An iterative receiver for a multiple-input-multiple-output (MIMO) orthogonal frequency-division multiplexing (OFDM) system is considered to jointly decode the transmitted bits and estimate the channel state. The receiver consists of the a posteriori probability (APP) algorithm, the repeat-accumulate (RA) decoder, and the least-squares (LS) channel estimator. An obvious problem, with more than two blocks in an iterative receiver, is to find the optimal activation schedule of the different blocks. This paper proposes to use extrinsic information transfer (EXIT) charts to characterize the behavior of the receiver blocks and find out the optimal activation schedule for them. A semi-analytical expression of the EXIT function is derived for the decision directed LS channel estimator. An algorithm is proposed to generate the EXIT function of the APP algorithm as a function of channel estimate mutual information (MI). Surface fitting is used to get closed form expressions for the EXIT functions of the APP algorithm and the RA decoder. Trellis search based algorithms are shown to find the convergence with the lowest possible complexity using the EXIT charts. With the proposed concept, the activation scheduling can be adapted to prevailing channel circumstances and unnecessary iterations will be avoided.

A three-dimensional (3D) multi-cell multi-user system model along with spatial array processing are presented and studied in this paper. Most of the radio propagation in the literature has been studied and modeled in a two dimensional plane. The 3D propagation channel modeling has drawn attentions recently. This is due to the fact that, in some scenarios, the assumption of small elevation domain angular spread does not hold. With multiple antenna elements displaced in a 3D space, e.g., uniform linear array (ULA) or uniform planar array (UPA), the base station (BS) is capable of forming beams in vertical domain. This is particularly interesting for the interference limited scenario, since the 3D array processing provides another degree of freedom to transmit signal and eliminate the interference. Using the 3D multicell channel model, the impact of elevation domain angular spread and antenna analog beam pattern on the base station antenna design, digital beamforming and power control are studied.

This paper presents a higher performance uplink concept for the evolution of GSM/EDGE radio access network (GERAN). The EGPRS2 UL concept can operate at 1.2 times the current GERAN symbol rate, and employ 4-, 16- and 32-QAM constellations, concatenated with convolutional codes. The interference introduced by the higher transmit signal bandwidth is mitigated by the use of interference rejection combining (IRC) receivers. We explore both noise and interference limited scenarios, in link and system level simulations. The simulation study shows that the user peak data rates can be increased up to 100%, and that the average throughput over practical signal to noise ratio (SNR) conditions can gain up to 50%. Therefore, coverage and user bit rates can be significantly improved without requiring additional radio frequency spectrum.

Closed form ergodic capacity equations for multiple antenna radio links interfered by noise and interference are presented. So called burst channel with a predefined frame structure is assumed to facilitate the analysis in the case of frequency-selective channels. Perfect channel state information (CSI) is assumed at the receiver. Numerical examples are presented for flat Rayleigh fading channel in the case of different antenna setups and the characteristics of evolved universal terrestrial radio access network (UTRAN) system simulation model are evaluated with different open-loop power control options

In this paper the dual symbol rate (DSR) concept is proposed for GERAN networks. DSR doubles the modulation rate and user bit rate per radio time slot. Basic assumption is that the transmit signal of adjacent channels overlap in frequency domain. Interference rejection combining algorithms (IRC) are used to combat the introduced interference. Analysis of theoretical channel capacity suggests that data rates can be improved either by increasing the number of antennas or frequency bandwidth in the used radio links. Simulation results are given to demonstrate possible DSR deployment to the existing networks. Spectral efficiency, coverage and user bit rates are shown to be significantly improved

The further evolution of GSM/EDGE radio access network (GERAN) is currently a subject to standardization in the Third Generation Partnership Project (3GPP). The objectives include significant improvements on user data rates, spectral efficiency and coverage by avoiding concurrent impacts to the existing GSM networks. A variable bandwidth concept, a.k.a. dual symbol rate (DSR) is proposed to meet the demand for uplink enhancements, e.g. settled by camera phones. DSR doubles the modulation rate and user bit rate per radio time slot. It has overlapping transmission bands with regular GSM channel spacing. Numerical results are given to demonstrate possible DSR deployment to the existing networks. Spectral efficiency, coverage and user bit rates are shown to be significantly improved

The studies of wireless communication systems with multiple transmit and receive antennas have shown to provide large gains over single antenna systems. The recent studies have focused on the spectral efficiency improvements, while the achievable data rates of wireless radio links allowing flexible frequency band usage are much less considered. We present closed-from expressions for the achievable data rates of wireless radio links with multiple antennas operating over frequency selective fading channels including one-sided antenna correlation. In transmitter side the cases with no channel state information (CSI), covariance feedback and ideal CSI are evaluated. Perfect CSI is assumed at the receiver. Numerical results show examples of the upper limits for achievable data rates of the GSM radio links. Furthermore, the numerical results quantify the importance and the relation of different transmitter CSI, antenna correlation and bandwidth usage to the radio link behavior and data rates

An adaptive reduced-state sequence estimation (RSSE) receiver based on per-survivor processing (PS) in conjunction with a local decision feedback equalizer (DFE) for the EDGE system with 8PSK modulation is proposed in this paper. A pre-filter prior to the RSSE-PSP is exploited to reshape the channel impulse response (CIR) to have approximately a minimum phase property. The pre-filter settings are computed indirectly by estimating the CIR from the midamble with the least squares algorithm and then applying the minimum-mean square error (MMSE) criterion to reshape the estimated CIR. The proposed scheme handles the first part of the sampled CIR by the RSSE-PSP and the rest by the local DFE, which utilizes the decision per survivor mechanism for reducing the error propagation and complexity reduction. A good trade-off between the performance and complexity could be achieved by retaining the RSSE-PSP for good performance and local DFE for intersymbol interference cancellation. Computer simulations indicate that the RSSE-PSP with a pre-filter works well in ISI channels with a lower computation burden.

A new adaptive reduced-complexity MLSE receiver based on per-survivor processing (PSP-MLSE) and a decision feedback equalizer (DFE) for GSM/EDGE system with 8PSK modulation is proposed. The initial channel impulse response (CIR) is estimated by the LS algorithm with a mid-amble and updated by the LMS algorithm during data transmission. The proposed scheme handles the first part of the sampled CIR by the PSP-MLSE and the rest by the DFE for complexity reduction. Instead of the conventional global DFE (decision per sample), the local DFE (decision per state/survivor) is suggested to reduce error propagation. A good tradeoff between performance and complexity, by retaining the PSP-MLSE for good performance and using the local DFE for intersymbol interference cancellation, is obtained. Computer simulations indicate that the PSP-MLSE with the local DFE is clearly superior to that with the global DFE, especially in severe ISI channels.