Ho Chi Minh City University of Technology

Faculty of Electrical and Electronics Engineering

Department of Telecommunications
----------------------------------------------------------

Interference and Capacity in

Cellular Systems

Lectured by Ha Hoang Kha, Ph.D.

Ho Chi Minh City University of Technology
Email: [email protected]
Content

1. System Capacity of FDMA

2. System Capacity of CDMA

H. H. Kha 2
References

 Ta Tri Nghia, Wireless Communications, lecture notes.

 T.S. Rappaport, Wireless Communications, Prentice Hall
PTR, 1996.
 A. Goldsmith, Wireless Communications, 2004.
 Tallal Elshabrawy, wireless Communications, lecture
notes.

H. H. Kha 3
Interference & System Capacity

 The wireless environment constitutes a shared

medium.
 Interference is the major limiting factor in performance
of wireless systems in general.

 Types of Interference:
• Co-channel interference
• Adjacent channel interference

H. H. Kha 4
Co-channel Interference

Frequency reuse implies that several cells use the same set
of channels
B
B G C
G C A
A F D
F D E B
E B G C
B G C A
G C A F D
A F D E
F D E B
E B G C
G C A
A F D
F D E
E

Frequency reuse = 7
Co-channel interfering cells for cell allocated with channel group A

H. H. Kha 5
Co-channel Interference, SIR & System Capacity

Co-channel
P2 P1H11
Interfering Cells SIR1 =
H21
BS 2
P2H21
P1 H12 H22
P2H22
SIR 2 =
H11 R

P1H12
D
MT 2
BS 1
R MT 1

BS: Base Station

MT: Mobile Terminal
Px: Transmitter power by base station x
Hxy: Small-scale & Large-scale channel between base station x and mobile terminal y
SIRy: Signal-to-Interference Ratio at mobile terminal y

 Improving SIR1 by increasing P1 would result in a decrease in

SIR2
 Improving BOTH SIR1 & SIR2 is possible by increasing the
distance separation between BS1 and BS2
H. H. Kha 6
Distance Separation between Base Stations

3
R' = R
2
D where R is the Cell Radius

j(2R’)
i(2R’)

Q: Co-channel reuse Ratio

H. H. Kha 7
SIR Computations

Assume interference from first tier (ring)

of co-channel interferers

−n
 d 
Pr = P0  
d
 0 D
D
P0 (R d0 )
−n
D
R −n
SIR =
k NB

∑ 0 ( i 0)
−n X
i=1
P D d ∑D
i=1
i
−n R

D
D
( )=
n
3N Qn
⇒ SIR
= D
NB NB
Di: interfering distance from ith
co-channel interference
NB No. of co-channel
interfering sites
H. H. Kha 8
SIR Computations

Assume interference from first tier (ring)

of co-channel interferers

−n
 d 
Pr = P0  
d
 0 D+R

P0 (R d0 )
−n D+R
R −n
SIR =
k NB
D R
∑ 0 ( i 0)
−n

i=1
P D d ∑D
i=1
i
−n

X
Worst Case SIR D-R D
R −n D-R
⇒ SIR =
2 (D − R ) + 2 (D + R ) + 2 (D )
−n −n −n

1 Di: interfering distance from it

⇒ SIR = co-channel interference
2 ( Q − 1) + 2 ( Q + 1) + 2 ( Q )
−n −n −n
NB No. of co-channel
interfering sites

H. H. Kha 9
SIR & System Capacity

SIR ∝ Q , Q=D R =
n
3N

 Improving SIR means increasing cluster size, which

corresponds to a decrease in system capacity
 Decreasing the cell size does not affect the SIR as Q=D/R
remains constant. A decrease in cell size corresponds to an
increase in system capacity

H. H. Kha 10
Example

 In First Generation cellular systems, sufficient voice quality

is achieved when SIR = 18 dB

1
SIR =
2 ( Q − 1) + 2 ( Q + 1) + 2 ( Q)
−n −n −n

 N=7 Q=4.6. Worst Case SIR = 49.56 (17 dB)

To design cellular system with worst performance better than

18 dB, N=9

 Capacity reduction = 7/9

H. H. Kha 11
Adjacent Channel Interference

 Adjacent channel interference results from imperfect

receiver which allows nearby frequencies to leak into the
passband
 Adjacent channel interference can be minimized through
careful filtering and channel assignments

H. H. Kha 12
Improving Coverage and Capacity in Cellular
Systems: Cell Splitting

Subdividing a congested cell into smaller cells, each with

its own base station and a corresponding reduction in
antenna height and transmitter power
Cell splitting Increasing system capacity by increasing
the number of clusters in a given area
Decreasing Transmitter Power
The SIR is independent of
SIR ∝ Qn , Q=D R =
3N transmitted power as long as it
is the same for all base stations

Why not make Transmitter Power as low as possible?

