Search

We propose a novel method for interpreting neural networks, focusing on convolutional neural network-based receiver model. The method identifies which unit or units of the model contain most (or least) information about the channel parameter(s) of the interest, providing insights at both global and local levels -- with global explanations aggregating local ones. Experiments on link-level simulations demonstrate the method's effectiveness in identifying units that contribute most (and least) to signal-to-noise ratio processing. Although we focus on a radio receiver model, the method generalizes to other neural network architectures and applications, offering robust estimation even in high-dimensional settings., Comment: 7+1 pages, 8 figures, 1 equation

Recent research has shown that integrating artificial intelligence (AI) into wireless communication systems can significantly improve spectral efficiency. However, the prevalent use of simulated radio channel data for training and validating neural network-based radios raises concerns about their generalization capability to diverse real-world environments. To address this, we conducted empirical over-the-air (OTA) experiments using software-defined radio (SDR) technology to test the performance of an NN-based orthogonal frequency division multiplexing (OFDM) receiver in a real-world small cell scenario. Our assessment reveals that the performance of receivers trained on diverse 3GPP TS38.901 channel models and broad parameter ranges significantly surpasses conventional receivers in our testing environment, demonstrating strong generalization to a new environment. Conversely, setting simulation parameters to narrowly reflect the actual measurement environment led to suboptimal OTA performance, highlighting the crucial role of rich and randomized training data in improving the NN-based receiver's performance. While our empirical test results are promising, they also suggest that developing new channel models tailored for training these learned receivers would enhance their generalization capability and reduce training time. Our testing was limited to a relatively narrow environment, and we encourage further testing in more complex environments.

This paper investigates the feasibility of machine learning (ML)-based pilotless spatial multiplexing in multiple-input and multiple-output (MIMO) communication systems. Especially, it is shown that by training the transmitter and receiver jointly, the transmitter can learn such constellation shapes for the spatial streams which facilitate completely blind separation and detection by the simultaneously learned receiver. To the best of our knowledge, this is the first time ML-based spatial multiplexing without channel estimation pilots is demonstrated. The results show that the learned pilotless scheme can outperform a conventional pilot-based system by as much as 15-20% in terms of spectral efficiency, depending on the modulation order and signal-to-noise ratio., Comment: Presented in the 57th Asilomar Conference on Signals, Systems, and Computers

Machine learning algorithms have recently been considered for many tasks in the field of wireless communications. Previously, we have proposed the use of a deep fully convolutional neural network (CNN) for receiver processing and shown it to provide considerable performance gains. In this study, we focus on machine learning algorithms for the transmitter. In particular, we consider beamforming and propose a CNN which, for a given uplink channel estimate as input, outputs downlink channel information to be used for beamforming. The CNN is trained in a supervised manner considering both uplink and downlink transmissions with a loss function that is based on UE receiver performance. The main task of the neural network is to predict the channel evolution between uplink and downlink slots, but it can also learn to handle inefficiencies and errors in the whole chain, including the actual beamforming phase. The provided numerical experiments demonstrate the improved beamforming performance., Comment: 27 pages, this work has been submitted to the IEEE for possible publication; v2: Fixed typo in author name, v3: a revision

Machine learning (ML) has shown great promise in optimizing various aspects of the physical layer processing in wireless communication systems. In this paper, we use ML to learn jointly the transmit waveform and the frequency-domain receiver. In particular, we consider a scenario where the transmitter power amplifier is operating in a nonlinear manner, and ML is used to optimize the waveform to minimize the out-of-band emissions. The system also learns a constellation shape that facilitates pilotless detection by the simultaneously learned receiver. The simulation results show that such an end-to-end optimized system can communicate data more accurately and with less out-of-band emissions than conventional systems, thereby demonstrating the potential of ML in optimizing the air interface. To the best of our knowledge, there are no prior works considering the power amplifier induced emissions in an end-to-end learned system. These findings pave the way towards an ML-native air interface, which could be one of the building blocks of 6G.

