Η διερεύνηση της περιοχής χωρητικότητας και της περιοχής επιτεύξιμων ρυθμών μετάδοσης καναλιών αποτελεί βασικό αντικείμενο της Θεωρίας Πληροφορίας. Η Ευθυγράμμιση Παρεμβολών είναι μια καινούρια ιδέα που δίνει μια εναλλακτική οπτική στο αντικείμενο αυτό, μέσω της διαφορετικής λογικής που εισάγει σχετικά με την κωδικοποίηση και τη μετάδοση της πληροφορίας. Σε πρόσφατες δημοσιεύσεις έχουν προταθεί μοντέλα που επιτρέπουν την εφαρμογή της θεωρίας της Ευθυγράμμισης Παρεμβολών και τη χρήση της σε πρακτικά συστήματα επικοινωνιών και καταδεικνύουν την υπεροχή της σε σχέση με συμβατικές μεθόδους. Παράλληλα, παρόλο που προς το παρόν έχει δοθεί έμφαση στην εδραίωση της Ευθυγράμμισης Παρεμβολών στις επικοινωνίες, η μαθηματική της βάση καθιστά δυνατή την εφαρμογή της σε αντικείμενα που ανήκουν σε άλλους τομείς. Σκοπός της παρούσας εργασίας είναι η μελέτη και η εφαρμογή της Αναδρομικής Ευθυγράμμισης Παρεμβολών για μετάδοση πληροφορίας σε Συστήματα Επικοινωνιών, καθώς και η διερεύνηση της απόδοσης της μεθόδου σε πρακτικά συστήματα. / The study of the channel capacity region and the achievable rate region is one of the main topics of Information Theory. Interference Alignment is a new idea that provides new insights through the introduction of a different viewpoint on data encoding and transmission. In recent publications, models have been proposed that allow the application of the theory of Interference Alignment to practical communication systems and demonstrate its superiority compared to traditional approaches. Furthermore, although for the time being emphasis has been put on establishing the use of Interference Alignment to communication systems, its mathematical formulation makes possible its use to other areas. The purpose of this thesis is to study and to apply Retrospective Interference Alignment to data transmission in communication systems and to evaluate the performance of the method in practical systems.

The security of information is an indispensable element of a communication system when transmitted signals are vulnerable to eavesdropping. This issue is a challenging problem in a wireless network as propagated signals can be easily captured by unauthorized receivers, and so achieving a perfectly secure communication is a desire in such a wiretap channel. On the other hand, cryptographic algorithms usually lack to attain this goal due to the following restrictive assumptions made for their design. First, wiretappers basically have limited computational power and time. Second, each authorized party has often access to a reasonably large sequence of uniform random bits concealed from wiretappers. To guarantee the security of information, Information Theory (IT) offers the following two approaches based on physical-layer security. First, IT suggests using wiretap (block) codes to securely and reliably transmit messages over a noisy wiretap channel. No confidential common key is usually required for the wiretap codes. The secrecy problem investigates an optimum wiretap code that achieves the secrecy capacity of a given wiretap channel. Second, IT introduces key agreement (block) codes to exchange keys between legitimate parties over a wiretap model. The agreed keys are to be reliable, secure, and (uniformly) random, at least in an asymptotic sense, such that they can be finally employed in symmetric key cryptography for data transmission. The key agreement problem investigates an optimum key agreement code that obtains the key capacity of a given wiretap model. In this thesis, we study the key agreement problem for two wiretap models: a Discrete Memoryless (DM) model and a Gaussian model. Each model consists of a wiretap channel paralleled with an authenticated public channel. The wiretap channel is from a transmitter, called Alice, to an authorized receiver, called Bob, and to a wiretapper, called Eve. The Probability Transition Function (PTF) of the wiretap channel is controlled by a random sequence of Channel State Information (CSI), which is assumed to be non-causally available at Alice. The capacity of the public channel is C_P₁∈[0,∞) in the forward direction from Alice to Bob and C_P₂∈[0,∞) in the backward direction from Bob to Alice. For each model, the key capacity as a function of the pair (C_P₁, C_P₂) is denoted by C_K(C_P₁, C_P₂). We investigate the forward key capacity of each model, i.e., C_K(C_P₁, 0) in this thesis. We also study the key generation over the Gaussian model when Eve's channel is less noisy than