With the increase of the cell loadings, the interference among the adjacent cells becomes severer acting as the limiting factor for the cell-edge user performance. In this regard, there have been many standard activities in 3GPP (3rd Generation Partnership Project) to mitigate the ICI (Inter-Cell Interference). The eNB (Enhanced Node-B)-centric ICI mitigation having its own limit, the recent approaches incorporate the interference cancellation of a dominant interference from the adjacent cell at the receiver. However, for the LTE (Long Term Evolution) uplink, the SC-FDMA (Single-Carrier Frequency Division Multiple Access) structure prohibited the eNB to use a similar interference cancellation scheme as for the OFDMA (Orthogonal Frequency Division Multiple Access) UE (User Equipment) receiver. In this paper, we propose a blind interference cancellation receiver for LTE eNB, which performs ICI cancellation blindly without prior information of the adjacent cell interferences. It is shown that the proposed blind interference cancellation receiver for SC-FDMA provides significant performance gain against the conventional receiver.

In cellular systems, the interference has been the most critical issue that practically limits the performance of overall networks. Furthermore, the distribution of the inter-cell interference (ICI) in conventional cellular networks employing orthogonal frequency-division multiple-access (OFDMA) with quadrature amplitude modulation (QAM) tends to approach a Gaussian distribution, which is known to be the worst-case distribution of the ICI as additive noise resulting in poor channel capacity. Thus, a dramatic enhancement of the channel capacity for the cellular network is expected when the ICI could be designed properly so that it has a non-Gaussian distribution. In this context, a cellular OFDMA system with a novel modulation scheme, frequency and quadrature-amplitude modulation (FQAM), is proposed in this paper. The statistical distribution of the ICI is shown to deviate far from the Gaussian distribution for the proposed FQAM-based system. Accordingly, it is shown that the non-Gaussian distribution of the ICI incurred with FQAM results in significantly improved transmission rates for the cell-edge users. Also, from the measurement results using practically implemented FQAM-based OFDMA systems, it is verified that the transmission rate for the cell-edge users could be increased significantly over the conventional QAM-based OFDMA system.

The distribution of the intercell interference (ICI) in conventional cellular networks employing orthogonal frequency-division multiple-access (OFDMA) with quadrature-amplitude modulation (QAM) tends to approach a Gaussian distribution when all available subcarriers in each cell are fully loaded. Recently, it has been also shown that the worst-case distribution of the ICI as additive noise in wireless networks with respect to the channel capacity is Gaussian. Thus, the channel capacity in cellular networks is expected to be further enhanced when the ICI could be designed properly so that it has a non-Gaussian distribution. This observation motivates us to propose, in this paper, a downlink cellular OFDMA network employing a modulation scheme called frequency and QAM (FQAM). We also derive maximum-likelihood metrics for the binary or non-binary error-correcting codes employed in the proposed network and propose their practical sub-optimal versions. Numerical results demonstrate that the distribution of the ICI in the proposed network deviates far from the Gaussian distribution. As a result, the transmission rates for the cell-edge users in the proposed network are significantly improved. In addition, the measurement results using practically implemented FQAM-based OFDMA systems verify that the transmission rates for the cell-edge users can dramatically increase, compared with the conventional QAM-based OFDMA network.

A systematic, granulized piecewise linear approximation to the Jacobian logarithm is developed. This approximation reverts to the Jacobian approximation and the exact Jacobian logarithm at the far extremes. The utility of the derived approximation is demonstrated by applying it to the decoding of Turbo coded, MIMO/BICM-ID systems.

A very simple design rule for parallel concatenated Turbo codes over GF(q) with q-ary orthogonal modulation is proposed. Unlike previously proposed exhaustive search methods, the proposed design rule allows straightforward and systematic code designs for any value of q ≥ 2 and any constraint length greater than 2.