In this paper, we propose a machine learning (ML) based physical layer receiver solution for demodulating OFDM signals that are subject to a high level of nonlinear distortion. Specifically, a novel deep learning based convolutional neural network receiver is devised, containing layers in both time- and frequency domains, allowing to demodulate and decode the transmitted bits reliably despite the high error vector magnitude (EVM) in the transmit signal. Extensive set of numerical results is provided, in the context of 5G NR uplink incorporating also measured terminal power amplifier characteristics. The obtained results show that the proposed receiver system is able to clearly outperform classical linear receivers as well as existing ML receiver approaches, especially when the EVM is high in comparison with modulation order. The proposed ML receiver can thus facilitate pushing the terminal power amplifier (PA) systems deeper into saturation, and thereon improve the terminal power-efficiency, radiated power and network coverage., Comment: To be presented in the 2021 IEEE International Symposium on Personal, Indoor and Mobile Radio Communications

Recently, deep learning has been proposed as a potential technique for improving the physical layer performance of radio receivers. Despite the large amount of encouraging results, most works have not considered spatial multiplexing in the context of multiple-input and multiple-output (MIMO) receivers. In this paper, we present a deep learning-based MIMO receiver architecture that consists of a ResNet-based convolutional neural network, also known as DeepRx, combined with a so-called transformation layer, all trained together. We propose two novel alternatives for the transformation layer: a maximal ratio combining-based transformation, or a fully learned transformation. The former relies more on expert knowledge, while the latter utilizes learned multiplicative layers. Both proposed transformation layers are shown to clearly outperform the conventional baseline receiver, especially with sparse pilot configurations. To the best of our knowledge, these are some of the first results showing such high performance for a fully learned MIMO receiver.

Deep learning has solved many problems that are out of reach of heuristic algorithms. It has also been successfully applied in wireless communications, even though the current radio systems are well-understood and optimal algorithms exist for many tasks. While some gains have been obtained by learning individual parts of a receiver, a better approach is to jointly learn the whole receiver. This, however, often results in a challenging nonlinear problem, for which the optimal solution is infeasible to implement. To this end, we propose a deep fully convolutional neural network, DeepRx, which executes the whole receiver pipeline from frequency domain signal stream to uncoded bits in a 5G-compliant fashion. We facilitate accurate channel estimation by constructing the input of the convolutional neural network in a very specific manner using both the data and pilot symbols. Also, DeepRx outputs soft bits that are compatible with the channel coding used in 5G systems. Using 3GPP-defined channel models, we demonstrate that DeepRx outperforms traditional methods. We also show that the high performance can likely be attributed to DeepRx learning to utilize the known constellation points of the unknown data symbols, together with the local symbol distribution, for improved detection accuracy., Comment: 32 pages, this work has been submitted to the IEEE for possible publication

As low Earth orbit (LEO) satellite communication systems are gaining increasing popularity, new theoretical methodologies are required to investigate such networks' performance at large. This is because deterministic and location-based models that have previously been applied to analyze satellite systems are typically restricted to support simulations only. In this paper, we derive analytical expressions for the downlink coverage probability and average data rate of generic LEO networks, regardless of the actual satellites' locality and their service area geometry. Our solution stems from stochastic geometry, which abstracts the generic networks into uniform binomial point processes. Applying the proposed model, we then study the performance of the networks as a function of key constellation design parameters. Finally, to fit the theoretical modeling more precisely to real deterministic constellations, we introduce the effective number of satellites as a parameter to compensate for the practical uneven distribution of satellites on different latitudes. In addition to deriving exact network performance metrics, the study reveals several guidelines for selecting the design parameters for future massive LEO constellations, e.g., the number of frequency channels and altitude., Comment: Accepted for publication in the IEEE Transactions on Communications in April 2020

