Research On Adaptive Transmission Technologies For Decode-and-forward Cooperative Systems

Cooperative communication is one of the key technologies for future wireless communication systems to mitigate multipath fading and provide high-data-rate services. By sharing antennas among neighboring nodes to transmit signals, cooperative communication not only expands network coverage through relaying information, but also forms a virtual multi-antenna array that brings spatial diversity gain. However, since half-duplex relays generally cost extra time or spectral resources to forward information, the application of cooperative communication is severely hampered as a result of its low spectral efficiency. By adopting adaptive transmission technologies, the available channel state information of cooperative communication system can be fully exploited to improve spectral efficiency. Compared to adaptive modulation in amplify-and-forward systems, the adaptation to channel variations in decode-and-forward(DF) systems can be more flexible, allowing the source and relay nodes to use different modulation levels to transmit signals, which can further improve spectral efficiency. This paper focuses on applications of adaptive transmission technologies to DF cooperative systems, including constellation schemes of transmit signals, diversity combining techniques at the receiver, relaying schemes and adaptive modulation strategies, which have certain application value in improving spectral efficiency for cooperative communication.To mitigate the low power efficiency problem of multilevel modulation(M-QAM) popularly used in adaptive modulation, a constellation scheme is designed for maximal ratio combiner based multi-relay selection DF systems with adaptive modulation. By adjusting the bit labeling sequence at the source and relay nodes with constellation schemes, bit reliability variations can be averaged over all transmission branches, leading to improvement in power efficiency of M-QAM as well as increase in diversity combining gain. To fully reflect the effect of constellation information on the performance of signal combining and provide theoretical foundation for constellation design, a new BER expression is derived for soft-bit maximal ratio combiner that is applicable to the combining of signals with different modulation levels. And a heuristic algorithm with low computational complexity is proposed to obtain the suboptimal solution to constellation design.The selection combining techniques at the receiver are studied for fixed DF systems using adaptive modulation, so as to alleviate the effect of error propagation and solve the combining problem of signals with different modulation levels. Compared to maximum likelihood detector and maximal ratio combiner, the low-complexity selection combiner simplifies the structure of receivers. The traditional scaled selection combiner(SSC), assuming i.i.d fading channels and the same modulation levels at the source and relay nodes, is modified for DF systems with adaptive modulation, assuming i.ni.d fading channels and different modulation levels. The different scale factors generated by the modified SSC for diversity branches reflect not only various source-relay channels reliabilities but also signal transmission reliabilities affected by different modulations levels. In addition, a slow adaptive modulation scheme is designed for SSC based DF systems to improve spectral efficiency.To weight diversity branches with more accurate scale factors when combining signals, an enhanced scaled selection combiner(ESSC) is proposed for single-relay fixed DF systems with adaptive modulation. Compared to the traditional SSC using average SNR of all channels, ESSC can fully exploit the instantaneous relay-to-destination SNR to generate scale factors. ESSC can further mitigate error propagation effect at the expense of slight increase in computation complexity. Besides, a power allocation scheme is proposed for ESSC based DF systems, and the slow and fast adaptive modulation schemes are designed.Three smart relaying schemes with different complexity are designed for selection combining based single-relay DF systems with adaptive modulation, including smart relaying based on discrete power control, continuous power control, and power allocation. Since smart relaying enables the relay to fully exploit the instantaneous source-to-relay SNR to optimize the transmit power of the source and relay nodes, the system power efficiency can be effectively improved and the error propagation problem is greatly alleviated. To improve spectral efficiency, a slow adaptive modulation scheme is designed for DF systems using smart relaying.