ABSTRACT
Amplifier nonlinearities affect drastically the performance of Orthogonal Frequency Domain Multiplexing OFDM systems. In this paper, the effects of amplifier nonlinearities is evaluated in OFDM with Space-Division-Multiple-Access (SDMA).We showed that nonlinear power amplifier reduced dramatically the performance of multiple-antennas OFDM systems.
1. INTRODUCTION
The introduction of an array antenna system in the base station of a mobile communication system increases the spectral efficiency, reducing the co-channel interference and increasing the reuse factor of the systems. It helps to increase the numbers of users placed in the same spectral portion. Space-Division-Multiple-Access (SDMA) or spatial multiplexing is a new option that can help to increase the
spectral efficiency in wireless communication systems. In SDMA, different users using a single antenna element are separated at the base-station, which have P antenna elements, with the aid of user-specific spatial signature, which is constituted by the channel transfer factors between the users with a single transmit antenna and the different receiver antenna elements at the Base Station (BS). The spatial signature generated by the channel over the transmitted signal acts like the spreading code in a conventional CDMA system.
Multiple access is a significant issue in OFDM systems. OFDM, that provide orthogonal subcarriers, is a system in which the multiacces techniques can be applied directly. The subcarriers can be divided into sub-channels and assigned to different users. In this case, only one carrier is assigned to each user preventing multiple-access interference among users. OFDM divides the frequency band in multiple carriers assigning to each users the best subchannels. The advantage of its application in multiuser cases is that carriers with high attenuation for some user can be less attenuated or high Signal to Noise Ratio SNR carrier to other users, due to the spatial distribution of the users, maximizing the performance of the system. Allocation methods considering the spatial signature generated by the MIMO structure eliminates the restriction that only one user is allowed per carrier. In this case, more complex receiver structures will be needed because multiuser detection techniques must be implemented in order to reach an suitable performance.
Supported by the Programme ALβ AN, European Union Programme
of High Level Scholarships for Latin America. Identification Number:
E03D19254AR
Fernando Gregorio and Timo Laakso are with the Smart and Novel Radios
Research Group (SMARAD) Center of Excellence.
In OFDM systems, the combination of different signals with different phase and frequency give a large Peak-to- Average Ratio PAR, which results in severe clipping effects if the compose signal is amplified by a power amplifier, which have nonlinear transfer function. In real applications, power amplifiers are used in the mobiles and base stations in order to transmit appropriate power levels. The high PAR levels of OFDM signals forces to the utilization of linear power amplifiers. However, linear amplifiers have low power efficiency which is problematic considering that battery life is a critical resource in mobile systems.
Several options appear in the literature related with OFDM systems and nonlinearities. PAR reduction techniques using mapping or coding [1], Low PAR OFDM implementations like Clustered OFDM [2], allocation methods minimizing the Intermodulation Products IMP [3] and beamforming design adding PAR constraints [4] are the tools to combat nonlinearities used in the transmitter. Other interesting approach is the application of multiuser detection and clipping noise mitigation techniques in the receiver [5]. Multiuser detection techniques known from Code-Division-Multiple-Access (CDMA) can be applied in SDMA-OFDM transceivers. In this paper, the effect of power amplifier nonlinearities is evaluated in SDMA-OFDM systems. Section 2 introduces power amplifier models. The derivation of clipping noise and distortion and their effect over the bit error rate performance is presented in Section 3. Simulation results are included in Section 4. Finally, the conclusions are drawn.
F. H. Gregorio and T. I. Laakso
Helsinki University of Technology, Signal Processing Laboratory, P.O. Box 3000, FIN-02015 HUT, Finland.
fernando.gregorio@hut.fi
No comments:
Post a Comment