In this paper, new digital predistortion (DPD) solutions for power amplifier (PA) linearization are proposed, with particular emphasis on reduced processing complexity in future 5G and beyond wideband radio systems. The first proposed method, referred to as the spline-based Hammerstein (SPH) approach, builds on complex spline-interpolated lookup table (LUT) followed by a linear finite impulse response (FIR) filter. The second proposed method, the spline-based memory polynomial (SMP) approach, contains multiple parallel complex spline-interpolated LUTs together with an input delay line such that more versatile memory modeling can be achieved. For both structures, gradient-based learning algorithms are derived to efficiently estimate the LUT control points and other related DPD parameters. Large set of experimental results are provided, with specific focus on 5G New Radio (NR) systems, showing successful linearization of multiple sub-6 GHz PA samples as well as a 28 GHz active antenna array, incorporating channel bandwidths up to 200 MHz. Explicit performance-complexity comparisons are also reported between the SPH and SMP DPD systems and the widely-applied ordinary memory-polynomial (MP) DPD solution. The results show that the linearization capabilities of the proposed methods are very close to that of the ordinary MP DPD, particularly with the proposed SMP approach, while having substantially lower processing complexity., Comment: The manuscript has 14 pages, 19 figures

Modern radio systems and transceivers utilize carrier aggregation (CA) to meet the demands for higher and higher data rates. However, the adoption of CA in the existing Long Term Evolution (LTE)-Advanced and emerging 5G New Radio (NR) mobile networks, in case of frequency division duplexing (FDD), may incur self-interference challenges with certain band combinations. More specifically, the nonlinear distortion products of the transmit signals or component carriers (CCs), stemming from the passive radio frequency (RF) front-end components of the transceiver, can appear in one or more of the configured receiver bands, potentially leading to the receiver desensitization. In this paper, we present advanced baseband equivalent signal models for such passive intermodulation (PIM) distortion viewed from the RX point of view, considering also potential memory effects in the PIM generation. Then, building on these signal models, a digital self-interference cancellation technique operating in the transceiver digital front-end is presented. The performance of the proposed solution is evaluated with real-life RF measurements for LTE-Advanced type user equipment (UE) with dual CC inter-band CA, demonstrating excellent suppression properties. The findings in this work indicate that digital cancellation is a feasible approach for improving the receiver sensitivity of mobile devices that may be prone to RF front-end induced PIM challenges., Comment: 7 pages, 7 figures, accepted to 2018 Asilomar Conference on Signals, Systems, and Computers

This paper proposes an active radio frequency (RF) cancellation solution to suppress the transmitter (TX) passband leakage signal in radio transceivers supporting simultaneous transmission and reception. The proposed technique is based on creating an opposite-phase baseband equivalent replica of the TX leakage signal in the transceiver digital front-end through adaptive nonlinear filtering of the known transmit data, to facilitate highly accurate cancellation under a nonlinear TX power amplifier (PA). The active RF cancellation is then accomplished by employing an auxiliary transmitter chain, to generate the actual RF cancellation signal, and combining it with the received signal at the receiver (RX) low noise amplifier (LNA) input. A closed-loop parameter learning approach, based on the decorrelation principle, is also developed to efficiently estimate the coefficients of the nonlinear cancellation filter in the presence of a nonlinear TX PA with memory, finite passive isolation, and a nonlinear RX LNA. The performance of the proposed cancellation technique is evaluated through comprehensive RF measurements adopting commercial LTE-Advanced transceiver hardware components. The results show that the proposed technique can provide an additional suppression of up to 54 dB for the TX passband leakage signal at the RX LNA input, even at considerably high transmit power levels and with wide transmission bandwidths. Such novel cancellation solution can therefore substantially improve the TX-RX isolation, hence reducing the requirements on passive isolation and RF component linearity, as well as increasing the efficiency and flexibility of the RF spectrum use in the emerging 5G radio networks., Comment: accepted to IEEE