Bob's. In the DM model, the wiretap channel is a Discrete Memoryless State-dependent Wiretap Channel (DM-SWC) in which Bob and Eve each may also have access to a sequence of Side Information (SI) dependent on the CSI. We establish a Lower Bound (LB) and an Upper Bound (UB) on the forward key capacity of the DM model. When the model is less noisy in Bob's favor, another UB on the forward key capacity is derived. The achievable key agreement code is asymptotically optimum as C_P₁→ ∞. For any given DM model, there also exists a finite capacity C⁰_P₁, which is determined by the DM-SWC, such that the forward key capacity is achievable if C_P₁≥ C⁰_P₁. Moreover, the key generation is saturated at capacity C_P₁= C⁰_P₁, and thus increasing the public channel capacity beyond C⁰_P₁ makes no improvement on the forward key capacity of the DM model. If the CSI is fully known at Bob in addition to Alice, C⁰_P₁=0, and so the public channel has no contribution in key generation when the public channel is in the forward direction. The achievable key agreement code of the DM model exploits both a random generator and the CSI as resources for key generation at Alice. The randomness property of channel states can be employed for key generation, and so the agreed keys depend on the CSI in general. However, a message is independent of the CSI in a secrecy problem. Hence, we justify that the forward key capacity can exceed both the main channel capacity and the secrecy capacity of the DM-SWC. In the Gaussian model, the wiretap channel is a Gaussian State-dependent Wiretap Channel (G-SWC) with Additive White Gaussian Interference (AWGI) having average power Λ. For simplicity, no side information is assumed at Bob and Eve. Bob's channel and Eve's channel suffer from Additive White Gaussian Noise (AWGN), where the correlation coefficient between noise of Bob's channel and that of Eve's channel is given by ϱ. We prove that the forward key capacity of the Gaussian model is independent of ϱ. Moreover, we establish that the forward key capacity is positive unless Eve's channel is less noisy than Bob's. We also prove that the key capacity of the Gaussian model vanishes if the G-SWC is physically degraded in Eve's favor. However, we justify that obtaining a positive key capacity is feasible even if Eve's channel is less noisy than Bob's according to our achieved LB on the key capacity for case (C_P₁, C_P₂)→ (∞, ∞). Hence, the key capacity of the Gaussian model is a function of ϱ. In this thesis, an LB on the forward key capacity of the Gaussian model is achieved. For a fixed Λ, the achievable key agreement code is optimum for any C_P₁∈[0,∞) in both low Signal-to-Interference Ratio (SIR) and high SIR regimes. We show that the forward key capacity is asymptotically independent of C_P₁ and Λ as the SIR goes to infinity, and thus the public channel and the interference have negligible contributions in key generation in the high SIR regime. On the other hand, the forward key capacity is a function of C_P₁ and Λ in the low SIR regime. Contributions of the interference and the public channel in key generation are significant in the low SIR regime that will be illustrated by simulations. The proposed key agreement code asymptotically achieves the forward key capacity of the Gaussian model for any SIR as C_P₁→ ∞. Hence, C_K(∞,0) is calculated, and it is suggested as a UB on C_K(C_P₁,0). Using simulations, we also compute the minimum required C_P₁ for which the forward key capacity is upper bounded within a given tolerance. The achievable key agreement code is designed based on a generalized version of the Dirty Paper Coding (DPC) in which transmitted signals are correlated with the CSI. The correlation coefficient is to be determined by C_P₁. In contrast to the DM model, the LB on the forward key capacity of a Gaussian model is a strictly increasing function of C_P₁ according to our simulations. This fact is an essential difference between this model and the DM model. For C_P₁=0 and a fixed Λ, the forward key capacity of the Gaussian model exceeds the main channel capacity of the G-SWC in the low SIR regime. By simulations, we show that the interference enhances key generation in the low SIR regime. In this regime, we also justify that the positive effect of the interference on the (forward) key capacity is generally more than its positive effect on the secrecy capacity of the G-SWC, while the interference has no influence on the main channel capacity of the G-SWC.

