The joint use of Soft-Input Soft-Output (SISO) detectors and channel decoders in an iterative manner has received growing attention for Multi-User Multiple-Input Multiple-Output (MU-MIMO) transmission schemes since several years, as it has been shown to operate close to fundamental limits, at least asymptotically. Amongst SISO detectors, message passing algorithms such as Vector Expectation Propagation (VEP) proved to outperform significantly linear detectors such as Linear Minimum Mean Square Error (LMMSE). Aside from its higher computational complexity, turbo VEP receivers rely on different hyper-parameters that can be optimized.In this context, we propose a joint optimization through deep-unfolding of the hyper-parameters that naturally arise in this kind of doubly iterative turbo VEP receivers. One of the difficulties arising for this type of receiver is when and how to choose between an extrinsic or an A Posteriori (APP) information feedback within the turbo receiver. The optimal selection is shown here to depend on the type of the considered SISO components. By properly choosing the hyper-parameters to be optimized, we show that deep-unfolding can naturally optimize the trade-off between extrinsic and APP information feedback and bring performance gains.

Multi-User Multiple-Input Multiple-Output (MU-MIMO) communication systems have gained significant attention due to their potential to enhance spectral efficiency and improve overall system performance. Soft-Input Soft-Output (SISO) detectors and decoders for iterative algorithms are commonly employed for the estimation of transmitted bits in such system. Message passing SISO detectors such as Vector Expectation Propagation (VEP) have demonstrated significant superiority over linear detectors such as Linear Minimum Mean Square Error (LMMSE) at the cost of higher computational complexity. SISO receivers rely on different hyper-parameters that can be optimized. In this article, we propose a modification of Gated Recurrent Unit (GRU), a type of neural network architecture from the Recurrent Neural Networks (RNN) field, suitable to learn hyperparameters of unfolded SISO iterative algorithms for multi-user MIMO communications. Our proposed algorithm outperforms both non-learned and state-of-the-art unfolded algorithms with a fair complexity increase.

In the context of the Gaussian linear model, recent works have studied factor graph modification using QR decomposition that enables the derivation of scalar Expectation Propagation (EP) based detectors. In this paper, we investigate on new factor graph representations induced by the use of the Golub-Kahan bi-diagonal Decomposition (GKD) and of the Singular Value Decomposition (SVD). New EP messages induced by the GKD or SVD underlying graphs are derived, that can be both scalar or vector messages. Complexity and performance of the resulting algorithms are studied for digital communications applications.

In this study, we intend to make an in-depth investigation of the performance-complexity trade-off of low complexity Multiple-Input Multiple-Output (MIMO) signal detection based on message passing algorithms. Several detection algorithms such as Belief Propagation (BP) and Expectation Propagation (EP) have been proposed to approximate symbol Maximum A Posteriori (MAP) for high dimensional signaling. We propose a thorough examination of those algorithms and some of their low-complexity versions, through a complexity/performance trade-off analysis to identify modes of operation depending on the number of antennas and constellation order. Finally, we propose a new simplified BP detection scheme, which combines the advantages of QR precoding and Interference Cancellation (IC).

In this paper, a novel category of expectation propagation (EP) based frequency domain (FD) semi-blind receivers are proposed for single-carrier block transmissions. A recently proposed EP-based framework for deriving double-loop turbo detectors is extended to handle joint data-aided channel estimation along with EP-based soft interference cancellation (IC). When addressing this problem in a message passing framework, an unconventional probability density function prevent us from establishing analytical update functions for estimating data and channel estimates. This is solved with variational inference methods, such as mean-field (MF), expectation maximization (EM) or EP, using a three-loop receiver structure with flexible performance-complexity trade-off, thanks to fast Fourier transform (FFT) based processing.

