A New User Mobility Based Adaptive Power Control in CDMA Systems
ABSTRACT We propose a new closed-loop power control scheme for wireless mobile communication systems using an adaptive step size. The proposed scheme selects the basic power control step size by considering the speed of the mobile station and a variable step size by using instantaneous companding logic based on power control command bit patterns. We show its improved performance in view of the standard deviation of received power at the base station in consideration of channel BER.
- [show abstract] [hide abstract]
ABSTRACT: For pt.I see ibid., vol 41, p.1626-34 (1993). Power control is essential in the use of direct-sequence code division multiple-access (CDMA) techniques. Early system-level performance analyses of a CDMA approach to wireless mobile and personal communications have assumed the ability of power control to equalize the absolute signal powers of CDMA users received at each base station. The present paper studies a more practical, although analytically more complicated, uplink power control technique that uses measurements of the received signal-to-interference ratio (SIR) instead. A combination of discrete-event link simulation and analysis of the obtained SIR statistics is used to explore the previously little-known behavior of a CDMA system using SIR-based power control and to obtain performance estimates for such a system under various operating assumptions. The overall results indicate that power control based on SIR has the potential for somewhat higher system performance than power control based on absolute signal strength assumed in the early analysesIEEE Transactions on Communications 03/1994; · 1.75 Impact Factor
Article: Hybrid Companding Delta Modulation[show abstract] [hide abstract]
ABSTRACT: We present a reasonably complete account of an improved adaptive delta modulation (ADM) system called hybrid companding delta modulation (HCDM). The HCDM system that is far superior to continuously variable slope DM (CVSD) or constant factor DM (CFDM) is advantageous, particularly for speech coding. It employs both syllabic and instantaneous companding schemes. Performance analysis of the system has been done and verified by computer simulation. In getting the mathematical formula for HCDM granular noise, a new method based on amplitude distribution is proposed. Optimization of the system parameter values by simulation is also discussed. In addition, an efficient method of hardware implementation is considered.IEEE Transactions on Communications 10/1981; · 1.75 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Performance of a reverse link code-division multiple-access (CDMA) system with fast close-loop power control algorithms is studied. It is found that if the fast close-loop power control algorithm functions effectively, the speed of the mobile unit is in the range such that its Doppler frequency is less than one tenth of the power control updating rate. This paper also proposes a new predictive power control algorithm with better performance in terms of system capacity than the conventional and adaptive step size algorithms. An increase in system capacity as high as 22% compared with the conventional algorithm can be achieved depending on the mobile velocityIEEE Transactions on Vehicular Technology 06/1999; · 2.06 Impact Factor
IEICE TRANS. COMMUN., VOL.E86–B, NO.5 MAY 2003
A New User Mobility Based Adaptive Power Control in
HyeJeong LEE†a), Student Member and Dong-Ho CHO†b), Regular Member
scheme for wireless mobile communication systems using an
adaptive step size. The proposed scheme selects the basic power
control step size by considering the speed of the mobile station
and a variable step size by using instantaneous companding logic
based on power control command bit patterns. We show its im-
proved performance in view of the standard deviation of received
power at the base station in consideration of channel BER.
power control, adaptive step size, mobile velocity
We propose a new closed-loop power control
Code Division Multiple Access (CDMA) technology of-
fers a significant spectral efficiency compared to other
fades against interference from other users and is weak-
ened by the near-far problem.
that power control, by maintaining the average received
power at the base station (BS) at a constant level, can
reduce shadowing, near-far problem and multipath fad-
ing . Power control is performed by using both open-
loop power control (OLPC) and closed-loop power con-
trol (CLPC). When the correlation between the for-
ward link and the reverse link is small, it is impossible
to achieve precise power control effect by OLPC only.
Therefore, an accurate and efficient CLPC scheme is
required to increase system capacity and to minimize
the transmitted power of the mobile station (MS).
