Performance Analysis of CDMA WLL Systems with Imperfect ...downloads.hindawi.com/journals/ijvt/2008/413821.pdf · control and imperfect sectorization on the performance of code division

Hindawi Publishing CorporationInternational Journal of Vehicular TechnologyVolume 2008, Article ID 413821, 8 pagesdoi:10.1155/2008/413821

Research Article

Performance Analysis of CDMA WLL Systems withImperfect Power Control and Imperfect Sectorization

Sami A. El-Dolil

Department of Electronic and Electrical Communication Engineering, Faculty of Electronic Engineering,Menoufiya University, Menouf 32952, Egypt

Correspondence should be addressed to Sami A. El-Dolil, msel [email protected]

Received 20 September 2007; Accepted 18 July 2008

Recommended by Mohsen Guizani

Wireless local loop (WLL) provides reliable, flexible, and economical access to the local telephone service using radio technologyin the place of traditional wireline. In this paper, an analytical model is derived to evaluate the effect of both imperfect powercontrol and imperfect sectorization on the performance of code division multiple access (CDMA) WLL systems. The resultsshow that the capacity degradation, due to imperfect power control, is about 25.8% and 11.5% for single cell and multiple cellsystems, respectively. Increasing the overlapping angle from 0◦ to 5◦ causes the capacity gain to decrease from 6 to 5.53, while thecorresponding sectorization efficiency drops from 100% to 92.3%.

Copyright © 2008 Sami A. El-Dolil. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1. Introduction

Wireless local loop (WLL) is a system that connects sub-scribers to the public switched telephone network (PSTN)using radio signals as a substitute for wireline for all or part ofthe connection between the subscribers and the switch. It isbelieved to be a fast and cost-effective mean to provide localphone service in rural areas and third world countries [1].Since WLL is a fixed radio communication system; narrow-beam antennas can be employed at both the base station(BS) and subscriber’s side so that the propagation betweenBS and subscriber’s equipment is very close to free spacepropagation. This gives many inherent advantages to theWLL system over the traditional cellular systems, such aswider coverage area, reduced interference, higher capacity,no fast fading, and no handoff [2, 3].

CDMA technology has the potential to provide a signif-icant improvement in the capacity of cellular radio systemscompared with FDMA and TDMA systems [4]. However,this improvement is dependent upon the effectiveness of thepower control system, especially on the reverse link. In theabsence of power control, a BS would receive a much strongersignal from a subscriber unit that is geographically close to itthan from a subscriber unit that is farther away. This is theso-called near-far problem [5, 6].

This paper presents a theoretical model to evaluate thereverse-link capacity of CDMA WLL systems in terms ofoutage probability, taking into account the power controlerror.

Sectorization in cellular CDMA systems increases thecapacity in proportion to the number of sectors per cell.In practice, the antenna patterns do not fit the sectorarea perfectly, and there are overlapping between sectors,which causes additional interference on both the reverse andforward link [7]. The imperfect sectorization effect on theperformance of CDMA WLL systems is also considered.

2. Effect of Imperfect Power Control inSingle Cell CDMA WLL Systems

Consider a CDMA WLL single cell system consists of Nsubscriber units transmitting to a BS receiver on the reverse-link. A simplified CDMA transmitter is shown in Figure 1.

The signal transmitted from the ith user to its BS is givenby [5]

Si(t) =√

2Aibi(t)ci(t)cos(w1t + θi

), (1)

where Ai is the transmitted power of the ith user, bi(t) is thedata sequence of the ith user, where each bit has an amplitude

mailto:[email protected]

2 International Journal of Vehicular Technology

ith userbi(t)

Ci(t)√

2Ai cos(w1t + θi)

Spreader Modulator

Spreading codegenerator

Carriergenerator

Figure 1: A simplified CDMA transmitter diagram.

of ±1 and a duration of Tb, ci(t) is the spreading codesequence of the ith user and each of the M chips per code hasa duration Tc, wi is the reverse link-carrier frequency, and θiis the random phase of the ith user carrier.

