Padam L. Kafle

Spatial correlation and capacity measurements for wideband MIMO channels in indoor office environment

This paper describes a broadband multiple input, multiple output (MIMO) channel characterization platform and capacity measurement results in indoor office environment. The MIMO testbed has been designed for broadband MIMO channel sounding, capacity measurements and for characterizing the directional-multipaths of the radio propagation channel. The MIMO channel data have been collected in 5 GHz band inside a modern office environment. Capacity results from these experiments are discussed for different propagation conditions, including non line-of-sight (NLOS) and LOS propagations with various spacings between array elements. Spatial correlations are analyzed from the measured data and a frequency selective MIMO channel model based on the correlation statistics is validated. Post-processing of the measured data with a sequential ESPRIT (Estimation of Signal Parameters via Rotational Invariance Techniques) algorithm is utilized to extract the directions of departures and arrivals of multipath components at the transmitting and receiving arrays, respectively. A method of obtaining broadband MIMO capacity, indirectly, from a single directional measurement of MIMO propagation channel is also proposed and verified from the measurement results.

Spatial correlation measurements for broadband MIMO wireless channels

In this paper, we propose a method to determine the transmit and receive side spatial correlations for an arbitrary antenna spacing in the arrays from a directional MIMO channel measurement. A broadband MIMO channel measurement platform has been developed, which is used to obtain the angles of arrival (AOA) and angles of departure (AOD) of resolvable multipath components (MPC). The AOAs and AODs are estimated from the measured MIMO channel data by using a sequential unitary ESPRIT algorithm. The measured AOAs, AODs and corresponding powers of the MPCs are used to compute the power angle spectrum for the transmitter and receiver sides of the MIMO propagation channel. The spatial correlation matrices and MIMO capacity can then be obtained from these directional measurements. Capacity results from the correlation measurements using a stochastic broadband MIMO channel model are observed to closely match the directly measured MIMO capacity for different propagation conditions and antenna spacings.

Performance of turbo coded multicarrier CDMA with iterative multiuser detection and decoding

The performance of a turbo coded MC-CDMA uplink with iterative multiuser detection is evaluated. Optimal likelihood ratios are computed based on the maximum a posteriori criterion using the multiuser signal structure and a priori information of code bits of all other users available from channel decoders. Combining turbo coding and multiuser detection in MC-CDMA has shown performance close to the single user bound with a few iterations of joint detection and decoding. Additionally, a suboptimal approach using soft interference cancellation is also implemented for the uplink transmission and compared with the conventional methods. Simulation results are presented in a Rayleigh multipath fading environment for a typical data transmission scheme.

Capacity of MIMO–OFDM systems in spatially correlated indoor fading channels

Multiple-input multiple-output (MIMO) wireless systems provide significantly higher capacity compared to the conventional single antenna systems in rich scattering environments. However, the real capacity advantage in a specific channel environment strongly depends on its propagation characteristics. In this paper, we analyze the capacity of an OFDM based spatial multiplexing system for indoor fading channels. We model the channel between the transmitter and receiver arrays by considering two indoor propagation models, the extended Saleh-Valenzuela (ESV) model and the geometrically based single bounce elliptical model (GBSBEM). Spatial correlation matrices for the realizations of the ESV and GBSBEM are obtained for the wideband MIMO fading channels, which are then used for evaluation of the mutual information and outage capacity using Monte Carlo simulations. The behavior of capacity versus outage is investigated for different physical parameters and antenna geometries. Capacity results are discussed here with similar power delay profiles, antenna spacing and angle spreads for uniform linear and circular arrays.

Capacity of MIMO OFDM systems in spatially correlated indoor fading channels

Multiple-input multiple-output (MIMO) wireless systems provide significantly higher capacity compared to the conventional single antenna systems in rich scattering environments. However, the real capacity advantage in a specific channel environment strongly depends on its propagation characteristics. In this paper, we analyze the capacity of an OFDM based spatial multiplexing system for indoor fading channels. We model the channel between the transmitter and receiver arrays by considering two indoor propagation models, the extended Saleh-Valenzuela (ESV) model and the geometrically based single bounce elliptical model (GBSBEM). Spatial correlation matrices for the realizations of the ESV and GBSBEM are obtained for the wideband MIMO fading channels, which are then used for evaluation of the mutual information and outage capacity using Monte Carlo simulations. The behavior of capacity versus outage is investigated for different physical parameters and antenna geometries. Capacity results are discussed here with similar power delay profiles, antenna spacing and angle spreads for uniform linear and circular arrays.

