Seminar Topics & Project Ideas On Computer Science Electronics Electrical Mechanical Engineering Civil MBA Medicine Nursing Science Physics Mathematics Chemistry ppt pdf doc presentation downloads and Abstract

Full Version: Multistage Interference Cancellation with Decision Directed Channel
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Multistage Interference Cancellation with Decision Directed Channel
Estimation in Multirate DS/CDMA on a Mobile Radio Channel


[attachment=25001]


INTRODUCTION

A possible multiple access technique for digital cellular mobile
communication systems is direct-sequence code division multiple
access (DS/CDMA). It is a flexible multiple access method,
suitable for transmission of variable data rates [1], which makes
it a good prospect for next generation systems.
Two of the most important factors that limit the capacity of a
multiuser DS/CDMA system are signal interference between
users and possibly large variations in the power of the base station’s
received signal from different users. The former is referred
to as multiple access interference (MAI) and the latter as the
near-far effect. The power variations that causes the near-far
effect is due to the difference in distance between the users and
the base station, as well as fading and shadowing. One way to
combat this is to use stringent power control [2]. Another
approach would be to use more sophisticated receivers which are
near-far resistant [3]. Because of MAI’s contribution to the nearfar
problem and its limiting of total system capacity, a lot of
attention has recently been given to the subject of multiuser
detectors that have the prospect of both mitigating the near-far
problem and cancelling the MAI [3]-[10].



MULTISTAGE INTERFERENCE CANCELLATION IN A
MULTIPATH ENVIRONMENT


A multistage hybrid IC [13] employs a combination of NDDIC
and DDIC together with a modified RAKE. This is illustrated in
Fig. 1. In a multipath environment we have to employ a RAKE
receiver to take advantage of the diversity of the channel. The
RAKE receiver demands knowledge of the channel parameters



DECISION DIRECTED CHANNEL ESTIMATION

A decision directed channel estimation scheme is proposed in
[14] and [15] for DQPSK and DQAM, respectively. In this paper
we consider only DQPSK. After the NDDIC stages most of the
interference, on each received path, from other paths and other
users is cancelled. Assuming that all the interference is removed,
the received sample for each symbol of each path is modelled as
the product of the transmitted symbol and the complex channel
gain (fading) plus additive white Gaussian noise.


NUMERICAL RESULTS

The simulation results from a parallel channel system with 15
QPSK users and two parallel channels per user are shown in Fig.
4. The channel is a Rayleigh fading channel with two paths and
the detector is a hybrid IC with two NDDIC stages and one stage
with RAKE and DDIC. Orthogonal Gold sequences of length
128 are used. The results are from a system with known channel
parameters and two systems where the channel is estimated. The
estimation is performed using a FIR filter of length 5 and 11. The
graph depicts that the loss caused by estimation with a filter of
length 5 is only about 1 dB compared to the results when using
known channel parameters.


DISCUSSION and CONCLUSIONS
This paper presents a novel combination of a multistage
hybrid IC scheme for multipath environments and a decision
directed minimum mean square error channel estimator, which is
evaluated together with a multiple data rate scheme denoted parallel
channels. The hybrid IC includes a combination of nondecision
directed and decision directed IC combined with a modified
RAKE receiver. The decision directed minimum mean
square error estimator is implemented by an F tap FIR filter on
each received path, whose input is the last received samples on
that particular path. The coefficients in the filter is chosen to minimize
the mean square error between the true and estimated channel
gain. Fading correlation between paths of a user’s parallel
channels is used to improve the performance.