Moon-Il Lee

Evolution of reference signals for LTE-advanced systems

3GPP LTE Release 10 standards (also known as LTE-Advanced) adopted some of the state-of-the-art radio access technologies that include carrier aggregation, eight-layer downlink spatial multiplexing, and four-layer uplink spatial multiplexing. For facilitating these enhancements, reference signals have significantly evolved in LTE-Advanced. This article examines underlying design principles of the LTE-Advanced reference signals. Specifically, newly introduced dedicated demodulation reference signals and channel state information reference signals for downlink and improvements of demodulation reference signals and sounding reference signals in uplink are discussed.

3D channel model extensions and characteristics study for future wireless systems

This work introduces a complete framework for three-dimensional (3D) channel extension based on two-dimensional (2D) channel models. In the 3D channel modeling, besides the azimuth angles, the multipath components of a receive-transmit pair are determined by elevation angles, correlations between elevation and azimuth, and the 3D configurations of a user equipments (UE) location and velocity. The 3D channel characteristics for an active antenna system (AAS) are studied in terms of aggregated array gain and channel covariance matrix. The 3D channel model is generated and examined through Monte Carlo simulations and a comparison of fixed vertical beamforming and adaptive vertical beamforming is presented. The comparison shows that adaptive vertical beamforming can potentially provide up to 8dB improvement in coupling loss. In addition, it is shown that for line-of-sight (LoS) the 3D channel covariance matrix can be decoupled into two independent covariance matrices with reduced sizes that effectively leads to reduced feedback overhead. For non-line-of-sight (NLoS), the correlation-based Kronecker model technique can be used to generate the 3D channel matrix with reduced complexity compared to the geometry-based technique.

Design of space-time codes achieving generalized optimal diversity

In this paper, we propose a generalized structure for constructing space-time codes (STCs), which achieves generalized optimum diversity (GOD). A GOD class of STCs can achieve the full diversity and an optimized coding gain with minimum delay for any number of transmit antennas (Nt) and any rate (R), if the rate is less than or equal to the rank of the channel matrix. The structure combines mutually exclusive R data symbols with complex weights at each transmit antenna and the same set of data symbols is repeated at every time slot in a code block, but is transmitted through different transmit antennas with different sets of complex weights. In addition, it does not need any restrictions to the complex weights, thus the code can be designed with the maximum degrees of freedom. Optimized codes based on the proposed structure are obtained under the well-known rank and determinant criteria by applying two design constraints in order to reduce the design complexity. Even using two constraints, however, the proposed structure has an unaffordable design complexity when the number of complex weights gets larger. Hence, we provide some guidelines for code design, which can further reduce the design complexity by examining the characteristics of the determinant of codeword difference matrices

An improved channel estimation scheme using polynomial-fitting and its weighted extension for an MC-CDMA/TDD uplink system with pre-equalization [mobile radio]

This paper deals with uplink channel estimation at a mobile transmitter for an MC-CDMA/TDD uplink system with pre-equalization. We present an enhanced version of a polynomial-fitting and extension (PF&E) scheme, which uses the idea of weighted extension. The scheme first approximates channel state information (CSI) estimates during some downlink slot interval to a polynomial of degree 2 in the least-squares sense, and then obtains the consecutive uplink CSI estimates by extending the approximated downlink CSI polynomial to the uplink slot interval, but with some temporal weighting corresponding to the maximum Doppler frequency. We perform computer simulations to analyze the overall bit error rate (BER) of the system under consideration, when the proposed scheme is used for uplink channel estimation. Extensive simulations show that the proposed scheme outperforms the PF&E scheme in terms of the overall BER under all the conditions considered.

Enhancing LTE-Advanced network performance using active antenna systems

This work exploits the benefits of adaptive downtilt and vertical sectorization schemes for Long Term Evolution Advanced (LTE-A) networks equipped with active antenna systems (AAS). We highlight how the additional control in the elevation domain (via AAS) enables use of adaptive downtilt and vertical sectorization techniques, thereby improving system spectrum efficiency. Our results, based on a full 3 dimensional (3D) channel, demonstrate that adaptive downtilt achieves up to 11% cell edge and 5% cell average spectrum efficiency gains when compared to a baseline system utilizing fixed downtilt, without the need for complex coordination among cells. In addition, vertical sectorization, especially high-order vertical sectorization utilizing multiple vertical beams, which increases spatial reuse of time and frequency resources, is shown to provide even higher performance gains.

Reevaluating cell wraparound techniques for 3D channel model based system-level simulations

Current wireless system-level simulation frameworks utilize cell wraparound in order to reduce boundary effects and provide an accurate evaluation of system performance. The current wraparound technique is purely distance based, where UEs always pick the closest cells, ignoring the radio propagation characteristics of the channel. While this wraparound methodology works well for a 2D channel model, it fails to account for the enhancements and spatial characteristics of the 3D channel model. In this paper we investigate the use of a radio distance based wraparound scheme for 3D channel models. This new wraparound methodology uses received downlink signal power to perform wraparound cell selection, allowing it to account for the radio propagation characteristics of the 3D channel model. We compare the performance of these two wraparound techniques with respect to coupling gain and other key metrics when considering a 3D channel model. Our results demonstrate the differences between the two schemes and highlight the improved accuracy offered by the radio distance based wraparound scheme.