In this article, a radio access system with a self-backhauling full-duplex access node serving legacy half-duplex mobile devices is studied and analyzed. In particular, it is assumed that the access node is using the same center frequency for all the transmissions, meaning that also the backhauling is done using the same frequency resources as the uplink and downlink transmissions. It is further assumed that the access node has a massive array to facilitate efficient beamforming and self-interference nulling in its own receiver. As a starting point, the signal model for the considered access node is first derived, including all the transmitted and received signals within the cell. This is then used as a basis for obtaining the sum-rate expressions, which depict the overall rates experienced by the mobile users that are served by the access node. In addition, the data rate for the bi-directional backhaul link is also derived, since the access node must be able to backhaul itself wirelessly. The maximum achievable sum-rate is then determined by numerically solving an optimization problem constructed from the data rate expressions. The full-duplex scheme is also compared to two alternative transmission schemes, which perform all or some of the transmissions in half-duplex mode. The results show that the full-duplex capability of the access node is beneficial for maximizing the sum-rate, meaning that a simple half-duplex transmission scheme is typically not optimal. In particular, the highest sum-rate is usually provided by a relay type solution, where the access node acts as a full-duplex relay between the mobiles and the backhaul node., Comment: 30 pages, submitted for review

This paper presents a novel RF circuit architecture for self-interference cancellation in inband full-duplex radio transceivers. The developed canceller is able to provide wideband cancellation with waveform bandwidths in the order of 100 MHz or beyond and contains also self-adaptive or self-healing features enabling automatic tracking of time-varying self-interference channel characteristics. In addition to architecture and operating principle descriptions, we also provide actual RF measurements at 2.4 GHz ISM band demonstrating the achievable cancellation levels with different bandwidths and when operating in different antenna configurations and under low-cost highly nonlinear power amplifier. In a very challenging example with a 100 MHz waveform bandwidth, around 41 dB total cancellation is obtained while the corresponding cancellation figure is close to 60 dB with the more conventional 20 MHz carrier bandwidth. Also, efficient tracking in time-varying reflection scenarios is demonstrated., Comment: 7 pages, to be presented in 2015 IEEE 81st Vehicular Technology Conference

In this article, we address the challenges of transmitter-receiver isolation in \emph{mobile full-duplex devices}, building on shared-antenna based transceiver architecture. Firstly, self-adaptive analog RF cancellation circuitry is required, since the capability to track time-varying self-interference coupling characteristics is of utmost importance in mobile devices. In addition, novel adaptive nonlinear DSP methods are also required for final self-interference suppression at digital baseband, since mobile-scale devices typically operate under highly nonlinear low-cost RF components. In addition to describing above kind of advanced circuit and signal processing solutions, comprehensive RF measurement results from a complete demonstrator implementation are also provided, evidencing beyond 40~dB of active RF cancellation over an 80 MHz waveform bandwidth with a highly nonlinear transmitter power amplifier. Measured examples also demonstrate the good self-healing characteristics of the developed control loop against fast changes in the coupling channel. Furthermore, when complemented with nonlinear digital cancellation processing, the residual self-interference level is pushed down to the noise floor of the demonstration system, despite the harsh nonlinear nature of the self-interference. These findings indicate that deploying the full-duplex principle can indeed be feasible also in mobile devices, and thus be one potential technology in, e.g., 5G and beyond radio systems., Comment: 18 pages, submitted for review

In this paper we analyze the effect of the calibration period, or lack of, on self-interference channel estimation in the digital domain of in-band full-duplex radio transceivers. In particular, we consider a scenario where the channel estimation must be performed without a separate calibration period, which means that the received signal of interest will act as an additional noise source from the estimation perspective. We will explicitly analyze its effect, and quantify the increase in the parameter estimation variance, or sample size, if similar accuracy for the channel estimate is to be achieved as with a separate calibration period. In addition, we will analyze how the calibration period, or its absence, affects the overall achievable rates. Full waveform simulations are then used to determine the validity of the obtained results, as well as provide numerical results regarding the achievable rates. It is shown that, even though a substantial increase in the parameter sample size is required if there is no calibration period, the achievable rates are still comparable for the two scenarios., Comment: 8 pages, to be presented in the Asilomar Conference on Signals, Systems, and Computers in November 2014