The received signal at the BS receiver Rup(t) consists ofthe following: interference from other users in the cell whichcalled intracell interference, the receiver noise n(t), and thereceived signal from the desired user.

From Figure 2, Rup(t) is given by

Rup(t) =N−1∑

i=0

aiSi(t − τi

)+ n(t), (2)

where ai represents the path loss of the ith user, τi is therandom delay of the ith user signal at the receiver, and n(t) isthe additive white Gaussian noise (AWGN) of the receiver.

The signal at the output of the matched filter is given by

Z(Tb) = 1

Tb

∫ Tb+τ0

τ0

Rup(t)2cos(w1t + θ

)ci(t − τi

)dt, (3)

where θ is the carrier phase angle in the receiver as shown inFigure 2.

The intracell interference at the output of the matchedfilter is given by

Zint(t) =N−1∑

i=1

Zi(Tb). (4)

2.1. Perfect Power Control

To reduce the near-far problem, as well as the interferencefrom other users and hence to increase the capacity of CDMAWLL system, it is important to apply a power control on thereverse link so that the received power from each user at theBS is controlled to be the constant target power, S, [5] where

a2i Ai = S for i = 0, 1, . . . ,N − 1. (5)

The total noise power is the sum of the intracell interferencepower and the AWGN power of the receiver.

The AWGN power at the output of the matched filter isgiven by [5, 7] as

η = NoRb =(No

Tc

)TcRb = NoW

Gp, (6)

where Rb = 1/Tb is the bitrate of the message sequence bi(t),W = 1/Tc is the chiprate, and N0W is the noise power at thereceiver input.

Thus, after despreading, the noise power η is the inputnoise power decreased by the processing gain Gp = Tb/Tc.The intracell interference power is given by [5]

Iint = var(Zint(t)

) = 1Gp

N−1∑

i=1

a2i Ai, (7)

where var(Zint(t)) is the variance of intracell interference atthe output of the matched filter.

By applying voice activity detection, users transmit onlywhen speech signal is present. We introduce a voice activityvariable (VAF) vi which equals 1 with probability of μ, andequals 0 with probability of 1 − μ. By multiplying (7) by viand using (5) so,

Iint

S= 1Gp

N−1∑

i=1

via2i AiS

= 1Gp

N−1∑

i=1

vi, (8)

the intracell interference to signal power ratio given by (8)is reduced by a factor of Gp after the process of matchedfiltering. Now, we define the ratio Eb/Io, which is the energyper bit to interference density ratio, where Eb = STb, I0 =I/Rb = ITb, and I is the total interference power (sum of Iint

and η) so

EbI0= S

I= 1Iint/S + η/S

. (9)

In (8), the summation of vi over (N − 1) users may beexpressed as

N−1∑

i=1

vi = μ(N − 1), (10)

so that

EbI0= 1μ(N − 1)/Gp + η/S

. (11)

The bit error rate (BER) for the binary phase shift keying(BPSK) modulation can be expressed as

Pb = 12

erfc

(√EbI0

), (12)

where erfc(σ) is the complementary error function [5]. For arequired BER, a required Eb/Io, (Eb/Io)req can be determinedfrom (12). Given (Eb/Io)req, the maximum number of activeusers, other than the ith user, that can be supported by thesystem is given by (11) as

m =N−1∑

i=1

vi =⌊

Gp(Eb/I0

)req

− Gp

S/η

⌋, (13)

where �x� represent the largest integer that is smaller thanor equal x. Provided that the number of active users does

International Journal of Vehicular Technology 3

S0(t)

S1(t)

Si(t)

SN−1(t)

a0

a1

...

ai...

aN−1 Channel

Rup(t)

n(t)

2 cos(w1t + θ)

BS receiver

c0(t)

c1(t)

ci(t)

cN−1(t)

1Tb

∫(·)dt

1Tb

∫(·)dt

1Tb

∫(·)dt

1Tb

∫(·)dt

Matched filters

Decision

Decision

Decision

Decision

...