Efficient adaptive vertical downtilt schemes in LTE-Advanced networks

This work introduces efficient schemes for adaptive vertical downtilt beamforming in Long Term Evolution Advanced (LTE-A) networks equipped with active antenna systems (AAS). A framework to determine cell-specific downtilt angles is proposed and solutions are provided with affordable complexities for practical applications. Instead of utilizing a single fixed downtilt, a cell-specific downtilt is obtained and adaptively adjusted across the system, which not only increases signal strength for desired users but also provides better control of inter-cell interference in the system. System level simulation shows that the proposed schemes achieve up to 10% cell edge and 5% cell average spectrum efficiency gain compared to the baseline system with fixed downtilt without involving complex coordination among cells.

A New STC Structure to Achieve Generalized Optimal Diversity with a Reduced Design Complexity

We propose a new space-time code (STC) structure that can achieve generalized optimal diversity (GOD) with a reduced design complexity, and also provide an increased coding gain. As a class of GOD STC structure, the new structure achieves the full diversity and an optimized coding gain with minimum delay irrespective of the number of transmit antennas (Nt) and the rate (R ), if the rate is less than or equal to the rank of the channel matrix. In the new structure, only R transmit antennas at a time slot are engaged in signal transmission, and a different set of R transmit antennas at a different time slot is used. Nt data symbols are combined together using complex weights for each antenna at a time slot. The same set of data symbols is repeated at all the remaining time slots in a code block, but is transmitted through different transmit antennas using different sets of combining weights at different time slots. In particular, the above structure eliminates cross-code product terms in the determinant of codeword difference matrices when R < Nt, thus increasing the coding gain. In addition, we propose a simplified code structure to facilitate the code design, in which a higher-rate GOD STC can be obtained by coupling R rate-one codes using different rotations. We apply two design constraints such as the power and orthogonality constraints, and provide some guidelines to reduce further the design complexity. In conclusion, the proposed structure provides a higher coding gain as compared with the original GOD structure [ 1 ] and the existing linear dispersion STCs (LD-STCs), when R < Nt and can reduce significantly the design complexity when the rate becomes higher and/or the number of transmit antennas gets larger.

Space-time block codes achieving generalized optimal diversity

We propose a generalized code structure for constructing space-time codes (STCs) that can achieve generalized optimum diversity (GOD) with minimum delay. As a GOD code, it can achieve the full diversity and a maximum coding gain for any number of transmit antennas ( t N ) and any rate ( R ), if the rate is less than or equal to the rank of the channel matrix. As a minimum delay code, the structure requires only t N time slots in order to achieve the full diversity. The proposed structure combines mutu- ally exclusive R data symbols with complex weights at each transmit antenna and the same set of data symbols is repeated at every time slot in a code block, but is transmitted through different transmit antennas with different sets of complex weights. In addi- tion, the structure does not need any restrictions to the complex weights, thus the code can be designed with the maximum degrees of freedom. The design of an optimum code would be done by finding the set of the complex weights so that the minimum abso- lute determinant of code word difference matrices could be maxi- mized. The optimum code could be easily designed so that two types of constraints, that is, the power constraints and the orthogo- nality constraint could be satisfied. The two constraints make the channel resource fully utilized when the channel state information (CSI) is not available at the transmitter and also reduce to the uni- tary constraint for capacity lossless code design in the case of t N R = . We give two STC design examples: (1) an STC with rate 2 for two transmit antennas and (2) an STC with rate 3 for three transmit antennas. We also provide some guidelines for code de- sign. Computer simulations are performed to demonstrate the per- formance of the proposed designs. From simulation results, we see that the proposed code structure and code designs can achieve GOD with minimum delay.

A multislot-interleaved turbo-coded MC-CDMA/TDD uplink system with pre-equalization using polynomial fitting and extension for uplink channel estimation

In this paper, we present a multislot-interleaved turbo-coded MC-CDMA/TDD uplink system with pre-equalization using polynomial fitting and extension (PF and E) for uplink channel estimation at a mobile transmitter. The PF and E scheme that extends the channel state information (CSI) polynomial approximated over some downlink interval to the uplink slot interval provides better uplink CSI estimates in a time-varying channel. To improve further the system performance, we propose a two-step multislot interleaving (TS-MSI) scheme that improves the decoding performance by fully randomizing burst errors near the end of the uplink slot interval due to the CSI inaccurateness when the PF and E scheme is used. In the scheme, the permutation is performed in two steps as follows: (1) the interslot group-level permutation to distribute as evenly as possible information bits in a slot over all slots in an interleaver span; and (2) the intraslot bit-level permutation to separate as distant as possible adjacent bits in a slot after group-level permutation. Computer simulations are performed to evaluate the performance of the proposed system under various system conditions and channel conditions. Simulation results show that using the PF and E scheme for uplink channel estimation at the mobile transmitter can track quite well the time variability of the uplink channel, even with some downlink CSI estimation error. We see that the system performance when using TS-MSI can be improved gradually as the interleaver size increases.