On-shelf availability is a measure of how available a product is for a shopper in a store. The products should be in the place where the shopper expects it and at the time the shopper wants to buy it. To ensure a high Onshelf availability store clerks must move around the store and look for products that must be replenished and products that are misplaced. However, this task is rather time consuming since there are usually hundreds of different products to keep track of in a store. This thesis therefore, aims to simplify this task for store clerks by creating an automatic system that can identify “out of stocks” and misplaced products. Different state of the art object detection algorithms and image classification methods were evaluated in order to solve this task. The object detection algorithms were used to find products and gaps on store shelves and were trained with a relatively small dataset. The bounding boxes obtained from the object detection algorithm was then forwarded to an image classification algorithm in order to predict the label of the product. The training data of the classification dataset consisted of a small dataset with 29 different detergent products. All the algorithms were evaluated on speed and F1-score while the object detection algorithms were also evaluated on an average precision with an intersection of union as 0.5 and 0.75. The results indicate that it is possible to use deep learning in order to improve on-shelf availability. However, the methods might only perform good on the small datasets used in this thesis. Furthermore, since no real test have been made in a real-life supermarket it is impossible to say for certain if it would indeed improve on-shelf availability.
Speech and language relearning are challenging for stroke survivors as well as for medical caregiv-ers. After a stroke, a patient’s ability to read, write, speak, and listen is decreased to different degrees, which results in a compromised independent life and a decreased quality of life for the patient. Tech-nology enhances systems can play a vital role in this context. However, the available software are not specifically built for after the stroke patient’s needs. This paper is therefore aimed to gather require-ments for designing a tailor-made speech relearning software application for stroke survivors. A de-sign science approach was adopted, where different stakeholders such as medical caregivers and in-formation technology consultants were involved in the process. The well-informed and experienced participants in their fields highlighted some important requirements such as different types of inter-face for a patient than speech therapist with extra management functionality for speech therapists so that they can adjust the relearning exercises according to the patient’s needs. Software requirements vary from patient to patient where the intensity of speech and language impairments, general medical condition of the patient, age, prior experience, and knowledge about the information of the patient and social setup of the patient plays an important role. Since stroke is most common in adults and adults learn differently than children, adult learning theory might help understand the patients' needs. There-fore, adult learning principles were involved in the requirement analysis process. The established re-quirements will be used for the development of speech and language relearning software.
Stroke is a common and severe disease that can be found in all regions across the globe, and not onlyamong older adults. Result of a stroke can be death, or a variety of disabilities caused by impairments indifferent brain functions. This chapter discusses technology enhanced stroke rehabilitation from a threefoldview of cognitive, motoric and speech rehabilitation. The important research question was: Whatwould be the requirements for technology-enhanced stroke rehabilitation in the areas of cognitive, motoricand speech rehabilitation? The study was carried out with a requirement-focused Design Science approachcollecting data with semi-structured interviews. Informants were selected in a purposive sampling choosingprofessionals with valuable knowledge and skills in stroke rehabilitation. The findings in this study havegenerated useful general requirements for a future implementation and testing of technology enhancedstroke rehabilitation. Within each of the three rehabilitation categories cognitive, motoric and speech, thereseems to be potential for successful use of technology enhanced services. This development ofrehabilitation services must follow the fundamental principle for all forms of stroke rehabilitation: eachpatient needs a personalised treatment. However, in all three rehabilitation categories, there is a need todefine more specific requirements based on feedback from stroke patients testing the rehabilitationservices.
Point clouds have become increasingly prevalent in representing 3D scenes within virtual environments, alongside 3D meshes. Their ease of capture has facilitated a wide array of applications on mobile devices, from smartphones to autonomous vehicles. Notably, point cloud compression has reached an advanced stage and has been standardized. However, the availability of quality assessment datasets, which are essential for developing improved objective quality metrics, remains limited. In this paper, we introduce BASICS, a large-scale quality assessment dataset tailored for static point clouds. The BASICS dataset comprises 75 unique point clouds, each compressed with four different algorithms including a learning-based method, resulting in the evaluation of nearly 1500 point clouds by 3500 unique participants. Furthermore, we conduct a comprehensive analysis of the gathered data, benchmark existing point cloud quality assessment metrics and identify their limitations. By publicly releasing the BASICS dataset, we lay the foundation for addressing these limitations and fostering the development of more precise quality metrics.
