A multiple-input-multiple-output, or MIMO, receiver receives multiple data streams in the same frequency band at the same time, significantly improving spectral efficiency. It has to preserve all the antenna aperture information and use it to deliver as many data streams as the antenna count. As the number of antennas increases, implementing a MIMO receiver system in the analog domain becomes difficult. A digital MIMO receiver architecture that digitizes all the antenna inputs on the element level offers multiple advantages. Digital MIMO signal processing is flexing and powerful. Complex space-time array processing is supported and so is digital array calibration. Therefore, the digital MIMO receiver architecture has become the most promising architecture for future massive MIMO systems. However, the digital MIMO receiver architecture has a disadvantage, namely that the spatial selectivity feature is missing in the RF/analog domain. At the target frequency band, multiple spatial signals can arrive at the antenna array at different power levels. Conventional spectral filtering is ineffective at in-band frequency so all the spatial signals have to co-exist in all the receiver elements and the following analog-to-digital converters (A/Ds). The instantaneous dynamic range required for these RF/analog and mixed-signal circuits will be limited by the strongest spatial signal on the upper bound, and the weakest spatial signal on the lower bound. A high instantaneous dynamic range requirement directly translates to high power consumption and high cost. Therefore, the recovery of spatial selectivity in the RF/analog domain is necessary. The first thrust toward recovering RF/analog spatial selectivity in a digital MIMO receiver is the scalable spatial notch suppression technique. Knowing the direction of a strong spatial blocker, a spatial notch, instead of beams, can be synthesized to the blocker direction to filter it out. This means that all the analog baseband outputs will show high conversion gains to signals from all directions but one, namely the blocker direction. In this way, high sensitivity is preserved in most directions to receiver multiple weak spatial signals simultaneously, which will be digitized, and separated in the digital domain. In the blocker direction, a low conversion gain filters the blocker out, preventing it from demanding high dynamic range for all of the RF/analog circuits and the A/Ds. In order to synthesize the scalable spatial notch, a spatial notch filter (SNF) is designed to provide lower input impedance in the blocker direction and high impedance in other directions. Using this spatially modulated impedance to load a current mode receiver leads to spatially modulated conversion gain. A transparent RF front-end translates this impedance to the antenna interface to achieve spatial notch suppression right at the antennas. A feedforward spatial notch canceler (FF SNC) uses the available isolated blocker information to improve spatial suppression ratio. The spatial notch suppression is scalable through a baseband node, allowing the tiling of multiple ICs on the same PCB for larger scale MIMO systems. A prototype receiver array was implemented with a 65nm CMOS process. Experimental results showed 32dB steerable spatial notch suppression, more than 19db of suppression inside the notch direction across all frequencies. In-band output-referred IP3 was improved from -10dBV to +24dBV, from outside to inside the notch direction, and IIP3 was also improved from +11dBm to +18dBm. Single-element equivalent double-sideband noise figure (NFDSB,eq) was 2.2 to 4.6dB across the 0.1 to 1.7GHz operating frequency range, also showing an improvement compared to other multi-antenna receivers at similar frequency ranges. A second thrust is an RF/analog arbitrary spatial filtering receiver. Instead of filtering out strong spatial blockers, a more general and robust way to recover spatial selectivity is to impose an arbitrary spatial response that adaptively equalizes the power levels of all the spatial signals. In this way, all the spatial signals should have the same power level