...

b0(t)

b1(t)

bi(t)

bN−1(t)

Figure 2: Channel and BS receiver block diagram.

not exceed m. However, when the number of active users islarger than m, the BER will be greater than the required BER,and this situation is referred to as system outage. The outageprobability of the single cell system is defined as

Pout = Pr(BER > BERreq) = Pr

(EbI0

<(EbI0

)

req

). (14)

The outage probability is defined as the probability that thenumber of active users being greater than m, that is,

Pout = Pr

(N−1∑

i=1

vi > m

)=

N−1∑

i=1

(N − 1

i

)μi(1− μ)N−1−i.

(15)

2.2. Imperfect Power Control

In a practical system, the power control is not perfect. So, thereceived signal power from the ith user at its BS will differfrom the target power level S by δi dB. This power errorδi is a random variable with a standard deviation σe. Thereare several reasons for δi being nonzero, such as the powermeasurement error at the BS and the inability to adjust thesubscriber unit transmitted power sufficiently fast to force δito zero [5, 8]. The signal power at the output of the matchedfilter for the ith user can be expressed as

S′ = 10δ0/10 · S, (16)

and the intracell interference power is

I′int =1Gp

N−1∑

i=1

vi10δi/10 · S. (17)

The intracell interference to signal power ratio at the outputof matched filter becomes

I′int

S′= 1Gp

N−1∑

i=1

vi10(δi−δ0)/10, (18)

where δ0 and δi are two mutually independent randomvariables of power control errors of the signal and theintracell interferers, respectively. By setting ε = δi − δ0 (18)becomes

I′int

S′= 1Gp

N−1∑

i=1

vi10ε/10 = Iint

S10ε/10, (19)

where ε is a random variable with zero mean and a standarddeviation σε =

√2σe. Following the same procedure as in the

perfect power control case, the ratio Eb/Io can be written as(EbI0

)

imp= 1I′int/S′ + η/S′

= 1(Iint/S

)10ε/10 + η/S′

. (20)

In order to evaluate the system performance, we intro-duce the outage probability that is defined as the probabilityof a system’s BER being greater than 10−3, that is,

Pout = Pr(BER > 10−3)

= Pr

((EbI0

)

imp= S′

I′int + η< γreq

)(21)

= Pr

(I′int + η

S′>

1γreq

)

= Pr

{10ε/10

N−1∑

i=0

vi > Gp

(1γreq

− η

S′

)},

(22)

where γreq is the required Eb/Io to ensure that the BER is lessthan 10−3. If the number of active users inside the cell is k,then (22) can be rewritten as

Pout = Pr

⎧⎨⎩

(k10ε/10 > Gp

(1γreq

− η

S′

))∣∣∣∣∣

(N−1∑

i=0

vi = k

)⎫⎬⎭

× Pr

(N−1∑

i=0

vi = k

)= P1P2.

(23)


The probability P1 in (23) is given by [5]

P1 = Q

(Gp(1/γreq − η/S′

)− kE(10ε/10)√k var

(10ε/10)

), (24)

where

Q(x) = 1√2π

∫∞xe(−y2/2)dy. (25)

The mean of the term 10ε/10 in (24) can be derived asfollowing:

E(10ε/10) =

∫∞−∞

exp[ε ln(10)

10

]exp

(− ε2/4σ2e

)√

4πσ2e

dε

= exp(σe

ln(10)10

)2

,

(26)

and the variance of the 10ε/10 is given by

var(10ε/10) = E

(10ε/10)2 − {E(10ε/10)}2

= exp(σe

ln(10)5

)2

−(

exp(σe

ln(10)10

)2)2

.

(27)

Now, consider the probability that there are k active intracellusers, P2, which is given by [5]

P2 = Pr

(N−1∑

i=0

vi = k

)=

N−1∑

k=0

(N − 1

k

)μk(1− μ)N−1−k.