Rapid development and massive use of Information Technology (IT) have since produced a massive amount of electronic data. In tandem, the demand for data outsourcing and the associated data security is increasing exponentially. Small organizations are often finding it expensive to save and process their huge amount of data, and keep the data secure from unauthorized access. Cloud computing is a suitable and affordable platform to provide services on user demand. The cloud platform is preferable used by individuals, Small, and Medium Enterprises (SMEs) that cannot afford large-scale hardware, software, and security maintenance cost. Storage and processing of big data in the cloud are becoming the key appealing features to SMEs and individuals. However, the processing of big data in the cloud is facing two issues such as security of stored data and system overload due to the volume of the data. These storage methods are plain text storage and encrypted text storage. Both methods have their strengths and limitations. The fundamental issue in plain text storage is the high risk of data security breaches; whereas, in encrypted text storage, the encryption of complete file data may cause system overload. This paper propose a feasible solution to address these issues with a new service model called Confidentiality-based Classification-as-a-Service (C2aaS) that performs data processing by treating data dynamically according to the data security level in preparation for data storing in the cloud. In comparison to the conventional methods, our proposed service model is strongly showing good security for confidential data and is proficient in reducing cloud system overloading.
The versatility and dimension of smart phone applications is increasing at magnificent rate and getting more and more advanced in a level that could solve complicated real time tasks. One of the important factors for such advancement has been the powerful sensors embedded on a Smartphone devices and sensory networks. Moreover, Context and Context-awareness would have remained a myth without the advent of sensors. The objective of this thesis has been to contribute to the research work carried out under the MediaSense project. Accordingly, the ultimate purpose of the thesis has been to evaluate and study the feasibility of the adaptive context view proposed in MediaSense Platform. In precise words, the thesis has done three core tasks. Firstly, the theoretical presentation of related works and the significance of the research question have been discussed through various social applications. Secondly, a proof-of-concept application has been developed to simulate what has been proposed in the research work. Finally, Android application has been designed and implemented in order to evaluate and study the techniques presented in a practical scenario. Moreover, in the android application known as SundsvallBIGBuddies, we have used the extensions designed for the existing MediaSense platform. The impact of using Android app relaying on a continuous stream of context data has been presented using graphs and tables. In order to study the impact we used smart phone and tablets from Samsung.
The thesis focuses on enabling the communication between Wireless Sensor Networks and Internet-of-Things applications. In order to achieve this goal, the first step has been to investigate the concept of the Internet-of-Things and then to understand how this scenario could be used to interconnect multiple Wireless Sensor Networks in order to develop context-aware applications which could handle sensor data coming from this type of network.
The second step was to design and implement a communication stack which enabled Wireless Sensor Networks to communicate with an Internet-of-Things platform. The CoAP protocol has been used as application protocol for the communication with the Wireless Sensor Networks. The solution has been developed in Java programming language and extended the sensor and actuator layer of the Sensible Things platform.
The third step of this thesis has been to investigate in which real world applications the developed solution could have been used. Next a Proof of Concept application has been implemented in order to simulate a simple fire detection system, where multiple Wireless Sensor Networks collaborate to send their temperature data to a control center. The last step was to evaluate the whole system, specifically the responsiveness and the overhead introduced by the developed communication stack.
This is a project that deals with network emulators for training purposes, where everything is based on open-source applications. The goal of this project was to evaluate GNS3 and CORE emulators and answer the question, if and how they can be used in educational purposes for students and teachers. The study begins by briefly describing the various emulators available through open-source, where it was chosen to focus on the following network emulators IMUNES: Marionnet, Mininet, NetKit, GNS3 and CORE. The evaluation was conducted using a form, and GNS3 and CORE emulators run on a Linux-based operating systems to test all functions and various applications available within the tools. The results showed that both emulators work great to make use of open-source applications that can emulate router functions to emulate network topologies with different routing protocols such as RIP, OSPF and BGP. The evaluation also showed that both emulators are excellent tools to be used by people with minimal knowledge in programming, because of its user-friendly interface that helps one to build complex topologies using drag-and-drop functionality only. The conclusion of the study is that both emulators work well for educational purposes to develop network technology, router protocols and Linux skills for students, as well as to create a virtual environment to develop their skills with which they also can experiment with their skills. To install the network emulators GNS3 and CORE and related applications took about 30 minutes per tool, as well as taking GNS3 23 MB and Core 10MB hard disk space to be installed without any accessory applications.
