Research On Efficient Transmission Technique And Spectrum Sensing For Next Generation Wireless Communication Systems

Radio spectrum is the most important resource in wireless communication.The kernel of the continuous development of wireless communication is to improve the spectrum utilization.There are two ways to improve spectrum efficiency.One is to enhance the existing systems' capability to make them can transmit more information in the same bandwidth by utiling the more efficient transmission technique.The other is spectrum reutilization,which is the emergying technique that reuses the licensed but underutilized spectrum without causing harmful interference to licensed systems by spectrum sensing.Multiple input multiple output-orthogonal frequency division multiplexing(MIMO-OFDM) has already become the physical transmission technique of next mobile communication systems for its ability to significantly improve the spectrum efficienty of the 3^rd mobile communication systems whose core technique is code-division multiple access (CDMA).Cognitive radio has already become a research hotspot since it allows the unlicensed users to sense and opportunistically access the licensed spectrum bands which are not currently being utilized by the licensed users.Thses two techniques respectively use the above two ways to improve the spectrum efficiency.On the basis of international current research works,this thesis investigates some key theories and technologies for MIMO-OFDM systems and cognitive radio systems in depth,and presents some new contributions in the following topics:multiple access based on MIMO-OFDM, OFDM channel estimation,and spetrum sengsing for cognitive radio systems.On the topic of multiple access based on MIMO-OFDM,the recently proposed space-time-frequency spreading CDMA(STFS-CDMA) scheme is analyzed.Target for obtaining the full diversity,a spreading code design criterion is first derived for STFS-CDMA.Based on this criterion,two novel spreading codes:permutated double-orthogonal code(PDOC) and zero-padded rotary Fourier transform code(ZPRFC).are proposed.PDOC can obtain full space and frequency diversity when the number of users in the system is only one,but it cannot obtain full space and frequency diversity when the number of users is larger than one.ZPRFC can always obtain full space and frequency diversity no matter how many users in the system,but the number of supporting users of it is reduced.On the topic of OFDM channel estimation.a novel parametric channel estimation scheme is proposed.Comparing with non-parametric channel estimation schemes,parametric channel estimation schemes have better performance for a module to estimate the channel parameters is added to assist the channel estimation.However,this added module also makes the complexity of parametric channel estimation schemes is higher than that of non-parametric channel estimation schemes.The proposed scheme successfully solves the above problem.Through complexity analysis and computer simulation,it is shown that the proposed scheme not only has better performance,but also has lower complexity.On the topic of spectrum sensing for cognitive radio systems,two adaptive single-band sensing schemes are first proposed.Unlike the existing schemes where a decision is made after receiving all samples,these two schemes make a decision with the sequential arrival of samples. Once a decision is made,the sampling is sloped.Therefore,they can use less detection time to achieve the desired performance.Then,a class of multi-band sensing schemes is proposed to fight against the noise uncertainty in cognitive radio systems.Unlike the existing schemes which detect each spectrum segment in tarn,these schemes simultaneously detect multiple spectrum segments. Among these schemes,two schemes are designed based on the information theoretical criterion, and the other scheme is designed based on the generalized likelihood ratio tests.Throught theoretical analysis and computer simulation,it is shown that these schemes can fight against the noise uncertainty effectively.