The standard in IS-95 uses a fixed-step-size, closed-
loop power control (FCLPC) scheme. In an FCLPC,
the BS measures the received power from the MS, com-
pares it with a target power, and sends a power con-
trol command bit to the MS. A logical 0 command bit
is generated if the received power is greater than the
target power; otherwise, a logical 1 command bit is
generated. When the MS receives the power control
command, it adjusts its transmitted power by ±1dB.
The FCLPC scheme is simple and performs effectively
for a low-speed mobile, but does not work well for a
high-speed mobile. Moreover, its performance varies
according to vehicular speed, propagation channel char-
However, CDMA system
It has been shown
Manuscript received May 27, 2002.
Manuscript revised September 19, 2002.
†The authors are with the Communication and Infor-
mation Systems Lab., Department of Electrical Engineering
and Computer Science, Korea Advanced Institute of Science
and Technology (KAIST), Daejeon 305-701, Korea.
acteristics, and power control step size at the MS.
Many researches on adaptive step-size, closed-loop
power control (ACLPC) scheme have been done. In
, the seven previous power control command bits are
stored at the MS and used to decide the power con-
trol step size to compensate for fading. However, this
scheme uses too many bits, and thus, in addition to
overcompensating at the end of the deep fade, it does
not work well for a high-speed mobile with a rapidly
varying channel. A single-bit, adaptive-step-size closed-
loop power control (S-ACLPC) scheme is proposed in
, in which the adaptive power control step size is in-
creased or decreased by factor K, in accordance with a
comparison between the present and the previous power
control command bits. It also applies the same power
control scheme to the entire mobile, resulting in poor
performance at high speeds.
In this letter, a new mobility based, adaptive
closed-loop power control (M-ACLPC) scheme is pre-
sented, which selects the power control step size ac-
cording to mobile speed and power control command
bit patterns. Compared with previous CLPC schemes,
the new scheme performs better when mobile speed and
power control command error vary. We compare the
performance of our proposed M-ACLPC scheme with
that of the FCLPC scheme and the S-ACLPC scheme
2.The Mobility Based Adaptive Closed-Loop
Power Control (M-ACLPC) Scheme
Our proposed scheme is intended mainly to track the
target power quickly and to minimize power-control er-
ror. Figure 1 contains a block diagram of this scheme.
When an MS receives a power control command bit
from the BS, the MS decides the power control step
size and adjusts its transmitted power. If the MS re-
ceives any identical successive power control command
bits (1’s or 0’s), the received power at the BS would be
much smaller or larger than the target power. If the
received power at the BS is approximately the same
as the target power, the power control command bit
changes alternatively. So, if we monitor the power con-
trol bit patterns, we can predict the amount of power
control error at the BS.
To obtain the performance improvements of our
power control scheme, a current power control com-
model with adaptive step-size.
Block diagram of a proposed closed-loop power control
adaptive step size.
The multiplication factor used for determining
mand bit (R0) and two previous power control com-
mand bits (R1,R2) are stored at the shift register of
the MS and are used to decide the power control step
size. The power control step size at time t is given by
∆(t) = ∆b× f(R0,R1,R2)
where ∆bis the basic step size determined by the mo-
bile speed, and f(R0,R1,R2), which depends on the
present and two previous power control command bits,
is a multiplication factor determined by the logic rule
shown in Table 1. The rule to configure the value of
multiplication factor f is similar to the instantaneous
companding logic of voice coding . The value of f is
set to the highest value 1.5 in the deep fading period
in case of identical successive power control commands
occurring. On the contrary, the value of f is assigned
to the lowest value 0.66 (the reciprocal of 1.5) when the
channel varies slowly.
Then, the transmitted power of the MS at time t,
P(t) is given by
P(t) = P(t − Tp) + sign(R0)∆(t)
where Tp is the power control period and the sign of
the current power control command bit, sign(R0), rep-
resents whether the current transmitted power is larger
or smaller than the target power.
If a small power control step size is used, we cannot
compensate for the severe and rapidly varying power
control error for a high-speed mobile. However, a large
power control step size may overcompensate the chan-
nel fading of a low-speed mobile. Therefore, we divide
the mobile velocity into three groups: pedestrian group,
low speed group, and the group above moderate speed.