In this paper we provide an overview regarding the feasibility of in-band full-duplex transceivers under imperfect RF components. We utilize results and findings from the recent research on full-duplex communications, while introducing also transmitter-induced thermal noise into the analysis. This means that the model of the RF impairments used in this paper is the most comprehensive thus far. By assuming realistic parameter values for the different transceiver components, it is shown that IQ imaging and transmitter-induced nonlinearities are the most significant sources of distortion in in-band full-duplex transceivers, in addition to linear self-interference. Motivated by this, we propose a novel augmented nonlinear digital self-interference canceller that is able to model and hence suppress all the essential transmitter imperfections jointly. This is also verified and demonstrated by extensive waveform simulations., Comment: 7 pages, presented in the CROWNCOM 2014 conference

This paper addresses the modeling and digital cancellation of self-interference in in-band full-duplex (FD) transceivers with multiple transmit and receive antennas. The self-interference modeling and the proposed nonlinear spatio-temporal digital canceller structure takes into account, by design, the effects of I/Q modulator imbalances and power amplifier (PA) nonlinearities with memory, in addition to the multipath self-interference propagation channels and the analog RF cancellation stage. The proposed solution is the first cancellation technique in the literature which can handle such a self-interference scenario. It is shown by comprehensive simulations with realistic RF component parameters and with two different PA models to clearly outperform the current state-of-the-art digital self-interference cancellers, and to clearly extend the usable transmit power range., Comment: 7 pages, 5 figures. To be presented in the 2014 International Workshop on Emerging Technologies for 5G Wireless Cellular Networks

In this paper we propose and analyze a novel self-interference cancellation structure for in-band MIMO full-duplex transceivers. The proposed structure utilizes reference receiver chains to obtain reference signals for digital self-interference cancellation, which means that all the transmitter-induced nonidealities will be included in the digital cancellation signal. To the best of our knowledge, this type of a structure has not been discussed before in the context of full-duplex transceivers. First, we will analyze the overall achievable performance of the proposed cancellation scheme, while also providing some insight into the possible bottlenecks. We also provide a detailed formulation of the actual cancellation procedure, and perform an analysis into the effect of the received signal of interest on self-interference coupling channel estimation. The achieved performance of the proposed reference receiver based digital cancellation procedure is then assessed and verified with full waveform simulations. The analysis and waveform simulation results show that under practical transmitter RF/analog impairment levels, the proposed reference receiver based cancellation architecture can provide substantially better self-interference suppression than any existing solution, despite deploying only low-complexity linear digital processing., Comment: 7 pages, 4 figures. To be presented in the 2014 IEEE Broadband Wireless Access Workshop

This article addresses the modeling and cancellation of self-interference in full-duplex direct-conversion radio transceivers, operating under practical imperfect radio frequency (RF) components. Firstly, detailed self-interference signal modeling is carried out, taking into account the most important RF imperfections, namely transmitter power amplifier nonlinear distortion as well as transmitter and receiver IQ mixer amplitude and phase imbalances. The analysis shows that after realistic antenna isolation and RF cancellation, the dominant self-interference waveform at receiver digital baseband can be modeled through a widely-linear transformation of the original transmit data, opposed to classical purely linear models. Such widely-linear self-interference waveform is physically stemming from the transmitter and receiver IQ imaging, and cannot be efficiently suppressed by classical linear digital cancellation. Motivated by this, novel widely-linear digital self-interference cancellation processing is then proposed and formulated, combined with efficient parameter estimation methods. Extensive simulation results demonstrate that the proposed widely-linear cancellation processing clearly outperforms the existing linear solutions, hence enabling the use of practical low-cost RF front-ends utilizing IQ mixing in full-duplex transceivers., Comment: IEEE Journal on Selected Areas in Communications, vol. 32, no. 9, pp. 1674-1687, September 2014