when reaching the A/Ds, allowing the use of low-power A/Ds with low dynamic ranges, which are essential for the realization of the digital massive MIMO solution. Such an arbitrary spatial filtering response requires the ability to synthesize multiple spatial notches that can be independently steered, the depth of the notches free adjusted. In addition, a few performance metrics need to be improved based on the first work. Spatial suppression ratio was limited by the lack of magnitude control in the first work. In-band in-notch linearity performance was limited by the use of voltage mode gyrators that requires a band-limiting high-impedance node, which also limits spatial suppression bandwidth. Also, the antenna array dimensions scale inversely with operating frequency. So pushing the receiver array to work at higher frequency is also desired. Toward these goals, a 65nm CMOS prototype receiver array was implemented. Wideband current-mode receiver front-ends that consist of inverter-based LNTAs and passive mixers can work up to 3.1GHz. A baseband current-mode beamformer can synthesize virtual grounds at the output nodes in the target notch directions, providing not only an arbitrary spatial response but also an baseband input impedance that is also spatially modulated, allowing spatial filtering at the LNTA output nodes. Current mode operation avoids the use of band-limiting high impedance nodes for strong spatial signals, leading to superior linearity and wideband spatial suppression. This 4-element prototype measured more than 50dB of spatial suppression ratios with single-notch settings across all measured directions. Up to three notches can be synthesized, each of which can be independently steered and its depth freely adjusted. An in-band OIP3 of +34dBV was measured, 10dB higher than the first work, due to the current mode operation. A 20dB suppression bandwidth of 320MHz, or equivalently 64% was measured, more than 20× improvement than the first work, also due to the current mode operation. On a separate note, an ultra-wideband LNTA was also designed for an RF channelizing receiver work. This two-stage LNTA makes use of a gm-boosted current mirror structure to harness the linearity advantage of a current mirror, the low-noise input matching of the feedback structure, the high transconductance gain of a two-stage structure and an ultra-wideband input matching advantage of a gyrator. The implemented 65nm CMOS prototype is fully integrated, and provides 242mS peak transconductance gain over 0.6-9.6GHz operating frequency range. It achieves 4.5dB of NF and +6.5dBm of IIP3. In summary, RF/analog spatial selectivity can be recovered in innovative methods to relax the dynamic range requirement for all the RF/analog circuits together with the following A/Ds in a digital MIMO receiver. The scalable spatial notch suppression technique and the arbitrary spatial filtering technique allow the use of low-power A/Ds, which are essential for truly massive MIMO systems with manageable power consumption.

由於在盲語音分離的應用，線性準平穩源訊號混合的盲識別獲得了巨大的研究興趣。在這個問題上，我們利用準穩態源訊號的時變特性來識別未知的混合系統系數。傳統的方法有二：i)基於張量分解的平行因子分析(PARAFAC);ii)基於對多個矩陣的聯合對角化的聯合對角化算法(JD)。一般來說，PARAFAC和JD 都採用了源聯合的提取方法；即是說，對應所有訊號源的系統係數在升法上是用時進行識別的。 / 在這篇論文中，我利用Khati-Rao(KR)子空間來設計一種新的盲識別算法。在我設計的算法中提出一種與傳統的方法不同的提法。在我設計的算法中，盲識別問題被分解成數個結構上相對簡單的子問題，分別對應不同的源。在超定混合模型，我們提出了一個專門的交替投影算法(AP)。由此產生的算法，不但能從經驗發現是非常有競爭力的，而且更有理論上的利落收斂保證。另外，作為一個有趣的延伸，該算法可循一個簡單的方式應用於欠混合模型。對於欠定混合模型，我們提出啟發式的秩最小化算法從而提高算法的速度。 / Blind identification of linear instantaneous mixtures of quasi-stationary sources (BI-QSS) has received great research interest over the past few decades, motivated by its application in blind speech separation. In this problem, we identify the unknown mixing system coefcients by exploiting the time-varying characteristics of quasi-stationary sources. Traditional BI-QSS methods fall into two main categories: i) Parallel Factor Analysis (PARAFAC), which is based on tensor decomposition; ii) Joint Diagonalization (JD), which is based on approximate joint diagonalization of multiple matrices. In both PARAFAC and JD, the joint-source formulation is used in general; i.e., the