The SNR must be a above a minimum
SNR = Pr Noise threshold controlled by Pr

H. H. Kha 13
Capacity of FDMA Cellular Systems

Capacity: Maximum number of channels or users that could

be supported while maintaining a certain SIR requirement
( )=
n
3N Assume interference from first tier (ring) of
Qn
⇒ SIR
= co-channel interferers
NB 6
Qn 1
⇒ ≥ SIR th ⇒ Q ≥ ( 6 × SIR th ) n
6

B D
m= t channels/cell D
BcN D
R
X
Q = 3N
Bt Bt D D
=m =
Bc Q2 3 ( )  6 n
2
D
Bc  n 2 SIR th 
3 
For n=4
Di: interfering distance from ith
Bt m: Number of Channels per Cell
co-channel interference
m= Bt : Total available bandwdith
2 Bc : Channel bandwdith NB No. of co-channel interfering
Bc SIR th
3 N: Cluster Size sites

H. H. Kha 14
SINR in CDMA Systems: Base Station-to-Mobile
(Forward Link)

( )
n
P0 d0 dj
SINR j = K

∑P (d )
n
0 0 dj + Noise
i=1
dj
i≠ j
MTj
d2
Effective SINR because of Codes Cross-
dK
Correlation (Pseudo-Random Codes)
MT2 d1
( ) ( )
n n
P0 d0 dj GP0 d0 dj
SINR j =
1 1
(K − 1) P0 ( d0 )
eff n
( )
n

G
( K − 1) P0 d0 dj + Noise
G
dj + Noise MTK

MT1
G is called the processing gain (Tsymbol/Tchip), reflects the decline in
perceived interference due to spreading and de-spreading
The term Noise reflects noise power in the spreaded bandwidth
Single cell with K active
Effective SINR because of Codes Cross- Mobile Terminals
Correlation (Orothgonal Codes)

( ) ( )
n n
P0 d0 dj GP0 d0 dj
SINR j =
eff
1 Noise
( )
n
0 × (K − 1) P0 d0 dj + Noise
G

H. H. Kha 15
SINR in CDMA Systems: Mobile-to-Base Station
(Reverse Link)

( )
n
P0 d0 dj
SINR j = K

∑ P (d di ) + Noise
n
0 0 dj
i=1
i≠ j
MTj
d2
Effective SINR because of Codes Cross-
dK
Correlation (Pseudo-Random Codes)
( ) ( )
n n
P0 d0 dj GP0 d0 dj MT2 d1
SINR j eff =
1 K 1 K

∑ 0 ( 0 i) ∑ P0 ( d0 di ) + Noise
n n
P d d + Noise
G i 1= G i 1
MTK
i≠ j i≠ j

MT1
G is called the processing gain (Tsymbol/Tchip), reflects the decline in
perceived interference due to spreading and de-spreading
The term Noise reflects noise power in the spreaded bandwidth
Single cell with K active
Effective SINR because of Codes Cross- Mobile Terminals
Correlation (Orothgonal Codes)
( ) ( )
n n
P0 d0 dj GP0 d0 dj
SINR j eff = K
1 Noise
0 × ∑ P0 ( d0 di ) + Noise
n

i=1 G
i≠ j

H. H. Kha 16
The Near-Far Effect
Mobile-to-Base Station Communication
(Uplink or Reverse Link)

GP0 ( d0 d1 )
n
d1
SINR1 =
P0 ( d0 d2 ) + Noise
eff n
MT1
d2

GP0 ( d0 d2 )
n

SINR 2 =
P0 ( d0 d1 ) + Noise
eff n

MT2

The Near-Far Effect

SINR2|eff is much worse than SINR1|eff
The strongest received mobile signal may capture the demodulator at
the base station
The Near-Far effect is the resultant of multiple users using the same
transmit power level to communicate with the base station over the
same frequency and in the same time within the same cell.
H. H. Kha 17
Strength-Based Power Control

To address the Near-Far Effect

Control the transmission power dj

of each mobile terminal such d2

MTj

that it is received at the base dK

station at an equal level S d1
MT2
Mobile-to-Base Station Communication
MTK
(Uplink or Reverse Link)
MT1
Strength-Based Power Control

( )
n
0 ( d0 d1 ) 0 ( d0 d2 )
n n
P= P= P=
0 d0 d j S

GS
SINR j =
eff
(K − 1) S + Noise
H. H. Kha 18
Capacity of CDMA Cellular Systems

Assume Forward Link and Ignore Noise

GP0 ( d 0 R )
n

⇒ SIR =
(K − 1) P0 ( d 0 R ) + 6KP0 ( d 0 D )
n n

G
⇒ SIR =
(K − 1) + 6K (R D )
n
D
D
G D
⇒ ≥ SIR th X
(K − 1) + 6K ( Q )
−n R
D D
G
≥ (K − 1) + 6K ( Q )
−n
⇒
SIR th D
 G 
 SIR + 1 
K≤ th  Di: interfering distance from ith
 6 ( Q) + 1 
−n

  co-channel interference
  NB No. of co-channel interfering
 G  sites
 SIR + 1 
For n=4,Q= 3N ⇒ K ≤  th 
2
 N−2 + 1  K: Number of Channels (i.e., Codes) per Cell

 3  N: Cluster Size
 

H. H. Kha 19