This paper addresses the analysis of oscillator phase-noise effects on the self-interference cancellation capability of full-duplex direct-conversion radio transceivers. Closed-form solutions are derived for the power of the residual self-interference stemming from phase noise in two alternative cases of having either independent oscillators or the same oscillator at the transmitter and receiver chains of the full-duplex transceiver. The results show that phase noise has a severe effect on self-interference cancellation in both of the considered cases, and that by using the common oscillator in upconversion and downconversion results in clearly lower residual self-interference levels. The results also show that it is in general vital to use high quality oscillators in full-duplex transceivers, or have some means for phase noise estimation and mitigation in order to suppress its effects. One of the main findings is that in practical scenarios the subcarrier-wise phase-noise spread of the multipath components of the self-interference channel causes most of the residual phase-noise effect when high amounts of self-interference cancellation is desired., Comment: Revised manuscript for IEEE Transactions on Wireless Communications, 2014

Despite the intensive recent research on wireless single-channel full-duplex communications, relatively little is known about the transceiver chain nonidealities of full-duplex devices. In this paper, the effect of nonlinear distortion occurring in the transmitter power amplifier (PA) and the receiver chain is analyzed, alongside with the dynamic range requirements of analog-to-digital converters (ADCs). This is done with detailed system calculations, which combine the properties of the individual electronics components to jointly model the complete transceiver chain, including self-interference cancellation. They also quantify the decrease in the dynamic range for the signal of interest caused by self-interference at the analog-to-digital interface. Using these system calculations, we provide comprehensive numerical results for typical transceiver parameters. The analytical results are also confirmed with full waveform simulations. We observe that the nonlinear distortion produced by the transmitter PA is a significant issue in a full-duplex transceiver and, when using cheaper and less linear components, also the receiver chain nonlinearities become considerable. It is also shown that, with digitally-intensive self-interference cancellation, the quantization noise of the ADCs is another significant problem., Comment: 16 pages, 12 figures

In an in-band full-duplex system, radios transmit and receive simultaneously in the same frequency band at the same time, providing a radical improvement in spectral efficiency over a half-duplex system. However, in order to design such a system, it is necessary to mitigate the self-interference due to simultaneous transmission and reception, which seriously limits the maximum transmit power of the full-duplex device. Especially, large differences in power levels in the receiver front-end sets stringent requirements for the linearity of the transceiver electronics. We present an advanced architecture for a compact full-duplex multiantenna transceiver combining antenna design with analog and digital cancellation, including both linear and nonlinear signal processing., Comment: Presented in 47th Annual Asilomar Conference on Signals, Systems, and Computers, 2013

Recently, full-duplex (FD) communications with simultaneous transmission and reception on the same channel has been proposed. The FD receiver, however, suffers from inevitable self-interference (SI) from the much more powerful transmit signal. Analogue radio-frequency (RF) and baseband, as well as digital baseband, cancellation techniques have been proposed for suppressing the SI, but so far most of the studies have failed to take into account the inherent nonlinearities of the transmitter and receiver front-ends. To fill this gap, this article proposes a novel digital nonlinear interference cancellation technique to mitigate the power amplifier (PA) induced nonlinear SI in a FD transceiver. The technique is based on modeling the nonlinear SI channel, which is comprised of the nonlinear PA, the linear multipath SI channel, and the RF SI canceller, with a parallel Hammerstein nonlinearity. Stemming from the modeling, and appropriate parameter estimation, the known transmit data is then processed with the developed nonlinear parallel Hammerstein structure and suppressed from the receiver path at digital baseband. The results illustrate that with a given IIP3 figure for the PA, the proposed technique enables higher transmit power to be used compared to existing linear SI cancellation methods. Alternatively, for a given maximum transmit power level, a lower-quality PA (i.e., lower IIP3) can be used., Comment: To appear in proceedings of the 2013 Asilomar Conference on Signals, Systems & Computers