algorithms are designed to identify the whole mixing system simultaneously. / In this thesis, I devise a novel blind identification framework using a Khatri-Rao (KR) subspace formulation. The proposed formulation is different from the traditional formulations in that it decomposes the blind identication problem into a number of per-source, structurally less complex subproblems. For the over determined mixing models, a specialized alternating projections algorithm is proposed for the KR subspace for¬mulation. The resulting algorithm is not only empirically found to be very competitive, but also has a theoretically neat convergence guarantee. Even better, the proposed algorithm can be applied to the underdetermined mixing models in a straightforward manner. Rank minimization heuristics are proposed to speed up the algorithm for the underdetermined mixing model. The advantages on employing the rank minimization heuristics are demonstrated by simulations. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Lee, Ka Kit. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 72-76). / Abstracts also in Chinese. / Abstract --- p.i / Acknowledgement --- p.ii / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Settings of Quasi-Stationary Signals based Blind Identification --- p.4 / Chapter 2.1 --- Signal Model --- p.4 / Chapter 2.2 --- Assumptions --- p.5 / Chapter 2.3 --- Local Covariance Model --- p.7 / Chapter 2.4 --- Noise Covariance Removal --- p.8 / Chapter 2.5 --- Prewhitening --- p.9 / Chapter 2.6 --- Summary --- p.10 / Chapter 3 --- Review on Some Existing BI-QSS Algorithms --- p.11 / Chapter 3.1 --- Joint Diagonalization --- p.11 / Chapter 3.1.1 --- Fast Frobenius Diagonalization [4] --- p.12 / Chapter 3.1.2 --- Pham’s JD [5, 6] --- p.14 / Chapter 3.2 --- Parallel Factor Analysis --- p.16 / Chapter 3.2.1 --- Tensor Decomposition [37] --- p.17 / Chapter 3.2.2 --- Alternating-Columns Diagonal-Centers [12] --- p.21 / Chapter 3.2.3 --- Trilinear Alternating Least-Squares [10, 11] --- p.23 / Chapter 3.3 --- Summary --- p.25 / Chapter 4 --- Proposed Algorithms --- p.26 / Chapter 4.1 --- KR Subspace Criterion --- p.27 / Chapter 4.2 --- Blind Identification using Alternating Projections --- p.29 / Chapter 4.2.1 --- All-Columns Identification --- p.31 / Chapter 4.3 --- Overdetermined Mixing Models (N > K): Prewhitened Alternating Projection Algorithm (PAPA) --- p.32 / Chapter 4.4 --- Underdetermined Mixing Models (N <K) --- p.34 / Chapter 4.4.1 --- Rank Minimization Heuristic --- p.34 / Chapter 4.4.2 --- Alternating Projections Algorithm with Huber Function Regularization --- p.37 / Chapter 4.5 --- Robust KR Subspace Extraction --- p.40 / Chapter 4.6 --- Summary --- p.44 / Chapter 5 --- Simulation Results --- p.47 / Chapter 5.1 --- General Settings --- p.47 / Chapter 5.2 --- Overdetermined Mixing Models --- p.49 / Chapter 5.2.1 --- Simulation 1 - Performance w.r.t. SNR --- p.49 / Chapter 5.2.2 --- Simulation 2 - Performance w.r.t. the Number of Available Frames M --- p.49 / Chapter 5.2.3 --- Simulation 3 - Performance w.r.t. the Number of Sources K --- p.50 / Chapter 5.3 --- Underdetermined Mixing Models --- p.52 / Chapter 5.3.1 --- Simulation 1 - Success Rate of KR Huber --- p.53 / Chapter 5.3.2 --- Simulation 2 - Performance w.r.t. SNR --- p.54 / Chapter 5.3.3 --- Simulation 3 - Performance w.r.t. M --- p.54 / Chapter 5.3.4 --- Simulation 4 - Performance w.r.t. N --- p.56 / Chapter 5.4 --- Summary --- p.56 / Chapter 6 --- Conclusion and Future Works --- p.58 / Chapter A --- Convolutive Mixing Model --- p.60 / Chapter B --- Proofs --- p.63 / Chapter B.1 --- Proof of Theorem 4.1 --- p.63 / Chapter B.2 --- Proof of Theorem 4.2 --- p.65 / Chapter B.3 --- Proof of Observation 4.1 --- p.65 / Chapter B.4 --- Proof of Proposition 4.1 --- p.66 / Chapter C --- Singular Value Thresholding --- p.67 / Chapter D --- Categories of Speech Sounds and Their Impact on SOSs-based BI-QSS Algorithms --- p.69 / Chapter D.1 --- Vowels --- p.69 / Chapter D.2 --- Consonants --- p.69 / Chapter D.1 --- Silent Pauses --- p.70 / Bibliography --- p.72

