Use this URL to cite or link to this record in EThOS:
Title: OFDM turbo transceivers
Author: Xu, Lei
ISNI:       0000 0001 3573 8986
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2008
Availability of Full Text:
Access from EThOS:
Access from Institution:
In this thesis a variety of different techniques designed for the sake of improving the attainable efficiency of wireless system are considered in the context of Orthogonal Frequency Division Multiplexing (OFDM). More explicitly, powerful turbo transceivers invoking Sphere Packing (SP) modulation are designed using EXtrinsic Information Transfer (EXIT) charts for both single-user and multi-user transmission. In order to provide an attractive suite of single-user OFDM transceivers, we contrive a SP-aided OFDM scheme for maximizing the achievable coding advantage. The corresponding capacity equation shows that the SP-OFDM scheme exhibits a higher capacity than its counterpart dispensing with SP. Furthermore, a carefully designed Sphere Packing symbol To Sub-Carrier Mapping (SPTSCM) scheme is contrived, which results in an improved Discrete Input Continuous Output Memoryless Channel (DCMC) capacity. This is achieved by reducing the correlation between the multiple OFDM sub-carriers conveying a specific SP symbol. In order to enhance the attainable BER performance of the SP-OFDM scheme, iterative detection is invoked for exchanging extrinsic information between the SP symbol-to-bit demapper and the channel decoder. In contrast to the classic SP signal construction, sophisticated SP schemes are designed with the aid of EXIT charts for the sake of improving the convergence behaviour of the iteratively detected schemes. Explicitly, an approximately 0.15bit/s/Hz DCMC capacity improvement is achieved and the proposed iterative detection aided SP-OFDM scheme becomes capable of performing within 2.1dB of the maximum achievable rate limit obtained using EXIT charts at BER = 10−4. Some of the proposed advances are then extended to a multi-user Space Division Multiple Access (SDMA) OFDM scheme. Novel turbo MUDs are designed and investigated in terms of their complexity, EXIT characteristics and BER performance. More explicitly, the linear turbo MUDs considered include the Complex-valued Minimum Mean Square Error (SIC-CMMSE) scheme, the Soft Interference Cancellation based Real-valued Minimum Mean Square Error (SIC-RMMSE) arrangement and the Soft Interference Cancellation based Minimum Bit-Error Rate (SIC-MBER) MUD schemes. The SIC-RMMSE is only applicable in the context of BPSK modulated schemes. A Reduced-complexity Minimum Bit-Error Rate (RMBER) is also designed for the sake of arriving at a complexity reduction by slightly compromising the attainable performance in comparison to the SIC-MBER MUD. Furthermore, a hybrid scheme was designed to eliminate the performance degradation imposed by the RMBER MUD. In contrast to above-mentioned linear schemes, Bayesian turbo MUD and the K-best iterative Sphere Decoder (SD) belong to the family of non-linear MUDs. Then a general complexityreduction technique was contrived in order to reduce the complexity imposed by the turbo MUDs, which was referred to as the A-priori-LLR-Threshold (ALT) aided MUD algorithm, which was introduced for the sake of striking an attractive trade-off between the attainable complexity-reduction and the performance degradation imposed. Finally, the novel framework of Transmit Domain Processing based Detectors (TDPD) was contrived for a generalized multi-access or multiplexing based channel model in order to circumvent the exponentially increasing complexity imposed by the Maximum Likelihood (ML) MUD as a function of both the number of antennas and the number of bits per symbol. Explicitly, the TDPD may be separated into three basic steps including the MUD’s Search Center (SC) calculation, detection candidate list generation and LLR output calculation. A range of different schemes were considered for each of the three steps, and the SIC-CMMSE SC aided hybrid K-best subset combination based iterative TDPD was capable of achieving a similar BER performance to the optimal ML detector at a significantly reduced-complexity.
Supervisor: Hanzo, Lajos ; Chen, Sheng Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: TK Electrical engineering. Electronics Nuclear engineering