(28)

The performance of the reverse link in a single cell CDMAWLL system, shown in Figure 3, is evaluated for 20 dB/dec,where WLL has a fixed-to-fixed link so propagation exponentof 2 is used [2], Eb/Io = 5 dB [9], W = 1.25 MHz, Rb =8 Kbps, VAF = 3/8, a signal-to-AWGN ratio of 20 dB at theoutput of the matched filter and a BER outage threshold of10−3 were used in the calculations.

For perfect power control and an outage probability of2%, the single cell system can support up to 89 users/cell asshown in Figure 3. The outage probability of the imperfectpower-controlled system having different standard devia-tions of power control error is also shown in the figure. Foran outage probability of 2% and a standard deviation ofpower control errors of 2 dB, the system can support 66 usersper cell. The capacity degradation, due to imperfect powercontrol, is about 25%.

Table 1 shows the number of users per cell for differentvalues of the standard deviation (STD) of power controlerrors and the percentage decrease in users due to imperfectpower control, for an outage of 2%.

3. Effect of Imperfect Power Control inMultiple Cell CDMA WLL Systems

In addition to the intracell interference, there is nowinterference from neighboring cells, which called intercell

40 45 50 55 60 65 70 75 80 85 90

Number of users/cell

10−4

10−3

10−2

10−1

100

Ou

tage

prob

abili

ty

Standard deviation = 2.5 dBStandard deviation = 2 dBStandard deviation = 1.5 dBPerfect power control

Figure 3: Outage probability of a single cell CDMA WLL system.

Table 1: Number of users per cell for different values of σe, andPout = 2%.

STD Users/cell % decrease

0 dB 89 0%

1.5 dB 69 22.4%

2 dB 66 25.8%

2.5 dB 62 30.3%

interference. The received signal at the BS includes: thedesired signal, intracell interference, the AWGN at thereceiver input, and intercell interference. Figure 4 shows thereverse-link communication system, where the arrangementfor the transmitter and BS receiver is the same as those shownin Figures 1 and 2, respectively.

The received signal at the BS is given by [5]

Rup(t) =N−1∑

i=0

aiSi(t − τi

)+J−1∑

j=1

N−1∑

i=0

ai jSi j(t − τi j

)+ n(t),

(29)

where the intercell interference from the J − 1 surroundingcells is

J−1∑

j=1

N−1∑

i=0

ai jSi j(t − τi j

), (30)

where ai j represents the effects of path loss, τi j is the randomtime delay of the ith user in the jth cell, and si j(t) is the signaltransmitted by the ith user in the jth cell.

For a particular user, say the zeroth one, the signal at theoutput of the matched filter is given by

Z(Tb) = a0

√2A0b0cosϕ0 + Zint

(Tb)

+ Zext(Tb)

+ Zn(Tb),

(31)


Zeroth mobilein the zeroth

Intra-cellinterference

Inter-cellinterference

from J − 1 cells

a0S0(t)

N−1∑

i=1

aiSi(t)

J−1∑

j=1

N−1∑

i=0

ai jSi j(t)

BS receiver...

Rup(t)

n(t)

b0(t)

bN−1(t)

Figure 4: Block diagram of the multicell CDMA reverse-link system.

where ϕ0 is the carrier phase difference. The first term is thedesired signal, the second term is the intracell interferencecomponent, the third term is the intercell interferencecomponent, while the last term is the AWGN component.The intercell interference at the output of the matched filteris given by

Zext(t) = 1Tb

∫ Tb0

J−1∑

j=1

N−1∑

i=0

ai j√

2Aijbi j(t − τi j

)ci j(t − τi j

).

(32)

3.1. Perfect Power Control

Similar to the approach in deriving the intracell interferencepower, the intercell interference power at the output of thematched filter, Iext, can be shown to be

Iext = E((Zext

(Tb))2) = 1

Gp

J−1∑

j=1

N−1∑

i=0

a2i jAi j . (33)

The intercell interference to signal ratio is given by

Iext

S= 1Gp

J−1∑

j=1

N−1∑

i=0

vi ja2i jAi j

S= 1Gp

J−1∑

j=1

I jS

, (34)

where I j /S is the interference to signal power ratio from thejth cell.