This dissertation addresses multi-layer optimization aspects of multiple input multiple output (MIMO) and deep learning-based communication systems. The initial focus is on the rate optimization for multi-user MIMO (MU-MIMO) configurations; specifically, multiple access channel (MAC) and interference channel (IC). First, the ergodic sum rates of MIMO MAC and IC configurations are determined by jointly integrating the error and overhead effects due to channel estimation (training) and feedback into the rate optimization. Then, we investigated methods that will increase the achievable rate for parallel Gaussian IC (PGIC) which is a special case of MIMO IC where there is no interference between multiple antenna elements. We derive a generalized iterative waterfilling algorithm for power allocation that maximizes the ergodic achievable rate. We verified the sum rate improvement with our proposed scheme through extensive simulation tests. Next, we introduce a novel physical layer scheme for single user MIMO spatial multiplexing systems based on unsupervised deep learning using an autoencoder. Both transmitter and receiver are designed as feedforward neural networks (FNN) and constellation diagrams are optimized to minimize the symbol error rate (SER) based on the channel characteristics. We first evaluate the SER in the presence of a constant Rayleigh-fading channel as a performance upper bound. Then, we quantize the Gaussian distribution and train the autoencoder with multiple quantized channel matrices. The channel is provided as an input to both the transmitter and the receiver. The performance exceeds that of conventional communication systems both when the autoencoder is trained and tested with single and multiple channels and the performance gain is sustained after accounting for the channel estimation error. Moreover, we evaluate the performance with increasing number of quantization points and when there is a difference between training and test channels. We show that the performance loss is minimal when training is performed with sufficiently large number of quantization points and number of channels. Finally, we develop a distributed and decentralized MU-MIMO link selection and activation protocol that enables MU-MIMO operation in wireless networks. We verified the performance gains with the proposed protocol in terms of average network throughput. / Doctor of Philosophy / Multiple Input Multiple Output (MIMO) wireless systems include multiple antennas both at the transmitter and receiver and they are widely used today in cellular and wireless local area network systems to increase robustness, reliability and data rate. Multi-user MIMO (MU-MIMO) configurations include multiple access channel (MAC) where multiple transmitters communicate simultaneously with a single receiver; interference channel (IC) where multiple transmitters communicate simultaneously with their intended receivers; and broadcast channel (BC) where a single transmitter communicates simultaneously with multiple receivers. Channel state information (CSI) is required at the transmitter to precode the signal and mitigate interference effects. This requires CSI to be estimated at the receiver and transmitted back to the transmitter in a feedback loop. Errors occur during both channel estimation and feedback processes. We initially analyze the achievable rate of MAC and IC configurations when both channel estimation and feedback errors are taken into account in the capacity formulations. We treat the errors associated with channel estimation and feedback as additional noise. Next, we develop methods to maximize the achievable rate for IC by using interference cancellation techniques at the receivers when the interference is very strong. We consider parallel Gaussian IC (PGIC) which is a special case of MIMO IC where there is no interference between multiple antenna elements. We develop a power allocation scheme which maximizes the ergodic achievable rate of the communication systems. We verify the performance improvement with our proposed scheme through simulation tests. Standard optimization techniques are used to determine the fundamental limits of MIMO communications systems. However, there is still a gap between current operational systems and these limits due to complexity of these solutions and limitations in their assumptions. Next, we introduce a novel physical layer scheme for MIMO systems based on machine learning; specifically, unsupervised deep learning using an autoencoder. An autoencoder consists of an encoder and a decoder that compresses and decompresses data, respectively. We designed both the encoder and the decoder as feedforward neural networks (FNNs). In our case, encoder performs transmitter functionalities such as modulation and error correction coding and decoder performs receiver functionalities such as demodulation and decoding as part of the communication system. Channel is included as an additional layer between the encoder and decoder. By incorporating the channel effects in the design process of the autoencoder and jointly optimizing the transmitter and receiver, we demonstrate the performance gains over conventional MIMO communication schemes. Finally, we develop a distributed and decentralized MU-MIMO link selection and activation protocol that enables MU-MIMO operation in wireless networks. We verified the performance gains with the proposed protocol in terms of average network throughput.

