips

 Chip rate (W) is typically much higher than bit rate (R)

 Codes need to be in some respect orthogonal to each

other

 defines how many chips are used to spread a single

information bit and thus determines the end bit rate

 Shorter code equals to higher bit rate but better Signal

to Interference and Noise Ratio (SINR) is required

 Also the shorter the code, the fewer number of codes are available

 Different bit rates have different geographical areas

covered based on the interference levels

Trang 5

Direct Sequence

 Information signal is multiplied with

channelization code => spread signal

 Spread signal is multiplied with channelization

code

 Multiplied signal (spread signal x code) is then

integrated (i.e summed together)

 If the integration results in adequately high (or low) values, the

signal is meant for the receiver

Direct Sequence

Trang 6

Direct Sequence

Processing gain and Spreading

Frequency

Despread narrowband signal

Spread wideband signal

W R

Trang 7

processing gain

 G[dB]=10*log 10 (W/R), where ’W’ is the chip rate and ’R’ is the user bit rate

loss is defined as:

 L p = 10*log 10 k, where ’k’ is the amount of users

 G tot =10*log 10 (W/kR)

smaller coverage

network (voice, web browsing, videophone) due to different bit rates

 Thus, the coverage area and capacity might be different for different

services depending on the radio network planning issues

Trang 8

self-interference that is necessary in order to reuse the available 5 MHz

carrier frequency over geographically close distances.

 processing gain 10 log10(3.84e6/12.2e3) = 25 dB

 For speech service the required SINR is typically in the

order of 5.0 dB, so the required wideband

signal-to-interference ratio (also called “carrier-to-signal-to-interference

ratio, C/I ) is therefore “5.0 dB minus the processing” =

-20.0 dB

 In other words, the signal power can be 20 dB under

the interference or thermal noise power, and the

WCDMA receiver can still detect the signal

 Notice: in GSM, a good quality speech connection

requires C/I = 9–12 dB

Principle of spreading

Trang 9

Trải phổ và giải trải phổ

Trang 10

Rx * Code2 Data de-spreaded

Trang 11

 WCDMA is the radio interface for UMTS systems

 Wide bandwidth, 3.84 Mcps (Megachips per second)

 Maps to 5 MHz due to pulse shaping and small guard

bands between the carriers

 Users share the same 5 MHz frequency band and time

 UL and DL have separate 5 MHz frequency bands

 High bit rates

 With Release ’99 theoretically 2 Mbps both UL and DL

 384 kbps highest implemented

 Fast Power Control (PC)

=> Reduces the impact of channel fading and minimizes

the interference

Trang 12

Codes in UMTS

Codes in WCDMA

 Channelization Codes (=short code)

 Codes from different branches of the code tree are orthogonal

 Length is dependent on the spreading factor

 Used for

 channel separation from the single source in downlink

 separation of data and control channels from each other in the uplink

 Same channelization codes in every cell / mobiles and therefore the additional

scrambling code is needed

 Scrambling codes (=long code)

 Very long (38400 chips = 10 ms =1 radio frame), many codes available

 Does not spread the signal

 Uplink: to separate different mobiles

 Downlink: to separate different cells

 The correlation between two codes (two mobiles/NodeBs) is low

 Not fully orthogonal

Trang 13

Codes in UMTS

Channelization codes separate different connection

Scrambling codes separate different mobiles

Trang 14

Codes in UMTS

Channelization codes separate different connection

Scrambling codes separate different mobiles

Codes in UMTS

 Channelization Codes (=short codes)

 Defines how many chips are used to spread a

single information bit and thus determines the

end bit rate

 Length is referred as spreading factor

 Used for:

 Downlink: Separation of downlink connections to different users within one cell

 Uplink: Separation of data and control channels from same terminal

 Same channelization codes in every cell /

mobiles

 additional scrambling code is needed

Trang 15

Codes in UMTS

 Scrambling codes (=long codes)

 Very long (38400 chips), many codes available

 Does not spread the signal

 Used for

Downlink: to separate different cells/sectors

Uplink: to separate different mobiles

 The correlation between two codes (two

mobiles/NodeBs) is low

Codes used

Trang 16

31

Codes and their Use

Codes in WCDMA

 The relation between downlink physical layer bit rates and codes

Trang 17

Channelisation code generation process

Trang 18

An example of the generation of two channelisation codes

Trang 19

The orthogonality

property of Walsh codes

The orthogonality property

of different length OVSF codes

Trang 20

The cross correlation of two

different Walsh codes with a non-zero offset

An example cross correlation function of two

eight-chip Walsh codes

Trang 23

The cross correlation properties of scrambling codes

Định dạng
Số trang	23
Dung lượng	1,2 MB