This paper investigates the dynamic behaviour of doubly iterative bit-interleaved coded modulation (BICM) receivers based on expectation propagation (EP). When implemented in the frequency domain, for single-carrier (SC) systems, such receivers achieve attractive performance-complexity trade-offs in quasi-static wideband channels. With this category of receivers, conventional binary extrinsic information transfer (EXIT) functions are subject to a great number of parameters, including channel realizations, constellation and inner iteration parameters. Hence, this paper proposes a novel extrinsic information evolution analysis method which simplifies the receiver's EXIT function into independent inner transfer functions. The core idea is to track state-evolution dynamics of EP through numerically stable extrinsic variance/information transfer (EXVIT) functions. Numerical results attest to the accuracy of this method for tracking the asymptotic receiver behaviour.

This paper analyzes some emerging techniques from the broad area of Bayesian learning for the design of iterative receivers for single-carrier transmissions using bit-interleaved coded-modulation (BICM) in wideband channels. In particular, approximate Bayesian inference methods, such as expectation propagation (EP), and iterative signal-recovery methods, such as approximate message passing (AMP) algorithms are evaluated as frequency domain equalizers (FDE). These algorithms show that decoding performance can be improved by going beyond the established turbo-detection principles, by iterating over inner detection loops before decoding. A comparative analysis is performed for the case of quasistatic wideband communications channels, showing that the EP-based approach is more advantageous. Moreover, recent advances in structured learning are revisited for the iterative EP-based receiver by unfolding the inner detection loop, and obtaining a deep detection network with learnable parameters. To this end, a novel, mutual-information dependent learning cost function is proposed, suited to turbo detectors, and through learning, the detection performance of the deep EP network is optimized.

A novel multiuser multiple-input multiple-output (MU-MIMO) receiver for single-carrier frequency-division multiple-access (SC-FDMA) is proposed within the expectation propagation (EP) framework. Recent advances in iterative receiver design show the ability of low complexity EP-based iterative algorithms to achieve performance close to maximum a posteriori (MAP) detection. In this paper, based on the exact derivation of frequency-domain EP-based turbo receivers, we propose a novel spectrally and computationally efficient soft interference canceller (IC) with a doubly-iterative architecture. Asymptotic and finite-length analysis show that our proposal outperforms existing schemes with comparable complexity, in various spatial-multiplexing configurations.

Third Generation Partnership Project (3GPP) has defined a solution for direct-mode communications by introducing SideLink (SL) channels in LTE-Advanced Pro. First, we point out that in case of autonomous SL operation, half-duplexing constraints combined with the current SL Hybrid Automatic Repeat Query (HARQ) protocol can heavily affect the decoding performance. These degradations are taken into account for the derivation of an analytical model of the SL data channel successful MAC access probability. This model, which has been validated by simulation, is of particular interest for Public Safety (PS) applications and it avoids overestimation of the SL throughput at high spectral efficiencies. Finally, by using the proposed model, it is shown that if a minimum requirement on successful access probability is imposed e.g. by PS applications, then SL communications as defined in LTE-Advanced Pro can support only a small number of concurring transmitters. For instance, for a minimum successful access probability of the SL data channel equal to 90%, only up to 8 transmitters can be supported by the SL data channel using a typical SL configuration.

This paper investigates the design and analysis of minimum mean square error (MMSE) turbo decision feedback equalization (DFE), with expectation propagation (EP), for single carrier modulations. Classical non iterative DFE structures have substantial advantages at high-data rates, even compared with turbo linear equalizers-interference cancellers (LE-IC), hence making turbo DFE-IC schemes an attractive solution. In this paper, we derive an iterative DFE-IC, capitalizing on the use of soft feedback based on expectation propagation, along with the use of prior information for improved filtering and interference cancellation. This turbo iterative DFE-IC significantly outperforms turbo LE-IC, especially at high-spectral efficiency and also exhibits performance improvements over existing DFE-IC variants. The proposed scheme can also be self-iterated, as done in the recent trend on EP-based equalizers, and it is shown to be an attractive alternative to linear self-iterated receivers. For time-varying (TV) filter equalizers, an efficient matrix inversion scheme is also proposed, considerably reducing the computational complexity relative to existing methods. Using finite-length and asymptotic analysis on a severely selective channel, the proposed DFE-IC is shown to achieve higher rates than known alternatives, with better waterfall thresholds and faster convergence, while keeping a similar computational complexity.