Le spectre électromagnétique est une ressource naturelle dont l'usage doit être optimisé. Un grand nombre de travaux actuels visent à améliorer l'utilisation des fréquences radio en y introduisant un degré de flexibilité rendu possible par l'agilité en forme d'onde et en fréquence permise par la radio logicielle (SDR), ainsi que par les méthodes de traitement intelligent du signal (radio cognitive). Cette thèse se place dans ce contexte. Concrètement, nous considérons le problème de partage du spectre électromagnétique entre plusieurs utilisateurs sous contraintes d'interférence mutuelle. Notre objectif est de contribuer à l'évaluation du gain du partage de cette ressource rare qu'est le spectre électromagnétique. En étudiant le canal gaussien d'interférence avec l'interférence traitée comme du bruit additif gaussien aux différents récepteurs, nous avons trouvé une description géométrique et plusieurs caractérisations de la région des débits atteignables. Ensuite, considérant un cas plus réaliste où chaque utilisateur a une certaine qualité de service, nous avons trouvé une condition nécessaire et suffisante pour permettre la communication simultanée à travers le canal gaussien d'interférence pour deux utilisateurs. Dans un scénario de partage entre un utilisateur primaire ayant une plus grande priorité d'accès au spectre et un utilisateur secondaire, après avoir déterminé des bornes minimales pour le débit du primaire en fonction du schéma d'allocation de puissance de l'utilisateur secondaire, nous avons proposé une technique originale d'allocation de puissance pour l'utilisateur secondaire accédant de manière opportuniste au spectre sous contraintes de performance de coupure pour tous les utilisateurs. En particulier, cette technique d'allocation de puissance n'utilise que l'information sur l'état des canaux des liens directs allant de l'émetteur secondaire vers les autres points du réseau. Finalement, considérant des modèles de canaux plus réalistes; après avoir montré l'existence d'une zone d'exclusion autour du récepteur primaire (zone où il n'y a aucun transmetteur secondaire, dans le but de protéger l'utilisateur primaire contre les fortes interférences), nous avons caractérisé l'effet du shadowing et du path-loss sur cette zone d'exclusion du primaire. / In this thesis, we address the problem of spectrum-sharing for wireless communication where multiple users attempt to access a common spectrum resource under mutual interference constraints. Our objective is to evaluate the gains of sharing by investigating different scenarios of spectrum access. Studying the Gaussian Interference Channel with interferences considered as noise, we found a geometrical description and several characteristics of the achievable rate region. Considering a more realistic scenario, with each user having a certain QoS, we found necessary and sufficient condition to be fulfilled for simultaneous communication over the two-user Gaussian Interference Channel. Furthermore, we proposed two lower bounds for a single-primary-user mean rate, depending on the secondary user power control scheme. Specially, we investigated an original power control policy, for a secondary user, under outage performance requirement for both users and partial knowledge of the channel state information. Finally, considering a spectrums-haring with a licensee or primary user and several secondary or cognitive users, we showed the existence of an exclusive region around the primary receiver and we characterized the effects of shadowing and path-loss on this exclusive region (or no-talk zone).

