Abstract Integrating AI into wireless network design and management is essential for creating self-sustaining 6G networks. A key challenge is the development of automated network procedures with minimal human intervention, leveraging real-time monitoring data for immediate feedback. These advancements promote data-driven decision-making but pose risks related to data availability, safety, and the black-box nature of learning algorithms. This cumulative thesis proposes and evaluates novel procedures and algorithms for data- driven radio planning and cellular network optimization, addressing practical challenges in applying learning-based methods on real-world deployments. It emphasizes the utility of monitoring data and the integration of model-based and model-free methods, ensuring the scalability and safety of adaptive network procedures across diverse environments. The first part of the thesis explores the application of deep learning to radio propagation modeling in live cellular networks. The first paper presents a novel network ...

Technology advancements profoundly impact numerous aspects of life, including how we communicate and interact. For instance, hearing aids enable hearing-impaired or elderly people to participate comfortably in daily conversations; telecommunications equipment lifts distance constraints, enabling people to communicate remotely; smart machines are developed to interact with humans by understanding and responding to their instructions. These applications involve speech-based interaction not only between humans but also between humans and machines. However, the microphones mounted on these technical devices can capture both target speech and interfering sounds, posing challenges to the reliability of speech communication in noisy environments. For example, distorted speech signals may reduce communication fluency among participants during teleconferencing. Additionally, noise interference can negatively affect the speech recognition and understanding modules of a voice-controlled machine. This calls for speech enhancement algorithms to extract clean speech and suppress undesired interfering signals, ...

Audio effects modeling is the process of emulating an audio effect unit and seeks to recreate the sound, behaviour and main perceptual features of an analog reference device. Audio effect units are analog or digital signal processing systems that transform certain characteristics of the sound source. These transformations can be linear or nonlinear, time-invariant or time-varying and with short-term and long-term memory. Most typical audio effect transformations are based on dynamics, such as compression; tone such as distortion; frequency such as equalization; and time such as artificial reverberation or modulation based audio effects. The digital simulation of these audio processors is normally done by designing mathematical models of these systems. This is often difficult because it seeks to accurately model all components within the effect unit, which usually contains mechanical elements together with nonlinear and time-varying analog electronics. Most existing ...

Over the past years magnetic resonance spectroscopy (MRS) has been of significant importance both as a fundamental research technique in different fields, as well as a diagnostic tool in medical environments. With MRS, for example, spectroscopic information, such as the concentrations of chemical substances, can be determined non-invasively. To that end, the signals are first modeled by an appropriate model function and mathematical techniques are subsequently applied to determine the model parameters. In this thesis, signal processing algorithms are developed to quantify in-vivo and ex-vivo MRS signals. These are usually characterized by a poor signal-to-noise ratio, overlapping peaks, deviations from the model function and in some cases the presence of disturbing components (e.g. the residual water in proton spectra). The work presented in this thesis addresses a part of the total effort to provide accurate, efficient and automatic data analysis ...

This thesis deals with the coding of images with multiple coding schemes and their dynamic selection. In our society of information highways, electronic communication is taking everyday a bigger place in our lives. The number of transmitted images is also increasing everyday. Therefore, research on image compression is still an active area. However, the current trend is to add several functionalities to the compression scheme such as progressiveness for more comfortable browsing of web-sites or databases. Classical image coding schemes have a rigid structure. They usually process an image as a whole and treat the pixels as a simple signal with no particular characteristics. Second generation schemes use the concept of objects in an image, and introduce a model of the human visual system in the design of the coding scheme. Dynamic coding schemes, as their name tells us, make ...

Hyperspectral (HS) imaging, which consists of acquiring a same scene in several hundreds of contiguous spectral bands (a three dimensional data cube), has opened a new range of relevant applications, such as target detection [MS02], classification [C.-03] and spectral unmixing [BDPD+12]. However, while HS sensors provide abundant spectral information, their spatial resolution is generally more limited. Thus, fusing the HS image with other highly resolved images of the same scene, such as multispectral (MS) or panchromatic (PAN) images is an interesting problem. The problem of fusing a high spectral and low spatial resolution image with an auxiliary image of higher spatial but lower spectral resolution, also known as multi-resolution image fusion, has been explored for many years [AMV+11]. From an application point of view, this problem is also important as motivated by recent national programs, e.g., the Japanese next-generation space-borne ...

Given the widespread use of miniaturized audio interfaces, echo control systems are faced with increasing challenges to address a large variety of acoustic conditions observed by such interfaces. This motivates the use of sophisticated machine learning-based techniques to overcome the limitations of conventional methods. The contributions in this thesis can be outlined by decomposing the task of nonlinear acoustic echo control into two subtasks: Nonlinear Acoustic Echo Cancellation (NAEC) and Acoustic Echo Suppression (AES). In particular, by formulating the single-channel NAEC model-adaptation task as a Bayesian recursive filtering problem, an evolutionary resampling strategy for particle filtering is proposed. The resulting Elitist Resampling Particle Filter (ERPF) is shown experimentally to be an efficient and high-performing approach that can be extended to address challenging conditions such as non-stationary interferers. The fundamental problem of nonlinear model design is addressed by proposing a novel ...