An original expectation propagation-based message passing framework is introduced, wherein transmitted symbols are considered to belong to the multivariate white Gaussian distribution family. This approach allows deriving a novel class of single-tap frequency domain (FD) receivers with a quasi-linear computational complexity in block length, thanks to fast-Fourier transform-based implementation. This framework is exposed in detail, through the design of a novel double-loop single-carrier FD equalizer (FDE), where self-iterations of the equalizer with the demapper and turbo iterations with the decoder provide numerous combinations for the performance and complexity tradeoff. Furthermore, the flexibility of this framework is illustrated with the derivation of an overlap FDE, used for time-varying channel equalization, among others, and with the design of an FD multiple-input multiple-output detector, used for spatial multiplexing. Through these different receiver design problems, this framework is shown to improve the mitigation of inter-symbol, inter-block, and multi-antenna interferences, compared to alternative single-tap FD structures of previous works. Thanks to finite-length and asymptotic analysis, supported by numerical results, the improvement brought by the proposed structures is assessed and then completed by also accounting for computational costs.

In this work, we demonstrate and validate a novel architecture for next generation cellular networks that enables collaborative forwarding at Layer 2 among adjacent eNBs with the aid of enhanced user equipment (UE) devices, that act voluntarily as packet forwarders. We introduce an evolved-UE (eUE) which is capable of operating simultaneously over multiples eNBs in order to enable reliable multi-hop operation through relaying and to achieve low-latency communication through efficient L2/MAC forwarding. For the demonstration and the evaluation of this architecture, we used the OpenAirInterface emulation platform to implement it, and also to evaluate its performance. The obtained results show that, the proposed architecture achieves significant reduction in latency (up to 16.94%) and improvement on packet loss rate (up to 59.25%), as the number of the employed eUEs increases with increasing BLER up to 20%. Moreover, the proposed architecture enables eUEs to increase the aggregated data rate in downlink by exploiting data connection to multiple eNBs.

In this paper, we propose a novel architecture for next generation cellular networks that enables collaborative forwarding at Layer 2 among adjacent eNBs with the aid of enhanced user equipment (UE) devices, that act voluntarily as packet forwarders. Therefore, legacy UEs are leveraged as active network elements being capable of operating simultaneously over multiple base stations (eNBs). To this end, we introduce an evolved-UE (eUE) in order to enable reliable multi-hop operation through relaying and to achieve low-latency communication through efficient L2/MAC forwarding. Through extensive experimentation with OpenAirInterface emulation platform, we evaluated the performance and also validated the feasibility of the proposed architecture. Our results show that, in certain use cases corresponding to public safety and moving/small cell scenarios, the proposed architecture achieves significant reduction in latency (up to 16.94%) and improvement on packet loss rate (up to 59.25%), as the number of the employed eUEs increases with increasing BLER up to 20%. Moreover, the proposed architecture enables eUEs to increase the aggregated data rate in downlink by exploiting data connection to multiple eNBs at the expense of extra power consumption, which calls for the appropriate incentives to enable such a cooperation.

In this paper, we investigate the selection of cooperative relays in a cellular mobile network. We take into account the effect of the resource sharing of each relay by its served users on the relay selection strategy. At first, we propose a novel selection strategy of one cooperative relay considering both Amplify and Forward (AF) and Decode and Forwards (DF) protocols. Then, we extend our study to the case of selection of two cooperative relays in an attempt to improve the spatial diversity gain. The performances of the proposed selection algorithms are analyzed in terms of number of users assisted by cooperation and residual powers at the relays.

Frequency offset estimation for time-hopping (TH) ultra-wide-band (UWB) is addressed in the literature by relying on an AWGN assumption and by exploiting a periodic preamble appended to each packet. In this paper we generalize these techniques with two aims. First, we identify a solution which does not rely on any periodic structure, but can be implemented with a generic TH format. Second, we identify a solution which is robust to multiple access interference (MAI) by assuming a Gaussian mixture (GM) model for MAI. In fact, GMs have recently been identified as good descriptors of UWB interference, and they provide closed form and limited complexity results. With these ideas in mind, we build a data aided maximum likelihood (ML) estimator. The proposed ML solution shows quasi optimum performance in the Cramer-Rao bound sense, and proves to be robust in meaningful multiple user scenarios.

