Current SDN controllers have been designed based on a monolithic approach that integrates all of services and applications into one single, huge program. The monolithic design of SDN controllers restricts programmers who build management applications to specific programming interfaces and services that a given SDN controller provides, making application development dependent on the controller, and thereby restricting portability of management applications across controllers. Furthermore, the monolithic approach means an SDN controller must be recompiled whenever a change is made, and does not provide an easy way to add new functionality or scale to handle large networks. To overcome the weaknesses inherent in the monolithic approach, the next generation of SDN controllers must use a distributed, microservice architecture that disaggregates the control plane by dividing the monolithic controller into a set of cooperative microservices. Disaggregation allows a programmer to choose a programming language that is appropriate for each microservice. In this dissertation, we describe steps taken towards disaggregating the SDN control plane, consider potential ways to achieve the goal, and discuss the advantages and disadvantages of each. We propose a distributed architecture that disaggregates controller software into a small controller core and a set of cooperative microservices. In addition, we present a software defined network programming framework called Umbrella that provides a set of abstractions that programmers can use for writing of SDN management applications independent of NB APIs that SDN controllers provide. Finally, we present an intent-based network programming framework called OSDF to provide a high-level policy based API for programming of network devices using SDN. <br>

Standing up from a seated position, known as sit-to-stand (STS) movement, is one of the most frequently performed activities of daily living (ADLs). However, the aging generation are often encountered with STS issues owning to their declined motor functions and sensory capacity for postural control. The motivated is rooted from the contemporary market available STS assistive devices that are lack of genuine interaction with elderly users. Prior to the software implementation, the robot chair platform with integrated sensing footmat is developed with STS biomechanical concerns for the elderly. The work has its main emphasis on recognising the personalised behavioural patterns from the elderly users’ STS movements, namely the STS intentions and personalised STS feature prediction. The former is known as intention recognition while the latter is defined as assistance prediction, both achieved by innovative machine learning techniques. The proposed intention recognition performs well in multiple subjects scenarios with different postures involved thanks to its competence of handling these uncertainties. To the provision of providing the assistance needed by the elderly user, a time series prediction model is presented, aiming to configure the personalised ground reaction force (GRF) curve over time which suggests successful movement. This enables the computation of deficits between the predicted oncoming GRF curve and the personalised one. A multiple steps ahead prediction into the future is also implemented so that the completion time of actuation in reality is taken into account.

The differences between legacy and 5G networks in both capability, and their increasingly dynamic nature and complexity warrant searching for new kinds of ways to manage networks. This thesis work explored different options for interaction and interfaces for a declarative, intent-based control (IBC) based 5G network management system. IBC has not been previously applied on the transport layer and in addition to drafting interface concepts, this work also maps existing definitions and implementations of IBC and intents. The functioning of IBC on the transport layer was envisioned through these existing solutions. The work was done as an iterative codesign project following a user-centered design process. Concepts were both drafted and evaluated together with a group of domain experts. Special attention was paid to the forming of trust towards automation. A tentative task path for issuing an intent to the system was mapped and the drafted concepts approached the steps involved in the task path with varying levels of automation and different visual representations for the information the user needs and the functions of the system. No prior body research exists on IBC on the transport layer and trust towards automation in network management and the topics are novel. The envisioned application of IBC on the transport layer, and the drafted concepts on user interaction with the system are tentative in nature and more research is required to determine the feasibility of applying IBC on the transport layer, as well as the effectiveness of the presented concepts in promoting trust among users. / Skillnaderna mellan äldre och 5G-nätverk i båda funktionerna, och deras alltmer dynamiska karaktär och komplexitet motiverar sökning efter nya typer av sätt att hantera nätverk. Detta avhandlingsarbete undersöktes olika alternativ för interaktion och gränssnitt för en deklarativ, avsiktsbaserad kontroll (intent-based control, IBC) 5G-nätverkshanteringssystem. IBC har inte tidigare applicerats på transportskiktet och förutom utarbetande av gränssnittskoncept, kartlägger detta arbete också befintliga definitioner och implementeringar av IBC och avsikter. Funktion av IBC på transportskiktet förutsågs genom dessa befintliga lösningar. Arbetet gjordes som ett iterativt samdesignprojekt efter den användarcentrerad designprocess. Begrepp utarbetades och utvärderades tillsammans med en grupp domenexperter. Särskild uppmärksamhet ägnas åt bildandet av förtroende för automatisering. En preliminär arbetsväg för att utfärda en avsikt till systemet kartlades och de utarbetade koncepten närmade sig stegen i uppgiftsvägen med varierande nivåer av automatisering och olika visuella representationer för informationen användarens behov och systemets funktioner. Ingen tidigare undersökning finns på IBC på transportskit och förtroendet mot automatisering i nätverkshantering och ämnena är nya. De förutsåg tillämpning av IBC på transportlagret och utkastet begrepp om användarinteraktion med systemet är preliminära och mer forskning krävs för att bestämma genomförbarheten av att tillämpa IBC på transportskiktet, liksom effektiviteten på det presenterade begrepp för att främja förtroende bland användare. / 5G verkkojen eroavaisuudet aikaisempiin verkkoihin niin dynaamisuuden, kuin uusien kyvykkyyksien ja verkkojen monimutkaisuus vaativat uudenlaisia tapoja hallita verkkoja. Työssä konseptoitiin eri vuorovaikutus- ja käyttöliittymämalleja aiepohjaiselle- (intent-based control, IBC), deklaratiiviselle hallintamallille 5G-verkkojen hallintaan. IBC:ia ei ole aikaisemmin sovellettu kuljetusverkkotasolla ja käyttöliittymämallien lisäksi työssä kartoitettiin myös IBC:in ja aikeen (intent) olemassa olevia määritelmiä ja toteutustapoja sovellustasolla, joiden pohjalta hahmoteltiin, miten IBC toimisi kuljetusverkkotasolla. Työ toteutettiin osallistavana, iteratiivisena muotoiluprojektina seuraten käyttäjäkeskeistä suunnitteluprosessia. Konseptit luotiin, sekä arvioitiin pienessä ryhmässä alan asiantuntijoiden kanssa. Luottamuksen muodostamiseen automaatiota kohtaan kiinnitettiin erityistä huomiota. Työstetyissä konsepteissa käyttäjän toimintopolun eri askeleita aikeen antamiseksi järjestelmälle lähestyttiin eri kanteilta vaihdellen toimintopolun automatisaation tasoa, sekä verkon ja aikeen tilan ja toimintojen käyttäjälle näkyvää visuaalista esitystä. IBC:n soveltaminen kuljetusverkkotasolla ja luottamus automaatiota kohtaan verkonhallinnassa ovat aiheina uusia, eikä aiheista löydy aikaisempaa tutkimusmateriaalia. Työn hahmotelma IBC:ista kuljetusverkkotasolla, sekä työn aikana tuotetut konseptit käyttäjän vuorovaikutuksesta järjestelmän kanssa ovat luonteeltaan alustavia ja lisää tutkimustyötä vaaditaan mm. IBC:n käytön realistisuudesta kuljetusverkkotasolla, sekä hahmoteltujen konseptien toimivuudesta käyttäjän luottamuksen vahvistamisessa.

