Search

In this paper, an emerging wireless communication concept, which is termed as spatial modulation (SM) for large-scale multiple-input–multiple-output (MIMO), is considered. The results show that from the information-theoretic perspective, SM achieves capacity comparable to the open-loop MIMO capacity, although a subset of transmit antennas is activated in every channel use because both the channel coefficients and the input symbols carry information in SM. As a result, SM compensates the loss of information capacity due to a subset of antennas being activated by modulating information in the antenna index; therefore, the total information rate remains high. In particular, an upper bound for the capacity of SM is derived in closed form, and it is shown analytically that this upper bound is almost certainly achievable in the massive MIMO regime. Moreover, it is shown that the upper bound is achievable with no channel state information at the transmitter (CSIT) but with channel distribution information (CDI) at the transmitter (CDIT). The optimum transmission strategy should adapt the channel input distribution to fading using CDI such as the $K$ factor in Rician fading or the shape parameter $m$ in Nakagami- $m$ fading. [ABSTRACT FROM PUBLISHER]

In this paper, two major multiple-input-multiple-output (MIMO) transmission techniques such as spatial modulation (SM) and spatial multiplexing (SMX) are considered. In particular, the focus is on an interference scenario, where an SM MIMO system interferes with a conventional SMX MIMO system. The interference caused by SM has a peculiar behavior. In SMX MIMO networks, the interference subspace is fixed in a given coherence time interval. It can be estimated accurately, and interference suppression techniques can be applied accordingly. However, the interference caused by an SM system has no fixed interference subspace, and it changes in every single channel use. Therefore, SM interference could pose some unique challenges to SMX MIMO systems in some scenarios. In this paper, the effect of SM interference on SMX MIMO systems and methods to overcome/reduce the adverse effects caused by SM interference are studied. Furthermore, the effect of SMX interference on SM MIMO systems is also studied. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel Virtual Multiple-Input–Multiple-Output (VMIMO) architecture based on the concept of Spatial Modulation (SM). Using a dual-hop and Decode-and-Forward protocol, we form a distributed system, called Dual-Hop Hybrid SM (DH-HSM). DH-HSM conveys information from a Source Node (SN) to a Destination Node (DN) via multiple Relay Nodes (RNs). The spatial position of the RNs is exploited for transferring information in addition to, or even without, a conventional symbol. In order to increase the performance of our architecture, while keeping the complexity of the RNs and DN low, we employ linear precoding using Channel State Information (CSI) at the SN. In this way, we form a Receive-Spatial Modulation (R-SM) pattern from the SN to the RNs, which is able to employ a centralized coordinated or a distributed uncoordinated detection algorithm at the RNs. In addition, we focus on the SN and propose two regularized linear precoding methods that employ realistic Imperfect Channel State Information at the Transmitter. The power of each precoder is analyzed theoretically. Using the Bit Error Rate (BER) metric, we evaluate our architecture against the following benchmark systems: 1) single relay; 2) best relay selection; 3) distributed Space Time Block Coding (STBC) VMIMO scheme; and 4) the direct communication link. We show that DH-HSM is able to achieve significant Signal-to-Noise Ratio (SNR) gains, which can be as high as 10.5 dB for a very large scale system setup. In order to verify our simulation results, we provide an analytical framework for the evaluation of the Average Bit Error Probability (ABEP). [ABSTRACT FROM AUTHOR]

The prospect of base station cooperation leading to joint combining at widely separated antennas has led to increased interest in macrodiversity systems, where both sources and receive antennas are geographically distributed. In this scenario, analytical investigation of channel capacity is extremely challenging for finite-size systems since the channel matrices have a very general form where each path may have a different power. Hence, in this paper, we consider the ergodic sum capacity of a macrodiversity multiple-input multiple-output system with arbitrary numbers of sources and receive antennas operating over Rayleigh fading channels. For this system, we compute the exact ergodic capacity for a system with at most two transmit antennas and a compact approximation for the general system, which is shown to be very accurate over a wide range of cases. Finally, we compare our results with previous asymptotic results and bounds. Results are verified by Monte Carlo simulations and the impact on capacity of various channel power profiles is investigated. [ABSTRACT FROM AUTHOR]

The performance of linear receivers in the presence of co-channel interference in Rayleigh channels is a fundamental problem in wireless communications. Performance evaluation for these systems is well-known for receive arrays where the antennas are close enough to experience equal average SNRs from a source. In contrast, almost no analytical results are available for macrodiversity systems where both the sources and receive antennas are widely separated. Here, receive antennas experience unequal average SNRs from a source and a single receive antenna receives a different average SNR from each source. Although this is an extremely difficult problem, progress is possible for the two-user scenario. In this paper, we derive closed form results for the probability density function (pdf) and cumulative distribution function (cdf) of the output signal to interference plus noise ratio (SINR) and signal to noise ratio (SNR) of minimum mean squared error (MMSE) and zero forcing (ZF) receivers in independent Rayleigh channels with arbitrary numbers of receive antennas. The results are verified by Monte Carlo simulations and approximations and high SNR analysis are also derived. The results enable further system analysis such as the evaluation of outage probability, bit error rate (BER) and capacity. [ABSTRACT FROM PUBLISHER]