Why these two tools work well for training purposes is that both emulators has integrated support for a large numbers of applications and the use of simple user interface to emulate network environments. In addition, the tools are completely free to all students and teachers, therefore everyone have the same opportunities to access them. The report recommendation is to use the emulator CORE precisely because the utility has so many features integrated within itself, and it is such a simple tool to use.
The number of IoT devices and their respective data is increasing for each day impacting the traditional architecture model of solely using the cloud for processing and storage in a negative way. This model may therefore need a supporting model to alleviate the different challenges for future IoT applications. Several researchers have described and presented algorithms and models with focus on distributed architecture models. The main issues with these however is that they fall short when it comes to the implementation and distribution of tasks. The former issue is that they are not implemented on actual hardware but simulated in a constrained environment. The latter issue is that they are not considering sharing a single task but to distribute a whole task. The objective of this thesis is therefore to present the different challenges regarding the traditional architecture model, investigate the research gap for the IoT and the different computing paradigms. Together with this implementing and evaluating a future architecture model capable of collaboration for the completion of a generated task on multiple off-the-shelf hardware. This model is evaluated based on task completion time, data size, and scalability. The results show that the different testbeds are capable communicating and splitting a single task to be completed on multiple devices. They further show that the testbeds containing multiple devices are performing better regarding completion time and do not suffer from noticeable scalability issues. Lastly, they show that the completion time drops remarkably for tasks that are split and distributed.
Two different techniques are dominating the 3D TV market today i.e. active shutter glasses and passive film patterned retarder. Both the techniques have their pros and cons. In this paper we compare these two types of 3D TV3D TV by evaluating them with respect to some important visual ergonomic parameters such as angle dependent cross talk, luminance levels, flicker and resolution.
Embedded systems with integrated sensing, processing and wireless communication are driving future connectivity concepts such as Wireless Sensor Networks (WSNs) and Internet of Things (IoTs). Because of resource limitations, there still exists a number of challenges such as low latency and energy consumption to realize these concepts to full potential. To address and understand these challenges, we have developed and employed an intelligence partitioning method which generates different implementation alternatives by distributing processing load across multiple nodes. The task-to-node mapping has exponential complexity which is hard to compute for a large scale system. Regarding this, our method provides recommendation to handle and minimize such complexity for a large system. Experiments on a use-case concludes that the proposed method is able to identify unfavourable architecture solutions in which forward and backword communication paths exists in task-to-node mapping. These solution can be avoided for further architectural exploration, thus limiting the space for architecture exploration of a sensor node.
This thesis project concerns NAT traversal techniques and their application to P2P networking with regard to MediaSense platform. Since MediaSense open source platform, developed by Mid Sweden University, utilizes the benefits of P2P networking, it also suffers from the drawbacks provided by NAT. The issue of NAT traversal is not trivial due to the fact that the behavior of NAT devices is not standardized and vendors are free to provide their own implementations. The common knowledge is, that at least four main types of NATs exist, differing in the filtering and mapping algorithms employed. NAT traversal techniques vary accordingly. No single technique can handle all the cases. Most of the techniques can handle up to three main types of NAT. The last type is usually used in large corporate networks and is called the Symmetric NAT. The most viable, and basically the only available technique for its traversal, is data relaying. This thesis builds a NAT traversal module for the MediaSense platform. The main purpose of this module is to provide seamless NAT traversal capabilities to the platform. The module does this in several steps: UPnP enabled device discovery, NAT type determination and data relaying via the proxy. Firstly the module attempts to discover the presence of a UPnP enabled Internet Gateway Device on the network. If such a device is present on the network, a port mapping can be created, making the node located behind NAT accessible from the public Internet. If a UPnP enabled device was not found, the module will try to determine the type of NAT used. Based on the type of NAT used, the module can transit to either the proxy mode or request assistance of the STUN server to keep the created mapping alive. The resulting chapters provide the reader with the output produced by each step, conclusions the author has made while working on this project and some general ideas on future work within the subject.