Let us consider one of the interfering cells, say cell j , whereits BS is at a distance d from BS0 as shown in Figure 5. Theinterference term I j in (34) is the interference power from allthe users in cell j to the BS0, [5, 10].

If the interfering user in cell j is located at a distance rfrom its BS and ro from the BS0, the interfering user, whenactive, produces an interference to the BS0 given by [11]

I(ro, r

)

S=(

10(ζ0/10)

rα0

)(rα

10(ζm/10)

)=(r

r0

)α10ζ/10 ≤ 1,

(35)

where the first term is, due to the attenuation, caused bydistance and blockage to the given BS, while the secondterm is the effect of power control to compensate forthe corresponding attenuation to the BS of the out-of-cell

dr

MS

R

d

r0

dA

Figure 5: CDMA interference calculation.

interferer, ζ , ζ0, and ζm all are random variables with zeromean and standard deviation σ , (ζ = ζ0− ζm), and r0 is givenby

r0 =√d2 + r2 + 2 drcosθ. (36)

Replace the summation in (34) by an integration over thearea of the cell j , so I j /S will be

I jS=∫ 2π

0

∫ R0vi jI(r0, r

)

S·ϕ(ξ,r

r0

)ρ da, (37)

where ρ is the user density, assuming N users are uniformlydistributed in a circular cell of radius R, ρ = N/πR2, da =r dr dθ is the unit area in Figure 5, Φ(ζ0 − ζm, r/r0) is theconstraint function for the interfering users in the cell j ,which can be defined as [5, 11]

ϕ(ξ,r

r0

)=

⎧⎪⎨⎪⎩

1, if(r

r0

)α10ξ/10 ≤ 1,

0, otherwise.(38)

It is necessary to calculate the mean and variance of theintercell interference power in order to calculate the outageprobability. From (35) into (37) and by taking the mean, weobtain [11]

E(I jS

)= ρμ

∫ 2π

0

∫ R0

(r

r0

)αE(

10ξ/10 ·ϕ(ξ,r

r0

))r dr dθ,

(39)


where E(vi j) = μ. The variance of I j /S can be given by [11]

var(I jS

)

=∫ 2π

0

∫ R0E

[vi j

(r

r0

)α10ξ/10 ·ϕ

(ξ,r

r0

)]2

−{E

[vi j

(r

r0

)α10ξ/10 ·ϕ

(ξ,r

r0

)]}2

ρr dr dθ.

(40)

The total interference-power-to-signal-power ratio for all thesurrounding cells, Iext/S, in (34) has a mean and variance ofthe following:

E[Iext

S

]= 1Gp

J−1∑

j=1

E[I jS

],

var[Iext

S

]=(

1Gp

)2 J−1∑

j=1

var[I jS

].

(41)

The ratio Eb/Io at the output of the matched filter can bewritten as

EbI0= 1Iint/S + Iext/S + η/S

. (42)

The system performance in terms of the outage probabilitythat has a BER greater than 10−3 is

Pout =N−1∑

k=0

(N − 1

k

)μk(1− μ)N−1−k

×Q(

1/γreq − η/S− k/Gp − E(Iext/S

)√

var(Iext/S

)).

(43)

3.2. Imperfect Power Control

The received signal power S′ from a user at its BS will differfrom the target power level S by δ0 dB. This error power is arandom variable with standard deviation σe. Using (16) and(35), the interfering power to received signal power ratio willbe

I′(ro, r

)

S′=(r

r0

)α10ζ/1010δi j−δ0 =

(r

r0

)α10ζ/1010ε/10,

(44)

where δi j is the power error for the ith user in cell j . The totalintercell interference-to-signal ratio is

I′ext

S′= Iext

S10ε/10. (45)

The ratio Eb/Io can be written as(EbI0

)

imp= 1I′int/S′ + I′ext/S′ + η/S′

+1(

Iint/S)10ε/10 +

(Iext/S

)10ε/10 + η/S′

.