Different from multiple-input multiple-output (MIMO) systems, a major challenge for cooperative communications is the problem of synchronization because multiple transmissions undertaken by cooperative systems may not be synchronized in time and/or frequency. With synchronization errors, conventional space-time (ST) codes may not be directly applicable any longer. To tackle the problem of timing synchronization, space-frequency (SF) coded orthogonal frequency division multiplexing (OFDM) cooperative systems have recently been proposed to achieve asynchronous diversity due to their insensitivity to timing errors. However, these systems still need to face the problem of multiple carrier frequency offsets (CFOs). Since each node in a cooperative system is equipped with its own oscillator, the received signals from different relay nodes may have multiple CFOs which cannot be compensated simultaneously at the destination node. For SF coded OFDM cooperative systems, this problem becomes more complicated because CFOs can lead to inter-carrier interference (ICI). To address this challenge, in this thesis we consider the signal detection problem in cooperative systems having multiple CFOs. / First, we investigate the effect of multiple CFOs on two classic ST codes. They are delay diversity and the Alamouti code. For delay diversity, we find that both its achieved diversity order and diversity product are not decreased by multiple CFOs arising from maximum-likelihood (ML) detection. For the Alamouti code, the diversity product may be decreased by multiple CFOs. In the worst case situation, full diversity order 2 cannot be achieved. / For deeper insights into the SF coded communication system with multiple CFOs, we then carry out diversity analysis. By treating the CFOs as part of the SF codeword matrix, we show that if all the absolute values of normalized CFOs are less than 0.5, then the full diversity order for the SF codes are not affected by the multiple CFOs in the SF coded OFDM cooperative system. We further prove that this full diversity property can still be preserved if the zero forcing (ZF) method is used to equalize the multiple CFOs. This method, by some reasonable approximations, is actually equivalent to the MMSE-F detection method. To improve the robustness of the SF codes to multiple CFOs, we propose a novel permutation method. With this method, the achieved diversity order of SF codes remains the same even when the absolute values of normalized CFOs are equal to or greater than 0.5. To reduce computational complexity, we further propose two full diversity achievable detection methods, namely the ZF-ML-Zn and ZF-ML-PIC detection methods, which are suitable for the case when the ICI matrix is singular. / In summary, in this study, we demonstrate that with proper design, the SF coded OFDM approach can be made robust to both timing errors and CFOs in a cooperative communication system. / Since OFDM systems are robust to timing errors, we turn to an SF coded cooperative communication system with multiple CFOs, where the SF codes are rotational based and can achieve both full cooperative and full multipath diversity orders. We begin with the traditional way of ICI mitigation. To preserve the performance of the SF code, we suggest increasing the SINR of each subcarrier but not equalizing the SF precoding matrix. By exploiting the structure of the SF codes, we propose three signal detection methods to deal with the multiple CFOs problem in SF coded OFDM systems. They are the minimum mean-squared filtering (MMSE-F) method, the two-stage simple frequency shift Q taps (FS-Q-T) method, and the multiple fast Fourier transform (M-FFT) method, all of which offer different tradeoffs between performance and computational complexity. Our simulation results indicate that the proposed detection methods perform well as long as the CFOs between nodes are small. / Tian, Feng. / Adviser: Ching Pak-Chung. / Source: Dissertation Abstracts International, Volume: 71-01, Section: B, page: 0559. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2009. / Includes bibliographical references (leaves 146-160). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese.