A basic power control step size ∆bis differently assigned
to each group. Since too large power control step size
may cause much interference to other users, we limit
the maximum value of basic power control step size ∆b
to 3dB and set the minimum value of ∆bto standard
1dB. The speed of the MS can be estimated using GPS
(Global Positioning System), MEMS (MicroElectroMe-
chanical Systems), the received signal of the MS and
cell-sojourn time . Because we have no need to know
the exact speed of the MS, we can implement the pro-
posed power control scheme using a simple method such
as measurement of cell-sojourn time. How to estimate
the speed of MS is another research issue addressed in
 and is outside the scope of this paper.
3. Simulation and Results
It is well known that it is impossible to evaluate the
performance of a CDMA system with closed-loop power
control without oversimplification, and it is difficult as
well to derive the amount of power control error. There-
fore, we investigate through computer simulation the
performance of the CLPC schemes we have discussed.
We consider three CLPC algorithms, including the con-
ventional FCLPC scheme, the S-ACLPC scheme  and
our M-ACLPC scheme.
In this simulation, we consider only an uplink
channel experiencing large signal variation and fast fad-
ing, and we choose a single path Rayleigh channel im-
plemented by the Jakes’ Method . We do not con-
sider slow fading such as shadowing because it varies
slowly enough to be compensated for by OLPC only. In
general, the performance of the power-controlled sys-
tem is affected by the background noise of the chan-
nel. However, since our objective is to show the per-
formance improvements of the proposed power control
scheme according to the user mobility, we do not con-
sider background channel noise for convenience. A car-
rier frequency of 2 GHz is used and the power control
command rate is 1500 times per second (1.5kHz). The
MS moves with a uniformly distributed velocity from
0km/h to 90km/h.
The power control step size at the MS is given
by (1).We use different basic step size ∆b accord-
ing to the speed of the MS. ∆b for the pedestrian
group [0–10km/h] is 1dB, ∆bfor the low-speed group
[10–30km/h] is 2dB, and ∆b for the group above a
moderate-speed range [30–90km/h] is 3dB. The re-
ceived power at the BS is averaged over 0.667ms. We
assume that there is no antenna diversity at the BS.
The performance of the CLPC schemes is shown by the
standard deviation of the received power at the BS.
IEICE TRANS. COMMUN., VOL.E86–B, NO.5 MAY 2003
(MS speed: 20km/h)
Received signal statistics at the BS for FCLPC scheme.
scheme. (MS speed: 20km/h)
Received signal statistics at the BS for M-ACLPC
3.2Results and Discussion
For an MS speed of 20km/h, the received power statis-
tics at the BS for the FCLPC and M-ACLPC schemes
are shown in Figs.2 and 3. In Fig.2, the MS adjusts its
transmitted power by ±1dB without considering the
amount of power control error at the BS. However, in
Fig.3, it is clear that the MS alters the power control
step size to compensate for the difference between the
received power at the BS and the target power. When
a channel varies slowly, the adaptive step size is small
and the variation of Tx power at the MS is also small,
but when deep fading occurs, the adaptive step size
becomes larger and Tx power at the MS rises at a com-
mensurate rate. The results show that our M-ACLPC
scheme adapts the power-control step size to variations
in channel fading and has a smaller standard deviation
of Rx power at the BS than the FCLPC scheme.
Figure 4 shows the standard deviation of Rx power
at the BS with respect to various mobile speeds when
different power control schemes. (no channel error)
Standard deviation of received power at the BS for
there is no channel error for the power control com-
mand. As mobile speed increases, channel fading varies
more rapidly, and power control error also increases for
all three power control schemes. However, among the
three schemes across the full range of mobile velocity
and especially at a high mobile speed, it can be shown
that our M-ACLPC scheme has the smallest standard
deviation of Rx power at the BS.