Dans cette thèse, nous menons une étude dans le cadre de la théorie de l'information sur deux questions importantes de la communication sans fil : l'amélioration du débit de données dans les réseaux avec interférence grâce à la coopération entre utilisateurs et le renforcement de la sécurité des transmissions à l'aide d'un signal de rétroaction.Dans la première partie de la thèse, nous nous concentrons sur le modèle le plus simple qui intègre à la fois l'interférence et la coopération, le canal à relais et interférence ou IRC (Interference Relay Channel). Notre objectif est de caractériser dans un nombre fixe de bits la région de capacité du IRC gaussien. À cette fin, nous dérivons une nouvelle limite supérieure de la capacité et deux stratégies de transmission. La limite supérieure est notamment obtenue grâce à une extension non triviale que nous proposons, de la classe de canaux semi-déterministe et injective à l'origine dérivée par Telatar et Tse pour le canal à interférence.Dans la seconde partie, nous étudions le canal avec espion et rétroaction généralisée ou WCGF (Wiretap Channel with Generalized Feedback). Notre objectif est de développer une stratégie de transmission générale qui englobe les résultats existants pour les différents modèles de rétroaction trouvés dans la littérature. À cette fin, nous proposons deux stratégies de transmission différentes sur la capacité du WCGF sans mémoire. Nous dérivons d'abord une stratégie qui est basée sur le codage source-canal conjoint. Nous introduisons ensuite une seconde stratégie où le signal de rétroaction est utilisé pour générer une clé secrète qui permet de chiffrer le message partiellement ou totalement. / In this thesis, we conduct an information-theoretic study on two important aspects of wireless communications: the improvement of data throughput in interference-limited networks by means of cooperation between users and the strengthening of the security of transmissions with the help of feedback.In the first part of the thesis, we focus on the simplest model that encompasses interference and cooperation, the Interference Relay Channel (IRC). Our goal is to characterize within a fixed number of bits the capacity region of the Gaussian IRC, independent of any channel conditions. To do so, we derive a novel outer bound and two inner bounds. Specifically, the outer bound is obtained thanks to a nontrivial extension we propose of the injective semideterministic class of channels, originally derived by Telatar and Tse for the Interference Channel (IC).In the second part of the thesis, we investigate the Wiretap Channel with Generalized Feedback (WCGF) and our goal is to provide a general transmission strategy that encompasses the existing results for different feedback models found in the literature. To this end, we propose two different inner bounds on the capacity of the memoryless WCGF. We first derive an inner bound that is based on the use of joint source-channel coding, which introduces time dependencies between the feedback outputs and the channel inputs through different time blocks. We then introduce a second inner bound where the feedback link is used to generate a key that encrypts the message partially or completely.

Dans cette thèse, nous menons une étude dans le cadre de la théorie de l'information sur deux questions importantes de la communication sans fil : l'amélioration du débit de données dans les réseaux avec interférence grâce à la coopération entre utilisateurs et le renforcement de la sécurité des transmissions à l'aide d'un signal de rétroaction.Dans la première partie de la thèse, nous nous concentrons sur le modèle le plus simple qui intègre à la fois l'interférence et la coopération, le canal à relais et interférence ou IRC (Interference Relay Channel). Notre objectif est de caractériser dans un nombre fixe de bits la région de capacité du IRC gaussien. À cette fin, nous dérivons une nouvelle limite supérieure de la capacité et deux stratégies de transmission. La limite supérieure est notamment obtenue grâce à une extension non triviale que nous proposons, de la classe de canaux semi-déterministe et injective à l'origine dérivée par Telatar et Tse pour le canal à interférence.Dans la seconde partie, nous étudions le canal avec espion et rétroaction généralisée ou WCGF (Wiretap Channel with Generalized Feedback). Notre objectif est de développer une stratégie de transmission générale qui englobe les résultats existants pour les différents modèles de rétroaction trouvés dans la littérature. À cette fin, nous proposons deux stratégies de transmission différentes sur la capacité du WCGF sans mémoire. Nous dérivons d'abord une stratégie qui est basée sur le codage source-canal conjoint. Nous introduisons ensuite une seconde stratégie où le signal de rétroaction est utilisé pour générer une clé secrète qui permet de chiffrer le message partiellement ou totalement. / In this thesis, we conduct an information-theoretic study on two important aspects of wireless communications: the improvement of data throughput in interference-limited networks by means of cooperation between users and the strengthening of the security of transmissions with the help of feedback.In the first part of the thesis, we focus on the simplest model that encompasses interference and cooperation, the Interference Relay Channel (IRC). Our goal is to characterize within a fixed number of bits the capacity region of the Gaussian IRC, independent of any channel conditions. To do so, we derive a novel outer bound and two inner bounds. Specifically, the outer bound is obtained thanks to a nontrivial extension we propose of the injective semideterministic class of channels, originally derived by Telatar and Tse for the Interference Channel (IC).In the second part of the thesis, we investigate the Wiretap Channel with Generalized Feedback (WCGF) and our goal is to provide a general transmission strategy that encompasses the existing results for different feedback models found in the literature. To this end, we propose two different inner bounds on the capacity of the memoryless WCGF. We first derive an inner bound that is based on the use of joint source-channel coding, which introduces time dependencies between the feedback outputs and the channel inputs through different time blocks. We then introduce a second inner bound where the feedback link is used to generate a key that encrypts the message partially or completely.