Multiple-input multiple-output (MIMO) systems are a mature technology that has been incorporated into current wireless broadband standards to improve the channel capacity and link reliability. Nevertheless, due to the continuous increasing demand for wireless data traffic new strategies are to be adopted. Very large MIMO antenna arrays represents a paradigm shift in terms of theory and implementation, where the use of tens or hundreds of antennas provides significant improvements in throughput and radiated energy efficiency compared to single antennas setups. Since design constraints limit the number of usable antennas, virtual systems can be seen as a promising technique due to their ability to mimic and exploit the gains of multi-antenna systems by means of wireless cooperation. Considering these arguments, in this work, energy efficient coding and network design for large virtual MIMO systems are presented. Firstly, a cooperative virtual MIMO (V-MIMO) system that uses a large multi-antenna transmitter and implements compress-and-forward (CF) relay cooperation is investigated. Since constructing a reliable codebook is the most computationally complex task performed by the relay nodes in CF cooperation, reduced complexity quantisation techniques are introduced. The analysis is focused on the block error probability (BLER) and the computational complexity for the uniform scalar quantiser (U-SQ) and the Lloyd-Max algorithm (LM-SQ). Numerical results show that the LM-SQ is simpler to design and can achieve a BLER performance comparable to the optimal vector quantiser. Furthermore, due to its low complexity, U-SQ could be consider particularly suitable for very large wireless systems. Even though very large MIMO systems enhance the spectral efficiency of wireless networks, this comes at the expense of linearly increasing the power consumption due to the use of multiple radio frequency chains to support the antennas. Thus, the energy efficiency and throughput of the cooperative V-MIMO system are analysed and the impact of the imperfect channel state information (CSI) on the system's performance is studied. Finally, a power allocation algorithm is implemented to reduce the total power consumption. Simulation results show that wireless cooperation between users is more energy efficient than using a high modulation order transmission and that the larger the number of transmit antennas the lower the impact of the imperfect CSI on the system's performance. Finally, the application of cooperative systems is extended to wireless self-backhauling heterogeneous networks, where the decode-and-forward (DF) protocol is employed to provide a cost-effective and reliable backhaul. The associated trade-offs for a heterogeneous network with inhomogeneous user distributions are investigated through the use of sleeping strategies. Three different policies for switching-off base stations are considered: random, load-based and greedy algorithms. The probability of coverage for the random and load-based sleeping policies is derived. Moreover, an energy efficient base station deployment and operation approach is presented. Numerical results show that the average number of base stations required to support the traffic load at peak-time can be reduced by using the greedy algorithm for base station deployment and that highly clustered networks exhibit a smaller average serving distance and thus, a better probability of coverage.

This thesis addresses and identifies outstanding challenges associated with the Multi user massive Multiple-Input Multiple-Output (MU massive MIMO) transmission, whereby various system scenarios have been considered to tackle these challenges. First, for a single cell scenario, the uplink effective capacity under statistical exponent constraints, the asymptotic error and outage probabilities in a multi user massive MIMO system are provided. The proposed approach establishes closed form expressions for the aforementioned metrics under both perfect and imperfect channel state information (CSI) scenarios. In addition, expressions for the asymptotically high signal-to-interference ratio (SIR) regimes are established. Second, the statistical queueing constraints, pilot contamination phenomenon and fractional power control in random or irregular cellular massive MIMO system are investigated, where base station locations are modelled based on the Poisson point process. Specifically, tractable analytical expressions are developed for the asymptotic SIR coverage, rate coverage and the effective capacity under the quality of service statistical exponent constraint. Laplace transform of interference is derived with the aid of mathematical tools from stochastic geometry. Simulation outcomes demonstrate that pilot reuse impairments can be alleviated by employing a cellular frequency reuse scheme. For example, with unity frequency reuse factor, we see that 40% of the total users have SIR above −10.5dB, whereas, with a reuse factor of 7, the same fraction of users have SIR above 20.5dB. In addition, for a certain parameters setting, the coverage probability in the lower 50th percentile can be maximized by adjusting power compensation fraction between 0.2 and 0.5. Also, for SIR threshold of 0dB, allocating 0.25 fraction of uplink transmit power can achieve approximately 6% improvement in coverage probability in the cell edge area compared to constant power policy and about 14% improvement compared to the full channel-inversion policy. Third and last, motivated by the powerful gains of incorporating small cells with macro cells, a massive MIMO aided heterogeneous cloud radio access network (H-CRAN) is investigated. More specific, based on Toeplitz matrix tool, tractable formulas for the link reliability and rate coverage of a typical user in H-CRAN are derived. Numerical outcomes confirm the powerful gain of the massive MIMO for enhancing the throughput of the H-CRAN while small remote radio heads (RRH cells) are capable of achieving higher energy efficiency.