(46)

30 35 40 45 50 55 60 65 70

Number of users/cell

10−4

10−3

10−2

10−1

100

Ou

tage

prob

abili

ty

Standard deviation = 2.5 dBStandard deviation = 2 dBStandard deviation = 1.5 dBPerfect power control

Figure 6: Outage probability of the multiple cell CDMA WLLsystem.

Following the same procedure as employed in the perfectpower control case, Pout is found as

Pout=N−1∑

k=0

(N − 1

k

)μk(1− μ)N−1−k

×Q⎛⎝1/γreq−η/S′−k/Gp−E

(I′int/S

′)−E(I′ext/S′)

√var(I′int/S′

)+ var

(I′ext/S′

)

⎞⎠ ,

(47)

where I′int/S′ and I′ext/S

′ are two independent Gaussiandistributed random variables, whose mean and variance maybe expressed as

E

[I′int

S′+I′ext

S′

]= E

[Iint

S10ε/10

]+ E

[Iext

S10ε/10

],

var

[I′int

S′+I′ext

S′

]= var

[Iint

S10ε/10

]+ var

[Iext

S10ε/10

].

(48)

The performance of the reverse-link CDMA WLL systemis shown in Figure 6. For an outage probability of 2%, theperfect power-controlled system can support 52 users/cell fora VAF of 3/8. The number of users per cell for different valuesof the standard deviation of power control errors and thepercentage decrease in users, due to imperfect power control,are displayed in Table 2 for an outage of 2%.

4. Effect of Imperfect Sectorization inCDMA WLL Systems

In the case of perfect directional antennas, there is a sharpseparation between the sectors. Due to overlap and sidelobe


Table 2: Number of users per cell for different values of σe, andPout = 2%.

STD Users/cell % decrease

0 dB 52 0%

1.5 dB 47 9.6%

2 dB 46 11.5%

2.5 dB 44 15.4%

360/Ns + θ0

θ0

(a)

360/Ns + θ

θ0

(b)

Figure 7: Imperfect sectorization: (a) three sector; (b) six sector percell.

of practical antenna, the BS still receives some interferencefrom users in other sectors [11, 12].

Figure 7 shows a sectorized cell arrangement having Ns

overlapping sectors, each with an angle of (2π/Ns)+θ0, whereθ0 is the overlapping angle. If there is no overlapping, that is,θ0 = 0, then 1/Ns of the total interference is received, andthe capacity gain, due to sectorization, is Ns times that of anunsectorized cell.

In a sectorized cell, only (2π/Ns)+θ0/2π of the total inter-ference from the surrounding is received. For this condition,the capacity gain, due to sectorization, is 2π/(2π/Ns) + θ0

times that of the unsectorized cell.We define the efficiency of sectorization es as the ratio of

the capacity gain with the sector antennae pattern having anoverlapping angle θ0 to the nonoverlapping antennae patternof 2π/Ns, so

es =2π/

(2π/Ns + θ0

)

2π/(2π/Ns

) = 2π/Ns

2π/Ns + θ0. (49)

So, the imperfect sectorization capacity gain Gimp will be

Gimp = Nses. (50)

In WLL system, due to fixed-to-fixed link, six sectors percell arrangement can be used, the capacity gain due tosectorization and its corresponding sectorization efficiencyfor different values of overlapping angle θ0 in degrees isshown in Figure 8. From the figure, the interference on thereverse link increases as the overlapping angle θ0 is increased,which causes the capacity gain and sectorization efficiency todecrease proportionally. Increasing θ0 from 0◦ to 5◦ causesthe capacity gain to decrease from 6 to 5.53, while thecorresponding sectorization efficiency drops from 100% to92.3%.

0 1 2 3 4 5 6 7 8 9 10

Overlapped angle (deg)

5

5.1

5.2

5.3

5.4

5.5

5.6

5.7

5.8

5.9

6

6.1

Cap

acit

yga

in

0.87

0.88

0.9

0.91

0.93

0.95

0.97

0.98

1

Sect

oriz

atio

neffi

cien

cy

Figure 8: Reverse-link sectorization gain and efficiency as afunction of the overlapping angle.