The use of single-antenna terminals for coherent cooperative wireless communications has received more attention in literature than for non-coherent communications. But acquiring knowledge of the fading coefficients in a cooperative scheme is very challenging in a fast fading case. In this paper, we develop several cooperative schemes suited for systems which do not require channel state information at either relays or destination. We analyze the performance of these schemes in terms of diversity gain and error probability. The diversity gained with an equal gain combiner at the destination is investigated too. The proposed cooperative schemes are compared to the literature and are shown to achieve significant performance gain depending on the type of relaying protocol.

Today link-to-system (L2S) interfaces are more and more used in order to speed up complete system-level simulations. In this paper a simple extension to generic incremental redundancy (IR) hybrid automatic repeat request (HARQ) strategies is presented for two well-known L2S interfaces: the exponential effective SNR metric (EESM) and the mutual information effective SNR metric (MIESM). Then we focus on the problem of the L2S interface tuning, which is necessary to achieve the highest accuracy of the prediction models. The standard calibration procedure is compared with a new method, based on the average (over channel and noise) physical (PHY) layer performance. The latter, called average calibration procedure, is less time consuming than the standard procedure. Moreover, we show by simulation that the optimal calibration factors, calculated with the two methods, converge to close values thus obtaining equivalent prediction accuracy for the same L2S interface.

We construct a new family of space-time codes suited for multiple-antenna non-coherent communications over Rayleigh flat fading channels. These codes use all the complex degrees of freedom of the system, i.e. (M times(1-M/T)) symbols per channel use, where T is the code length and M is the number of transmit antennas. Their codewords belong to the Grassmann manifold GTldrM(C), the set of the M-dimensional vector subspaces of CT. Our codes are built from space-time codes for coherent systems via a non linear map (parameterization) called the exponential map. The relationships between some properties of the non-coherent space-time codes and their corresponding coherent codes are investigated. We also propose a simplified decoding for these classes of unitary space-time codes. We analyse the performance of these codes in terms of complexity and error probability and we compare them with some current best propositions, in particular training-based codes. The performance of our codes is substantially equivalent or slightly better than the one of training-based codes, under GLRT (Generalized Likelihood Ratio Test) decoding. When a simplified decoder is applied at the receiver, the error probability curves of the two propositions seem to be basically equivalent in the SIMO case with small block size, high spectral efficiency and high number of receive antennas, and in the 2times2 MIMO case with low spectral efficiency. In the other cases the training codes with simplified decoding seem to outperform our proposition.

We consider multiple-antenna non-coherent communications over Rayleigh flat fading channels and single-antenna non-coherent systems with unknown channel phase. For both systems, we propose some sub-optimal simplified decodings for the class of unitary space-time codes obtained via the exponential map (I. Kammoun and J.-C. Belfiore, Nov. 2003). To this aim, we use the geometrical interpretation of these codes as sets of points over the Grassmann manifold G/sub TM/. We compare these codes with other propositions in the literature, for which a simplified decoder exists.

This letter proposes a new pulse amplitude modulation decomposition of continuous phase modulation (CPM) signals under the constraint of phase response separability. The separability condition is met by a broad class of CPM signals, including full-response signals. We show that the proposed decomposition has minimal cardinality of N/sub c/=L(M-1), where M is the alphabet size and L is the CPM memory length. The cardinality increases linearly with L, while previous proposed decompositions have cardinalities M/sup L/-M/sup L-1/, which grow exponentially with L.

A continuous phase modulation (CPM) is specified by the size M of the input alphabet, the modulation index h, the phase response /spl phi/(t) and the duration LT of the active part of /spl phi/(t). The integer L indicates the memory and, roughly, the complexity of a CPM system. In this contribution we show that a CPM with an arbitrary memory L is equivalent to a CPM with a unitary memory and a bigger alphabet with size L(M - 1) + 1. The assumption is that the phase response /spl phi/(t) is "separable", a requirement verified in some practical implementations. We discuss some consequences of this equivalence on modulator and demodulator structure.