If we define the system capacity as the maximum
number of users per cell such that the failure probability
is less than 1%, we can calculate the system capacity
easily by using well-known Fenton-Wilkinson method
. Here, the failure probability is the probability that
the BER at the BS is greater than 10−3. If the prob-
ability distribution function (pdf) of the each mobile’s
received power at the BS due to imperfect power con-
trol is assumed to be log-normal, the pdf of the to-
tal received interference power for n users follows ap-
proximately log-normal distribution with the logarith-
mic mean and logarithmic variance which can easily
determined from the mean and variance of the received
power at the BS .From the result of Fig.4, the
average standard deviations of the power control er-
ror at the BS are 3.724dB, 3.326dB and 3.165dB for
the FCLPC scheme, S-ACLPC scheme and M-ACLPC
scheme, respectively. Using these values, we can com-
pare the system capacity in consideration of the multi-
ple access interference by the Fenton’s method. The M-
ACLPC scheme has average capacity gain of 15% and
5.5% respectively compared with the FCLPC scheme
and the S-ACLPC scheme.
Figure 5 shows the standard deviation of the re-
ceived power at the BS when we apply 5% power control
command error. For the three power control schemes,
power control error at the BS increases somewhat in
comparison to Fig.4 in which there is no channel error.
The performance of the S-ACLPC scheme, in particu-
lar, declines severely at a high mobile speed. However,
the M-ACLPC scheme works relatively well for all mo-
ferent power control schemes. (5% power control command error)
Standard deviation of received power at the BS for dif-
bile speed ranges. We can reduce power control error
at the BS by about 0.5dB when there is slight power
control command error.
In this letter, we proposed a new adaptive-step-size,
closed-loop power control scheme that uses different ba-
sic power control step sizes based on mobile velocity.
This scheme is simple to implement and works well for
all mobile velocity ranges compared with other power
control schemes. Our M-ACLPC scheme is effective in
minimizing power control error at the BS when the MSs
are moving at various speeds. Also, we showed the in-
crease of system capacity in case that multiple access
interferences are considered by using Fenton-Wilkinson
method. It is further work area to suggest power con-
trol scheme that works well at very high mobile speed
such as 250km/h.
 S. Ariyavisitakul and L.F. Chang, “Signal and interference
statistics of a CDMA system with feedback power control,”
IEEE Trans. Commun., vol.41, no.11, pp.1626–1634, Nov.
 C.-C. Lee and R. Steele, “Closed-loop power control in
CDMA systems,” Proc. IEE, vol.143, no.4, pp.231–239,
 J.H. Kim, S.J. Lee, Y.W. Kim, M.Y. Chung, and D.K.
Sung, “Performance of single-bit adaptive step-size power
control scheme in DS-CDMA systems,” IEICE Trans. Com-
mun., vol.E81-B, no.7, pp.1548–1552, July 1998.
 C.K. Un, H.S. Lee, and J.S. Song, “Hybrid companding
delta modulation,” IEEE Trans. Commun., vol.29, no.9,
pp.1337–1344, Sept. 1981.
 D.J. Lee, “Resource management based on speed estimation
in mobile communication systems,” Doctorial Dissertation,
KAIST, May 2000.
 W.C. Jakes, Microwave mobile communications, Wiley,
New York, 1974.
 M.L. Sim, E. Gunawan, B.H. Soong, and C.B. Soh, “Per-
formance study of close-loop power control algorithms for
a cellular CDMA system,” IEEE Trans. Commun., vol.48,
no.3, pp.911–921, May 1999.
 L.F. Fenton, “The sum of a log-normal probability distribu-
tion in scattered transmission systems,” IRE Trans., vol.C5-
8, pp.57–67, March 1960.
 R. Prasad, M.G. Jansen, and A. Kegel, “Capacity analysis
of a cellular direct sequence code division multiple access
system with imperfect power control,” IEICE Trans. Com-
mun., vol.E76-B, no.8, pp.894–905, Aug. 1993.
 M.G. Jansen and R. Prasad, “Capacity, throughput, and
delay analysis of a cellular DS CDMA system with im-
perfect power control and imperfect sectorization,” IEEE
Trans. Veh. Technol., vol.44, no.1, pp.67–75, Feb. 1995.