Characterization Of Wideband-and Wideband-Virtual MIMO Radio Channels

The tremendous use of wireless systems in an indoor and outdoor areas has pushed scientists and engineers to develop new technologies capable of improving transmission data rate and system capacities. Since the available frequency spectrum assigned to wireless communication is limited, the multi-input multi-output (MIMO) or virtual multi-input multi-output (V-MIMO) are therefore to be applied in new wireless systems to improve spectral efficiency and enhance system capacities.In this thesis, based on an indoor to outdoor measurement data, two methods of noise cut are investigated such as dynamic and fixed level noise cut. It is found that the dynamic denoising is preferred if the noise floors and the DNRs (Dynamic Ranges) can be clearly observed and derived from the measured IRs (Impulse Responses). The PDPs (Power Delay Profiles) are almost overlapping when the fixed level noise cut is applied, however, obvious changes can be observed by using dynamic noise cut. The Ricean K-factors are not sensitive to the ways of noise cut even if the noise is not cut at all, the statistical values of the K-factors are still comparable to those by implementing noise cut.A detailed investigation of wideband channel models and channel characterizations based on channel measurements carried out in urban microcells at 5.25 and 2.25 GHz with 100 MHz bandwidth have been performed. The differences of the channel models and parameters as well as the correlations of the channel parameters are studied between 2.25 and 5.25 GHz with 100 MHz bandwidth when measurements were performed in the exact same routes for urban microcells. It is found that the same channel parameters at the two frequency bands are independent and their behaviors can be completely different in the same route, but their coherence bandwidths are highly correlatedThe performance of V-MIMO is also investigated by using the measurements carried out at a modern office building and compared to the performance of conventional MIMO by both theoretical and measuremental studies. The results show that the V-MIMO in many cases is desirable for capacity improvement. The capacity in wireless communication is a dependent parameter. The capacity is affected by many parameters such as K-factor, root mean square (rms) delay spread (DS), spatial correlation, etc. In this thesis, the parameters such as K-factor, rms delay spread and spatial correlation to affect V-MIMO capacity are investigated. The capacity, K-factor, rms DS and spatial correlation are evaluated for all pairs of users whose channels were measured while standing.In this thesis, we also investigate a V-MIMO wireless system performance using precoding scheme at a base station (BS) or an access point. The idea behind is to validate V-MIMO performance over SU-MIMO using channel coding techniques. The precoding scheme that we consider is MIMO receiver such as zero-forcing (ZF), minimum mean square error (MMSE) and successive interference cancelation (SIC). Finally, joint transmit and receive beamforming have been investigated.With precoding scheme at BS, the SIC achieves higher capacity when compared to the MMSE and ZF. Considering SIC precoding scheme at BS, the maximum achievable rate is seen with V-MIMO multi-user when compared to SU-MIMO.With joint transmit and receive beamforming, two modes of transmissions are investigated such as maximal eigenmode transmission (single beam) and eigenmode transmission (multiple beams). We can distinguish two kinds of eigenmode transmissions such as eigenmode transmission with water-filling (WF) or with equal power allocation. The eigenmode transmission with WF achieves capacity over all eigenmode transmissions. The maximum achievable rate is seen with the V-MIMO system when compared to the conventional MIMO system using transmit and receive beamforming.