Ke, Bingwen. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (p. 52-54). / Abstracts in Chinese and English. / Abstract --- p.3 / Acknowledgement --- p.5 / Content --- p.6 / Table of Figures --- p.8 / Chapter Chapter 1 --- Introduction --- p.9 / Chapter 1.1 --- Motivations and Contributions --- p.9 / Chapter 1.2 --- Organization of the Thesis --- p.11 / Chapter Chapter 2 --- Background --- p.12 / Chapter 2.1 --- Multiple-Input-Multiple-Output (MIMO) System --- p.12 / Chapter 2.1.1 --- Basic MIMO Structure --- p.12 / Chapter 2.1.2 --- Multiple User Detection (MUD) in MIMO Networks --- p.14 / Chapter 2.2 --- IEEE 802.11 --- p.16 / Chapter 2.2.1 --- CSMA/CA in 802.11 --- p.16 / Chapter 2.2.2 --- CSMA/CA(k) in 802.1 In --- p.18 / Chapter 2.2.3 --- Co-channel Transmission in MIMO WLAN --- p.19 / Chapter Chapter 3 --- Channel Correlation in MIMO Ad Hoc Networks --- p.20 / Chapter 3.1 --- Introduction of Channel Correlation --- p.20 / Chapter 3.2 --- Channel Correlation Threshold --- p.25 / Chapter Chapter 4 --- MAC with SINR Threshold --- p.28 / Chapter Chapter 5 --- Performance Evaluation of MWST in Fully-Connected Networks --- p.33 / Chapter Chapter 6 --- MAC with SINR Threshold (MWST) in Partially-Connected Networks --- p.38 / Chapter 6.1 --- Hidden Link Problem in Partially-Connected Networks --- p.38 / Chapter Chapter 7 --- Performance Evaluation in Partially-Connected Networks --- p.42 / Chapter 7.1 --- Fairness Issues in CSMA/CA(k) --- p.42 / Chapter 7.2 --- Fairness Performance of MWST --- p.45 / Conclusion --- p.50 / References --- p.52

Tese de doutoramento. Engenharia Electrotécnica e de Computadores. Faculdade de Engenharia. Universidade do Porto. 2009

Une meilleure compréhension des phénomènes de propagation de canal radio est la clé pour améliorer la performance globale des systèmes de communications sans-fil. Ceci est particulièrement vrai pour les environnements où sont observés de forts mécanismes de diffusion. Néanmoins, les modèles récents de canal radio n’incluent pas le diffus et doivent être réévalués en conséquence. Dans cette thèse, il est proposé de décomposer le canal radio polarimétrique MIMO en une composante multi-trajets spéculaire (SMC) et dense (DMC), cette dernière incluant le diffus et les faibles SMC. L’objectif de cette décomposition est de caractériser la contribution de la DMC et de développer un cadre de modélisation complet; cadre qui a été appliqué à deux scénarios de propagation présentant des mécanismes sévères de diffusion : milieu industriel et milieu végétal. Ici, des nouveaux modèles polarimétriques ont été développés et validés à partir de canaux radio mesurés. De plus, un algorithme de clustering basé sur la distance entre composante multi-trajets (MCD) a été proposé pour regrouper les SMC estimés. La performance et la robustesse de cet algorithme ont été comparées avec l’algorithme K-means MCD à partir de données générées par le modèle de canal WINNER II. L’algorithme validé a ensuite été directement appliqué aux scénarios avec l’hypothèse que la DMC est présente ou pas dans le modèle de données. Les résultats montrent sans ambiguïtés que les modèles proposés permettent non seulement une meilleure compréhension des mécanismes de propagation mais également que les modèles de canal radio sans DMC peuvent potentiellement induire en erreur l’interprétation de ces mécanismes. / A deeper understanding of the radio channel propagation phenomena is the key to improve the overall performance of wireless communication systems. This is particularly true for challenging propagation environments wherein strong diffuse scattering mechanisms are observed. However, the most recent radio channel models do not include this component and must be re-evaluated. In this thesis, it is proposed to decompose the polarimetric MIMO radio channel into specular and dense multipath components (SMC and DMC) where DMC includes diffuse scattering and weak SMC. The purpose of this decomposition is to investigate the contribution of DMC to the radio channel and develop a comprehensive modeling framework; framework which has been applied to two propagation scenarios presenting strong diffuse scattering mechanisms: indoor industrial and outdoor vegetation. Here, novel polarimetric models have been developed and validated from measured radio channels. Moreover, a multipath component distance (MCD)-based automatic clustering identification algorithm is proposed to group SMC obtained from measured radio channels. Its performance and robustness are compared with the K-means MCD algorithm using cluster data simulated by the WINNER II channel model. The validated clustering algorithm was then directly applied onto data which were estimated from the measured radio channels with or without DMC in the radio channel data model. The results unambiguously demonstrate that the proposed models not only provide a better understanding of the propagation mechanisms but also that radio channel models without DMC could potentially mislead the interpretation of those mechanisms.