5. Conclusion

CDMA technology has the potential to provide a significantimprovement in the capacity of WLL systems compared withFDMA and TDMA systems. However, this improvementis dependent upon the effectiveness of the power controlsystem, especially on the reverse link. In this paper, atheoretical model to evaluate the reverse-link capacity ofCDMA WLL systems in terms of outage probability, takinginto account the power control error, is obtained.The resultsshow that the capacity degradation, due to imperfect powercontrol, is about 25.8%, and 11.5% for single cell andmultiple cell systems, respectively. The effect of imperfectsectorization on the performance of CDMA WLL systemsis also considered. The interference on the reverse linkincreases as the overlapping angle is increased, which causesthe capacity gain and sectorization efficiency to decreaseproportionally as shown in Figure 8.

References

[1] W. Webb, Introduction to Wireless Local Loop, Artech House,Norwood, Mass, USA, 1998.

[2] D. Lee and C. Xu, “The effect of narrowbeam antenna andmultiple tiers on system capacity in CDMA wireless localloop,” IEEE Communications Magazine, vol. 35, no. 9, pp. 110–114, 1997.

[3] H. Stellakis, A. Giordano, A. Aksu, and W. Biagini, “Reverselink performance of wireless local loop CDMA networks,”IEEE Communications Letters, vol. 4, no. 2, pp. 49–51, 2000.

[4] P. Jung, P. W. Baier, and A. Steil, “Advantages of CDMAand spread spectrum techniques over FDMA and TDMAin cellular mobile radio applications,” IEEE Transactions onVehicular Technology, vol. 42, no. 3, pp. 357–364, 1993.

[5] R. Steele, C.-C. Lee, and P. Gould, GSM, cdmaOne and 3GSystems, John Wiley & Sons, New York, NY, USA, 2001.

[6] W. C. Y. Lee, “Overview of cellular CDMA,” IEEE Transactionson Vehicular Technology, vol. 40, no. 2, pp. 291–302, 1991.

[7] C.-C. Lee and R. Steele, “Effect of soft and softer handoffson CDMA system capacity,” IEEE Transactions on VehicularTechnology, vol. 47, no. 3, pp. 830–841, 1998.


[8] W.-M. Tam and F. C. M. Lau, “Analysis of power control andits imperfections in CDNA cellular systems,” IEEE Transactionson Vehicular Technology, vol. 48, no. 5, pp. 1706–1717, 1999.

[9] V. K. Garg and E. L. Sneed, “Digital wireless local loop system,”IEEE Communications Magazine, vol. 34, no. 10, pp. 112–115,1996.

[10] K. I. Kim, “CDMA cellular engineering issues,” IEEE Transac-tions on Vehicular Technology, vol. 42, no. 3, pp. 345–350, 1993.

[11] M. G. Jansen and R. Prasad, “Capacity, throughput, and delayanalysis of a cellular DS CDMA system with imperfect powercontrol and imperfect sectorization,” IEEE Transactions onVehicular Technology, vol. 44, no. 1, pp. 67–75, 1995.

[12] A. Ahmad, “A CDMA network architecture using optimizedsectoring,” IEEE Transactions on Vehicular Technology, vol. 51,no. 3, pp. 404–410, 2002.

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttp://www.hindawi.com Volume 2010

RoboticsJournal of

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation http://www.hindawi.com

Journal ofEngineeringVolume 2014

Submit your manuscripts athttp://www.hindawi.com

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation http://www.hindawi.com

Volume 2014

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

SensorsJournal of


Modelling & Simulation in EngineeringHindawi Publishing Corporation http://www.hindawi.com Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks


Documents

Performance Analysis of CDMA WLL Systems with Imperfect ...downloads.hindawi.com/journals/ijvt/2008/413821.pdf · control and imperfect sectorization on the performance of code division