Geometrically Based Broadband MIMO Channel Model With Tap-Gain Correlation

This paper proposes a broadband multiple-input-multiple-output (MIMO) channel model that includes the effect of tap-gain correlation based on the Geometrically Based Single Bounce Model (GBSBM). The GBSBM is extended to a broadband channel model by dividing a uniform scattering ring into a contour of constant delays. The geometry of the constant delay contours determines the amount of tap-gain correlation. A statistical model of the delays is developed based on the geometry of the contours of the constant delays. Then, the correlation among the delays is determined. The tap-gain correlation is analyzed for various values of resolvable delays and maximum delays. It is observed that the magnitude of tap-gain correlation decreases as the amount of resolvable delay increases but increases with an increase in the amount of maximum delay. The proposed tap-gain correlation model is used to investigate the outage capacity of the broadband MIMO system. The outage capacity results are analyzed in the presence of both tap-gain and spatial correlations.

On the performance of MC-CDMA with interleaved concatenated coding and interference cancellation for high-rate data transmission

A coded wideband MC-CDMA system is considered for high-bit-rate data transmission in a typical outdoor environment with Rayleigh multipath fading. First, two powerful interleaved concatenated coding techniques, turbo and serially concatenated convolutional codes (SCCC) codes are applied in the downlink system and important conclusions are derived on their relative performance behavior in this type of system. Simulation results indicate some of the relative merits of SCCC on achieving very low error rates, with relative mitigation of error floor behavior at the typical BER range of interest for wireless data applications. Next, the more critical uplink system is considered, which requires a multiuser detection technique in combination with powerful coding to achieve similar performance. Successive interference cancellation with SCCC coding is presented, with computation of optimal log-likelihood ratios required for this type of interference cancellation followed by SCCC decoding. Application of SCCC coding and interference cancellation has improved the system performance significantly even in the fully loaded case. Based on these results, the combination of SCCC coding and successive cancellation is found to be attractive for high-rate data transmission in the uplink MC-CDMA.

An iterative MMSE-decision feedback multiuser detector for space-time coded multicarrier CDMA system

Performance of wireless communication systems is primarily limited by multipath fading effect of the radio channel and multiple-access interference caused due to other users in the system. Two promising techniques to counter these problems are space-time coding and iterative multiuser detection. In this paper, an iterative receiver structure for space-time block coded MC-CDMA system employing multiple transmit and receive antennas is presented. The receiver is based on the minimum mean square error decision feedback multiuser detection. The log-likelihood ratios computed by the soft multiuser detector are refined successively during the iterative process by using the extrinsic information available through decoding of all users. Turbo code is used for channel coding. Simulation results in a frequency selective Rayleigh fading environment are presented to verify the performance of the proposed scheme.

Effect of tap gain and spatial correlations to wideband capacity and experimental validation

In this paper, the capacity of wideband multiple input, multiple output (MIMO) system is evaluated by using channel measurement techniques. The tap-gain and spatial correlations of the MIMO system are also analyzed. A MIMO-channel characterization platform has been developed for channel sounding in a frequency selective fading environment. Capacity results from these experiments are discussed in an indoor office environment. The measurement data are analyzed to obtain the correlation statistics as well as their effect to the MIMO capacity. It is shown that the MIMO capacity in the wideband case depends on all the correlation effects. It is thus important to quantify the spatial correlations at transmit and receive sides as well as the tap-gain correlations to accurately estimate the capacity. A unified wideband MIMO channel model that can be used for simulations is validated based in our experiments

A low-complexity iterative multiuser receiver for turbo-coded MC-CDMA system

A low-complexity iterative multiuser receiver is proposed using groupwise maximum likelihood sequence estimation (MLSE) combined with a modified interference cancellation scheme in a turbo coded MC-CDMA system. It is based on grouping the active users according to their signal strengths and computing the log-likelihood ratios by MLSE among each group of users at the first iteration. For subsequent iterations, a modified interference cancellation scheme that can benefit from code extrinsic information available through decoding is used, for lower complexity. Simulation results are presented in a Rayleigh multipath fading environment for a typical data transmission scheme. Comparisons are made with iterative receivers using the maximum a posteriori (MAP) criterion and soft interference cancellation. Performance very close to that of a MAP based iterative receiver is achieved by this novel scheme within a few iterations, with much lower computational cost. This approach also performs significantly well under near-far conditions.