17-03-2014, 12:45 PM
Abstract--The challenging issue in multiuser MIMO-OFDM
system is to cancel multi-access interference (MAI) when
different users share the same subcarrier. Also, the presence of
MAI makes the resource allocation problem very complicated.
Hence it is essential to find the optimal transmit and receive
weight vectors, to reduce MAI and to adaptively assign
resources to meet the QoS requirements. In this paper, Block
Diagonalization (BD) with Waterfilling power allocation is
considered. This BD with Waterfilling power allocation
algorithm increases the system capacity with zero MAI. In
order to increase the system capacity further, BD with power
and subcarrier allocation is proposed and analyzed their
performance. The simulation results show that there is
significant improvement in system capacity also guaranteeing
zero MAI.
I. INTRODUCTION
Multiple-input multiple-output orthogonal frequency
division multiplexing (MIMO-OFDM) combines the simple
equalization of OFDM modulation with the capacity,
diversity, and array gain of MIMO communication.
Currently MIMO-OFDM is being considered in various
multiuser systems including high-speed wireless local area
networks and next generation cellular systems.
Communication in multiuser MIMO-OFDM systems
however, requires dealing with a variety of sources of
interference including multiple access interference, coantenna
interference which is the interference caused by the
signals from multiple transmit antennas of a given user being
received on the same receive antenna, and intercarrier
interference (ICI). Transceiver techniques that can flexibly
deal with interference are thus an important component of a
MIMO-OFDM system.
Dirty paper coding (DPC) pre-cancels the interference at
the transmitter using perfect channel state information (CSI)
[2-3], [6-7]. DPC is a theoretically optimal technique and
achieves maximum sum capacity but it is proven to be
difficult to implement in practice. Consequently, several
practical near-DPC techniques based on the concept of
precoding have been proposed that offer different tradeoffs
between complexity and performance.
One of the simplest approaches to cancel MAI is zeroforcing
(ZF) or minimum mean squared error (MMSE)
multiuser precoding technique. Here the inversion of the
multiuser matrix channel is computed and premultiplied with
the transmitted signal to cancel MAI [9]. However, finding
the inversion of the channel matrix is highly complicated.
An alternative to implementing DPC is block
diagonalization (BD), which supports multiple stream
transmission as well. The basic concept of BD consists of
using special transmit vectors that ensure zero interference
between users but do not completely invert the channel.
The resulting multiuser MIMO-OFDM channel matrix
has a block diagonal form thus each user can apply a
standard point-to-point MIMO-OFDM receiver. The power
allocation is done based on Waterfilling algorithm. In the
proposed method Block Diagonalization with power and
subcarrier allocation is done to further enhance the system
performance. The presented results depicts that there is
improvement in system capacity than the existing one.
The remainder of this paper is organized as follows. In
Section2 we describe the MIMO-OFDM multiuser system.
Section3 presents BD followed by our proposed BD with
Waterfilling power allocation & subcarrier allocation
method. Section4 presents performance results. Section5
concludes this paper.
system is to cancel multi-access interference (MAI) when
different users share the same subcarrier. Also, the presence of
MAI makes the resource allocation problem very complicated.
Hence it is essential to find the optimal transmit and receive
weight vectors, to reduce MAI and to adaptively assign
resources to meet the QoS requirements. In this paper, Block
Diagonalization (BD) with Waterfilling power allocation is
considered. This BD with Waterfilling power allocation
algorithm increases the system capacity with zero MAI. In
order to increase the system capacity further, BD with power
and subcarrier allocation is proposed and analyzed their
performance. The simulation results show that there is
significant improvement in system capacity also guaranteeing
zero MAI.
I. INTRODUCTION
Multiple-input multiple-output orthogonal frequency
division multiplexing (MIMO-OFDM) combines the simple
equalization of OFDM modulation with the capacity,
diversity, and array gain of MIMO communication.
Currently MIMO-OFDM is being considered in various
multiuser systems including high-speed wireless local area
networks and next generation cellular systems.
Communication in multiuser MIMO-OFDM systems
however, requires dealing with a variety of sources of
interference including multiple access interference, coantenna
interference which is the interference caused by the
signals from multiple transmit antennas of a given user being
received on the same receive antenna, and intercarrier
interference (ICI). Transceiver techniques that can flexibly
deal with interference are thus an important component of a
MIMO-OFDM system.
Dirty paper coding (DPC) pre-cancels the interference at
the transmitter using perfect channel state information (CSI)
[2-3], [6-7]. DPC is a theoretically optimal technique and
achieves maximum sum capacity but it is proven to be
difficult to implement in practice. Consequently, several
practical near-DPC techniques based on the concept of
precoding have been proposed that offer different tradeoffs
between complexity and performance.
One of the simplest approaches to cancel MAI is zeroforcing
(ZF) or minimum mean squared error (MMSE)
multiuser precoding technique. Here the inversion of the
multiuser matrix channel is computed and premultiplied with
the transmitted signal to cancel MAI [9]. However, finding
the inversion of the channel matrix is highly complicated.
An alternative to implementing DPC is block
diagonalization (BD), which supports multiple stream
transmission as well. The basic concept of BD consists of
using special transmit vectors that ensure zero interference
between users but do not completely invert the channel.
The resulting multiuser MIMO-OFDM channel matrix
has a block diagonal form thus each user can apply a
standard point-to-point MIMO-OFDM receiver. The power
allocation is done based on Waterfilling algorithm. In the
proposed method Block Diagonalization with power and
subcarrier allocation is done to further enhance the system
performance. The presented results depicts that there is
improvement in system capacity than the existing one.
The remainder of this paper is organized as follows. In
Section2 we describe the MIMO-OFDM multiuser system.
Section3 presents BD followed by our proposed BD with
Waterfilling power allocation & subcarrier allocation
method. Section4 presents performance results. Section5
concludes this paper.