Pour atteindre avec la 4G des débits théoriques supérieurs à 100 Mb/s et espérer pouvoir multiplier par 20 l’efficacité énergétique des futurs réseaux 5G, toute la richesse du canal de propagation doit être exploitée. Cet axe de recherche innovant d’optimisation de la couche physique de la communication se base sur la caractérisation multidimensionnelle « temps réel » du canal. Cette notion de "temps réel" signifie que toutes les dimensions spatiales (à l'émission et à la réception), temporelle, fréquentielle et polarimétrique seront explorées simultanément pour s'assurer des conditions de stationnarité du canal pendant la mesure. Le sondeur associé, objet de la thèse, doit également être capable de s’adapter à différents scénarii de propagation, y compris à un contexte de haute mobilité pour une liaison sol-sol entre trains à grande vitesse par exemple. Le système qui a ainsi été développé à l'aide de composants numériques programmables, fournit avec un temps de latence inférieur à quelques dizaines de µs et sans post traitement, 128 fonctions de transfert associées à un canal MIMO (8,16) dans 80 MHz de bande, la durée du signal transmis étant de 150 µs. Sa reconfigurabilité aisée lui confère l'originalité d'être multi fonctions pour s'adapter aux challenges à venir. Citons par exemple la cybersécurité des communications sans fil pour laquelle l'équipement aura les rôles de système de communication, de générateur d'interférences et de sondeur de canal.Dans le cadre de la localisation, via leur téléphone portable de personnes situées en forêt, les résultats des campagnes de mesures menées avec le sondeur soit au sol, soit embarqué dans un ULM, sont également décrits. / To be able to reach a bit rate higher than 100Mb/s with 4G systems and to multiply by 20 the energy efficiency of future 5G networks, all the propagation channel richness must be exploited. This innovative research area dealing with the physical layer optimization is based on the multidimensional channel characterization in "real time". This concept of “real time” means that space, time, frequency and polarimetric dimensions are explored simultaneously to ensure stationarity conditions of the channel during measurements. The channel sounder, subject of the thesis, must also be able to adapt to different scenarios of propagation, including a context of high mobility as, for example, in the case of a communication between high-speed trains. The system that has been developed, based on programmable digital components, allows measuring in a 80Mz bandwidth, 128 transfer functions associated with a (8,16) MIMO channel in less than a few tens of µs and without post-processing. The duration of the transmitted signal is 150 µs. Another originality of this sounder is its easy reconfigurability and its multi-function ability. For example, for studying cyber security of wireless communications, it would play the role of communication system, interference source and channel sounder. In the frame of localization of people in forest owing to their mobile phone, results of channel characterization conducted with the sounder placed either on the ground, or in an ULM, are also described and analyzed.

A wide range of modern large-scale systems relies on recovering a signal from noisy linear measurements. In many applications, the useful signal has inherent properties, such as sparsity, low-rankness, or boundedness, and making use of these properties and structures allow a more efficient recovery. Hence, a significant amount of work has been dedicated to developing and analyzing algorithms that can take advantage of the signal structure. Especially, since the advent of Compressed Sensing (CS) there has been significant progress towards this direction. Generally speaking, the signal structure can be harnessed by solving an appropriate regularized or constrained M-estimator. In modern Multi-input Multi-output (MIMO) communication systems, all transmitted signals are drawn from finite constellations and are thus bounded. Besides, most recent modulation schemes such as Generalized Space Shift Keying (GSSK) or Generalized Spatial Modulation (GSM) yield signals that are inherently sparse. In the recovery procedure, boundedness and sparsity can be promoted by using the ℓ1 norm regularization and by imposing an ℓ∞ norm constraint respectively. In this thesis, we propose novel optimization algorithms to recover certain classes of structured signals with emphasis on MIMO communication systems. The exact analysis permits a clear characterization of how well these systems perform. Also, it allows an automatic tuning of the parameters. In each context, we define the appropriate performance metrics and we analyze them exactly in the High Dimentional Regime (HDR). The framework we use for the analysis is based on Gaussian process inequalities; in particular, on a new strong and tight version of a classical comparison inequality (due to Gordon, 1988) in the presence of additional convexity assumptions. The new framework that emerged from this inequality is coined as Convex Gaussian Min-max Theorem (CGMT).