Managing a system of network services is a complex and large-scale task that often lacks a trivial optimal solution. Deep Reinforcement Learning (RL) has shown great potential in being able to solve these tasks in static settings. However, in practice, the RL agents struggle to generalize their control policies enough to work in more dynamic real-world environments. To achieve a generality between environments, multiple contributions are made by this thesis. Low-level metrics are collected from each node in the system to help explain changes in the end-to-end delay of the system. To achieve generality in its control policy, more ways to observe and understand the dynamic environment and how it changes are provided to the RL agent by introducing the end-to-end delay of each service in the system to its observation space. Another approach to achieving more generality in RL policies is Model-Agnostic Meta-Learning (MAML), a type of Meta-Learning approach where instead of learning to solve a specific task, the model learns to learn how to quickly solve a new task based on prior knowledge. Results show that low-level metrics yield a much greater generality when helping to explain the delay of a system. Applying MAML to the problem is beneficial in adding generality to a learned RL policy and making the adaptation to a new task faster. If the RL agent can observe the changes to the underlying dynamics of the environment between tasks by itself, the policy can achieve this generality by itself without the need for a more complex method. However, if acquiring or observing the necessary data is too expensive or complex, switching to a Meta-Learning approach might be beneficial to increase generality. / Hanteringen av ett system med nätverksstjänster är ett komplext och stor skaligt problem där den optimal lösning inte är trivial. Djup förstärkningsinlärning har visat stor potential i att kunna lösa dessa problem i statiska miljöer. Dock är det svårt att generalisera lösningarna tillräckligt för att fungera i mer komplicerade och realistiska dynamiska miljöer. För att uppnå mer generella lösningar mellan miljöer presenterar denna masteruppsats flera möjliga lösningar. Lågnivåmetrik samlas in från varje nod i systemet för att hjälpa förklara skillnader i den totala responstiden för varje tjänst i systemet. För att generalisera förstärkningsinlärningsmodellen kan den förses med fler sätt att observera miljön, och därmed lära sig förstå hur den dynamiska miljön förändras. En annan metod för att uppnå mer generalitet inom förstärkningsinlärning är Model-Agnostic Meta-Learning (MAML), en typ av Meta-Learning där istället för att lära sig att lösa en specifik uppgift, modellen lär sig att lära sig att snabbt lösa en ny uppgift baserat på sin tidigare kunskap. Resultaten visar att lågnivåmetriken leder till en mycket högre generalitet i att hjälpa till att förklara responstiden av ett system. Att applicera MAML till problemet hjälper att bidra med generalitet till en förstärkningsinlärningsmodell och gör anpassningen till en ny uppgift snabbare. Om modellen själv kan observera ändringarna i underliggande dynamiken bakom miljön mellan uppgifter kan den uppnå mer generalitet utan ett behov av en mer komplex metod som MAML. Däremot, om det är svårt eller dyrt att få tag på eller observera den nödvändiga datan, kan ett byte till en Meta-Learning baserad metod vara mer fördelaktig för att öka generaliteten.

Chat-bots are computer programs designed to conduct textual or audible conversations with a single user. The job of a chat-bot is to be able to find the best response for any request the user issues. The best response is considered to answer the question and contain relevant information while following grammatical and lexical rules. Modern chat-bots often have trouble accomplishing all these tasks. State-of-the-art approaches, such as deep learning, and large datasets help chat-bots tackle this problem better. While there is a number of different approaches that can be applied for different kind of bots, datasets of suitable size are not always available. In this work, we introduce and evaluate a method of expanding the size of datasets. This will allow chat-bots, in combination with a good learning algorithm, to achieve higher precision while handling their tasks. The expansion method uses the continual learning approach that allows the bot to expand its own dataset while holding conversations with its users. In this work we test continual learning with IBM Watson Assistant chat-bot as well as a custom case study chat-bot implementation. We conduct the testing using a smaller and a larger datasets to find out if continual learning stays effective as the dataset size increases. The results show that the more conversations the chat-bot holds, the better it gets at guessing the intent of the user. They also show that continual learning works well for larger and smaller datasets, but the effect depends on the specifics of the chat-bot implementation. While continual learning makes good results better, it also turns bad results into worse ones, thus the chat-bot should be manually calibrated should the precision of the original results, measured before the expansion, decrease.