The central objective of this thesis is to suggest and develop one simple, unified method for communication over optical fiber networks, valid for all values of dispersion and nonlinearity parameters, and for a single-user channel or a multiple-user network. The method is based on the nonlinear Fourier transform (NFT), a powerful tool in soliton theory and exactly solvable models for solving integrable partial differential equations governing wave propagation in certain nonlinear media. The NFT decorrelates signal degrees of freedom in such models, in much the same way that the Fourier transform does for linear systems. In this thesis, this observation is exploited for data transmission over integrable channels such as optical fibers, where pulse propagation is governed by the nonlinear Schr\"odinger (NLS) equation. In this transmission scheme, which can be viewed as a nonlinear analogue of orthogonal frequency-division multiplexing commonly used in linear channels, information is encoded in the nonlinear spectrum of the signal. Just as the (ordinary) Fourier transform converts a linear convolutional channel into a number of parallel scalar channels, the nonlinear Fourier transform converts a nonlinear dispersive channel described by a \emph{Lax convolution} into a number of parallel scalar channels. Since, in the spectral coordinates the NLS equation is multiplicative, users of a network can operate in independent nonlinear frequency bands with no deterministic inter-channel interference. Unlike most other fiber-optic transmission schemes, this technique deals with both dispersion and nonlinearity directly and unconditionally without the need for dispersion or nonlinearity compensation methods. This thesis lays the foundations of such a nonlinear frequency-division multiplexing system.

The central objective of this thesis is to suggest and develop one simple, unified method for communication over optical fiber networks, valid for all values of dispersion and nonlinearity parameters, and for a single-user channel or a multiple-user network. The method is based on the nonlinear Fourier transform (NFT), a powerful tool in soliton theory and exactly solvable models for solving integrable partial differential equations governing wave propagation in certain nonlinear media. The NFT decorrelates signal degrees of freedom in such models, in much the same way that the Fourier transform does for linear systems. In this thesis, this observation is exploited for data transmission over integrable channels such as optical fibers, where pulse propagation is governed by the nonlinear Schr\"odinger (NLS) equation. In this transmission scheme, which can be viewed as a nonlinear analogue of orthogonal frequency-division multiplexing commonly used in linear channels, information is encoded in the nonlinear spectrum of the signal. Just as the (ordinary) Fourier transform converts a linear convolutional channel into a number of parallel scalar channels, the nonlinear Fourier transform converts a nonlinear dispersive channel described by a \emph{Lax convolution} into a number of parallel scalar channels. Since, in the spectral coordinates the NLS equation is multiplicative, users of a network can operate in independent nonlinear frequency bands with no deterministic inter-channel interference. Unlike most other fiber-optic transmission schemes, this technique deals with both dispersion and nonlinearity directly and unconditionally without the need for dispersion or nonlinearity compensation methods. This thesis lays the foundations of such a nonlinear frequency-division multiplexing system.