The unprecedented processing demand, posed by the explosion of big data, challenges researchers to design efficient and adaptive machine learning algorithms that do not require persistent retraining and avoid learning redundant information. Inspired from learning techniques of intelligent biological agents, identifying transferable knowledge across learning problems has been a significant research focus to improve machine learning algorithms. In this thesis, we address the challenges of knowledge transfer through embedding spaces that capture and store hierarchical knowledge. In the first part of the thesis, we focus on the problem of cross-domain knowledge transfer. We first address zero-shot image classification, where the goal is to identify images from unseen classes using semantic descriptions of these classes. We train two coupled dictionaries which align visual and semantic domains via an intermediate embedding space. We then extend this idea by training deep networks that ...

In this thesis, we consider the problem of decentralized estimation under communication constraints in the context of Collaborative Signal and Information Processing. Motivated by sensor network applications, a high volume of data collected at distinct locations and possibly in diverse modalities together with the spatially distributed nature and the resource limitations of the underlying system are of concern. Designing processing schemes which match the constraints imposed by the system while providing a reasonable accuracy has been a major challenge in which we are particularly interested in the tradeoff between the estimation performance and the utilization of communications subject to energy and bandwidth constraints. One remarkable approach for decentralized inference in sensor networks is to exploit graphical models together with message passing algorithms. In this framework, after the so-called information graph of the problem is constructed, it is mapped onto the ...

Machine learning and artificial intelligence methods have achieved remarkable success, matching and even surpassing human capabilities in various complex tasks. However, many demonstrations have generally neglected a critical part of the intelligence that is prevalent in the real world, namely, the one that emerges from the collective of interconnected individuals with diverse capabilities, perspectives and experiences. To explore this fact, the current dissertation utilizes mathematical models of collaborative learning and reasoning. These models are based on the following two concepts: Bayesian inference, which is used to model how agents update their beliefs in the face of uncertain data, and graphs, which represent the communication links and information exchange among individuals. Through these models, the current work examines the effect of dynamic models on learning, as well as the implications of causal interactions among agents on their decisions. In particular, this ...

Global Navigation Satellite Systems (GNSS) is the de facto technology for Position, Navigation, and Timing (PNT) applications when it is available. GNSS relies on one or more satellite constellations that transmit ranging signals, which a receiver can use to self-localize. Signal acquisition is a crucial step in GNSS receivers, which is typically solved by maximizing the so-called Cross Ambiguity Function (CAF) resulting from a hypothesis testing problem. The CAF is a two-dimensional function that is related to the correlation between the received signal and a local code replica for every possible delay/Doppler pair, which is then maximized for signal detection and coarse synchronization. The outcome of this statistical process decides whether the signal from a particular satellite is present or absent in the received signal, as well as provides a rough estimate of its associated code delay and Doppler frequency, ...

This dissertation explores the intersection of data fusion, federated learning, and Bayesian methods, with a focus on their applications in indoor localization, GNSS, and image processing. Data fusion involves integrating data and knowledge from multiple sources. It becomes essential when data is only available in a distributed fashion or when different sensors are used to infer a quantity of interest. Data fusion typically includes raw data fusion, feature fusion, and decision fusion. In this thesis, we will concentrate on feature fusion. Distributed data fusion involves merging sensor data from different sources to estimate an unknown process. Bayesian framework is often used because it can provide an optimal and explainable feature by preserving the full distribution of the unknown given the data, called posterior, over the estimated process at each agent. This allows for easy and recursive merging of sensor data ...

It is increasingly more common that an occasion is recorded by multiple individuals with the proliferation of recording devices such as smart phones. When properly aligned, these recordings may provide several audio and visual perspectives to a scene which leads to several applications in restoring, remastering and remixing frameworks in various fields. In this study, we interpret the problem of aligning multiple unsynchronized audio sequences in a probabilistic framework. In this manner, we propose a novel, model based approach where we define a template generative model. We define 6 different generative models using this template covering basically all kinds of features (real valued, positive, binary and categorical). Proper scoring functions that evaluates the quality of an alignment are derived from each model where we are able to penalize non-overlapping alignments and alignment of a single sequence against a pre-aligned sequences. ...

Functional near infrared spectroscopy (fNIRS) needs a standardization in signal processing tools before it is recognized as a reliable neuroimaging modality. This thesis study tries to present a comprehensive analysis of the feasibility of applying statistical inference methods to fNIRS signals. Using hierarchical linear models, both classical and Bayesian techniques are pursued and performance of different methods are presented on a comparative basis. The results obtained from a set of cognitive signals show that fNIRS can identify cognitive activity both at the subject and group levels. The analysis suggests that mixed or Bayesian hierarchical models are especially convenient for fNIRS signals. A related problem that is discussed in this thesis study is to guarantee that the outcome of the statistical analysis is congruent with underlying physiology. This problem is studied by putting constraints over the parameters to be estimated. Carrying ...

In an automotive environment, quality of speech communication using a hands-free equipment is often deteriorated by interfering car noise. In order to preserve the speech signal without car noise, a multichannel speech enhancement system including a beamformer and a post-filter can be applied. Since employing a beamformer alone is insufficient to substantially reducing the level of car noise, a post-filter has to be applied to provide further noise reduction, especially at low frequencies. In this thesis, two novel post-filter designs along with their optimization for different driving conditions are presented. The first post-filter design utilizes an adaptive smoothing factor for the power spectral density estimation as well as a hybrid noise coherence function. The hybrid noise coherence function is a mixture of the diffuse and the measured noise coherence functions for a specific driving condition. The second post-filter design applies ...