Aaltodoc

Aaltodoc is the institutional repository of Aalto University.

Aaltodoc has a new updated appearance. Instructions for searching and more information is available here.

Otakaari 1 grandhall. Photo: Esa Kapila
 

Communities in DSpace

Select a community to browse its collections.

Now showing 1 - 13 of 13

Recent Submissions

Neural Network Audio Preset Tagger
(2025-02-21) Puustinen, Mikko
 School of Electrical Engineering | Master's thesis
In recent years, audio presets, which are a set of predefined parameter values for signal processing algorithms, have become a popular way of saving and sharing certain desirable sound characteristics of these signal processing algorithms. The naming of these presets via natural language, conveys important information regarding the expected tonal characteristics and potential emotional aspects induced to the creator of the preset. This thesis aims to investigate and model the relationship between two modalities in audio preset creation, audio and text. To this aim, a multi-modal transformer is developed whose objective is to predict the name of a preset given an audio signal. Additionally, three data augmentation strategies are proposed to alleviate the problem of limited amount of training data. The performance of the model is assessed both by objective and subjective evaluation. Results from the objective assessment indicate that the model is capable of learning a connection between audio and text, although room for improvement is left in terms of generalization. Subjective evaluation highlights two key trends: the generated preset names are generally satisfactory according to human judgment and room for improvement is left in the grammatical natural language understanding of the model.
Statistical modelling of genetic background of haemoglobin deferral in blood donation
(2025-02-24) Karttunen, Krista
 School of Science | Master's thesis
Blood donation and blood supply are essential part of modern health care and part of national preparedness and readiness. In Finland, the Finnish Red Cross Blood Service, a non-profit organization, is centrally responsible for blood donation and the production of blood products. The health of the donor and the patient receiving the blood products is one of the most important principles. Each blood donation results in the donor losing about 250 mg of iron, which can lead to iron deficiency. To maintain the donor's health and the quality of blood products, a health questionnaire and haemoglobin measurement are conducted before each donation. If the donor's haemoglobin is lower than the threshold values, the donor is not allowed to donate. A low haemoglobin level results in a 90-day deferral, or in some cases, the deferral remains until the cause of the low haemoglobin is determined. Even a short-term deferral is known to affect the donor's motivation to donate in the future. If individuals known to be prone to low haemoglobin donated less frequently than the current recommendations, the number of haemoglobin deferrals could be decreased. Factors affecting haemoglobin deferral have been studied, but there is still little information on the impact of single nucleotide polymorphisms on haemoglobin deferrals. The aim of this thesis was to model the effect of single nucleotide variants associated with iron deficiency anaemia and iron metabolism disorders on haemoglobin deferral, alongside other commonly used variables, and to statistically compare the performance of these models. This study used data from the Finnish Red Cross Blood Service Biobank, including genotyped donors and their donation history. More complex models did not perform better in prediction of Hb deferral than simpler models. The SNP 17:58358769 variant was positively associated with haemoglobin deferral in all models used. Interestingly, the Cox proportional hazards model performed worse than other models in the prediction task, but detected a difference in the risk of haemoglobin deferral between pre- and postmenopausal women for two variants (SNP 1:169549811 and SNP 22:37066896), which other models did not detect. Variables derived from donation history predict Hb deferral well, but genetic variables can provide additional information for the prediction models.
Implementation and benchmarking of channel estimation for instantaneous beamforming
(2025-02-24) Ostrówka, Łukasz
 School of Electrical Engineering | Master's thesis
The evergrowing demand for the increased performance of mobile networks requires constantly introducing new technologies. In fifth-generation networks, massive multiple input multiple output (mMIMO) antenna arrays are essential for network performance in dense networks. However, their massive adoption is delayed by inadequate specification. The presence of multiple antennas vastly increases the throughput needed for the transfer of data toward the Distributed Unit (DU). To address this challenge the Open RAN Alliance has been working on the revised architecture, in which beam selection is performed in the Radio Unit (RU) and the transferred data dimension can be therefore reduced only to selected beams. This master's thesis utilizes an application-specific instruction set processor (ASIP) platform to implement and benchmark channel estimation for instantaneous beamforming. The presented solution aims to operate without existing information on channel conditions coming from the DU. The presented solution is implemented for 7 different combinations of antenna array size and layer number configurations. The calculated channel estimation error of the proposed solution provides satisfactory results, with the error reaching 2.55%. The performance of each configuration combination is measured and compared. The thesis also shows the impact of vector optimization in implementation, with an average performance gain of 57 times in comparison to scalar processing.
Techno-Economic analysis of Customer Relationship Management system development in a logistics company
(2025-02-19) Niemi, Julia
 School of Electrical Engineering | Master's thesis
Business environments are rapidly changing and evolving due to technological advancements, economic fluctuations, and increased client expectations. As a result, technological adaptability and efficient Customer Relationship Management (CRM) are fundamental for gaining a competitive edge in a dynamic market. A well-integrated CRM system significantly enhances company operations by improving customer relationships, increasing profitability, and streamlining processes. The advantages are particularly important in the logistics industry, in which effective processes and client value-creation are crucial factors for company perseverance. This thesis conducts a techno-economic analysis of CRM system development within a logistics company and aims to optimise the CRM processes of a specific unit within the company. The study focuses on addressing the limitations of the current CRM processes, including inefficiencies, scalability challenges, and data fragmentation. Based on the analysis, the objective is to define technical and economic criteria and utilise them to compare new CRM solutions with each other. The ultimate purpose is to propose a sustainable solution that aligns with the strategic goals of the company. The study compares two cloud-based CRM systems with each other, analysing their integration potential, scalability, cost-effectiveness, and alignment with strategic objectives. The findings highlight the strengths and limitations of each system. System 1 thrived in technical capabilities, providing superior scalability, integration potential, and advanced technologies such as predictive AI analytics, making it ideal for long-term growth. System 2 provided faster implementation, lower upfront costs, and localised support, providing a more immediate solution. The study recommends System 1 to the company due to stronger alignment with long-term scalability objectives and better support for seamless integrations. By implementing System 1, the company can streamline workflows, enhance customer experience, and remain competitive. This thesis contributes to CRM research by providing a comprehensive evaluation framework and practical insights for CRM implementation in the logistics industry.
Product Development in the Defence Industry: Transitioning from Project-Driven to Product-Driven Development
(2025-02-24) Sundell, Melina
 School of Electrical Engineering | Master's thesis
This thesis introduces and examines a new product development process in the defence industry. Traditionally, defence companies have developed products based on customer orders, with predefined requirements and funding from the customer. This type of product development follows a project-driven approach. However, long development cycles and rising costs are driving the need for a more agile method to keep pace with technological advancements and market demands. To address this, the thesis explores the adoption of product-driven approach, which leverages internal innovation and off-the-shelf products to accelerate development and provide greater control over the product portfolio. Methodology included a literature review and a case study with interviews and document analysis. This thesis was conducted in collaboration with Saab Finland Oy, and Saab Sensor Technology Centre in Tampere was the primary case-study to examine the implementation of product-driven development in a defence company. The findings indicate that transitioning to a product-driven approach can offer significant benefits, including faster innovation, improved cost efficiency, and greater customer value. The case study of Sirius Compact demonstrated how standardized hardware platforms, continuous software updates, and value-based pricing can enhance long-term product viability and profitability. Key success factors identified include fail-fast strategies for early investment in innovation, platform-based product families to leverage economies of scale, and a continuous feedback loop between customers, sales, and R&D. The study concludes that to remain competitive in a rapidly evolving landscape, defence companies must proactively shape product strategies rather than rely solely on traditional customer-driven project models.