Multiple users compete for a common resource like bandwidth to communicate data in interference networks. Existing approaches in dealing with interference limit the rate of communication due to paucity of shared resources. This limitation in the rate gets more glaring as the number of users in the network increases. For example, existing wireless systems either choose to orthogonalize the users (for example, Frequency Division Multiple Access (FDMA) systems or Code Division Multiple Access (CDMA) systems) or treat interference as Gaussian noise at the receivers. It is well known that these approaches are sub-optimal in general. Orthogonalization of users limit the number of available interference-free channels (known as degrees of freedom, abbreviated as DoF) and treating interference as noise means that the receiver cannot make use of the structure in the interfering signals. This motivates the need to analyze alternate transmit and decoding schemes in interference networks. This thesis mainly analyzes transmit schemes that use linear precoding for various configurations of interference networks with some practical constraints imposed by the use of finite input constellations, propagation delays, and channel state availability at the transmitters. The main contributions of this thesis are listed below. Achievable rates using precoding with finite constellation inputs in Gaussian Interference Channels (GIC) is analyzed. A metric for finding the approximate angle of rotation to maximally enlarge the Constellation Constrained (CC) capacity of two-user Gaussian Strong Interference Channel (GSIC) is proposed. Even as the Gaussian alphabet FDMA rate curve touches the capacity curve of the GSIC, with both the users using the same finite constellation, we show that the CC FDMA rate curve lies strictly inside the CC capacity curve at high powers. For a K-user MIMO GIC, a set of necessary and sufficient conditions on the precoders under which the mutual information between between relevant transmit-receive pairs saturate like in the single user case is derived. Gradient-ascent based algorithms to optimize the sum-rate achieved by precoding with finite constellation inputs and treating interference as noise are proposed. For a class of Gaussian interference networks with general message demands, identified as symmetrically connected interference networks, the expected sumspectral efficiency (in bits/sec/Hz) is shown to grow linearly with the number of transmitters at finite SNR, using a time-domain Interference Alignment (IA) scheme in the presence of line of sight (LOS) channels. For a 2×2 MIMO X-Network with M antennas at each node, we identify spacetime block codes that could be coupled with an appropriate precoding scheme to achieve the maximum possible sum-DoF of 4M 3 , for M = 3, 4. The proposed schemes are shown to achieve a diversity gain of M with SNR-independent finite constellation inputs. The proposed schemes have lower CSIT requirements compared to existing schemes. This thesis also makes an attempt to guarantee a minimum throughput when the zero-interference conditions cannot be satisfied in a wireline network with three unicast sessions with delays, using Precoding Based Network Alignment (PBNA). Three different PBNA schemes namely PBNA with time-varying local encoding coefficients (LECs), PBNA using transform approach and time-invariant LECs, and PBNA using transform approach and block time-varying LECs are proposed and their feasibility conditions analyzed.

The data traffic has been growing drastic over the past few years due to digitization and new technologies that are introduced to the market, such as autonomous cars. In order to meet this demand, the MIMO-OFDM system is used in the fifth generation wireless network, 5G. Designing the optimal wireless network is currently the main research within the area of telecommunication. In order to achieve such a system, multiple factors has to be taken into account, such as the suppression of interference from other users. A traditional method called linear minimum mean square error filter is currently used to suppress the interferences. To derive such a filter, a selection of parameters has to be estimated. One of these parameters is the ideal interference plus noise covariance matrix. By gathering prior knowledge of the underlying structure of the interference channel matrices in terms of the number of interferers and their corresponding bandwidths, the estimation of the ideal covariance matrix could be facilitated. As for this thesis, machine learning algorithms were used to extract these prior knowledge. More specifically, a two or three hidden layer feedforward neural network and a support vector machine with a linear kernel was used. The empirical findings implies promising results with accuracies above 95% for each model. / Under de senaste åren har dataanvändningen ökat drastiskt på grund av digitaliseringen och allteftersom nya teknologier introduceras på marknaden, exempelvis självkörande bilar. För att bemöta denna efterfrågan används ett s.k. MIMO-OFDM system i den femte generationens trådlösa nätverk, 5G. Att designa det optimala trådlösa nätverket är för närvarande huvudforskningen inom telekommunikation och för att uppnå ett sådant system måste flera faktorer beaktas, bland annat störningar från andra användare. En traditionell metod som används för att dämpa störningarna kallas för linjära minsta medelkvadratfelsfilter. För att hitta ett sådant filter måste flera olika parametrar estimeras, en av dessa är den ideala störning samt bruskovariansmatrisen. Genom att ta reda på den underliggande strukturen i störningsmatriserna i termer av antal störningar samt deras motsvarande bandbredd, är något som underlättar uppskattningen av den ideala kovariansmatrisen. I följande avhandling har olika maskininlärningsalgoritmer applicerats för att extrahera dessa informationer. Mer specifikt, ett neuralt nätverk med två eller tre gömda lager samt stödvektormaskin med en linjär kärna har använts. De slutliga resultaten är lovande med en noggrannhet på minst 95% för respektive modell.