Mobile devices and their applications are continuing to develop and the more advanced they are, the more they require high data ranges and the more they demand of the available wireless communication networks. At present, LTE (Long Term Evolution) is a good solution as it provides the users of mobile devices with a good throughput and a low latency. In the future, the two most important aspects for end users will be system spectral efficiency and system power controlling. This thesis deals with LTE downlink spectral efficiency and power controlling. The thesis will show how, by using IP multicasting for the LTE downlink, the base station is able to provide the necessary data through a significantly smaller spectrum and, additionally, how cooperative diversity, i.e. the cooperation between several base stations, can improve or even maximise the total network channel capacity, regardless of bandwidth size. A Packet and Resource Plan Scheduling algorithm (PARPS) is used to schedule the transmissions, and the results are calculated in MATLAB. By this means it is possible to analyse the efficiency of the spectrum management, the coverage probability and the power controlling for the different transmitters used for the LTE downlink.The LTE downlink scheme is simulated in Matlab for different numbers of transmitters (2-3). IP multicasting over the LTE downlink manages to transmit the same amount of data using less transmission power (50- 66.6%) with a better system spectral efficiency.
A big portion of software development is the method which controls the flow ofthe creative process. Thru experience it has been shown how a lack oforganisation can have a profoundly negative effect on the development workand finally on the product itself. We take the Scrum method in a two-man formand analyze how the a small-scale version effects the roles, artefacts andactivities associated with the original Scrum method. The result of the workusing a small scale Scrum is presented along with the changes that have beendone during the work. The result and the final discussion show a positive effecton using a structured development method.Lately the use of small screen devices and their use of the Internet has widelyincreased . With this change in user interface comes new challenges designing aWeb handling a wide variety of user devices. In the process of developing aweb application techniques for small-screen interfaces was analyzed from thewhere Responsive Design was found to be the best choice concidering relevantlimitations. Design principals from the concept of Responsive Design isanalyzed and applied.
The transmission of 3DTV sequences over packet based networks may result in degradations of the video quality due to packet loss. In the conventional 2D case, several different strategies are known for extrapolating the missing information and thus concealing the error. In 3D however, the residual error after concealment of one view might leads to binocular rivalry with the correctly received second view. In this paper, three simple alternatives are presented: frame freezing, a reduced playback speed, and displaying only a single view for both eyes, thus effectively switching to 2D presentation. In a subjective experiment the performance in terms of quality of experience of the three methods is evaluated for different packet loss scenarios. Error-free encoded videos at different bit rates have been included as anchor conditions. The subjective experiment method contains special precautions for measuring the Quality of Experience (QoE) for 3D content and also contains an indicator for visual discomfort. The results indicate that switching to 2D is currently the best choice but difficulties with visual discomfort should be expected even for this method.
Subjective assessment of Quality of Experience in stereoscopic 3D requires new guidelines for the environmental setup as existing standards such as ITU-R BT. 500 may no longer be appropriate. A first step is to perform cross-lab experiments in different viewing conditions on the same video sequences. Three international labs performed Absolute Category Rating studies on a freely available video database containing degradations that are mainly related to video quality degradations. Different conditions have been used in the labs: Passive polarized displays, active shutter displays, differences in viewing distance, the number of parallel viewers, and the voting device. Implicit variations were introduced due to the three different languages in Sweden, South Korea, and France. Although the obtained Mean Opinion Scores are comparable, slight differences occur in function of the video degradations and the viewing distance. An analysis on the statistical differences obtained between the MOS of the video sequences revealed that obtaining an equivalent number of differences may require more observers in some viewing conditions. It was also seen that the alignment of the meaning of the attributes used in Absolute Category Rating in different languages may be beneficial. Statistical analysis was performed showing influence of the viewing distance on votes and MOS results.
This report covers the implementation and evaluation of a stereo vision corre- spondence-based depth estimation algorithm on a GPU. The results and feed- back are used for a Multi-view camera system in combination with Jetson TK1 devices for parallelized image processing and the aim of this system is to esti- mate the depth of the scenery in front of it. The performance of the algorithm plays the key role. Alongside the implementation, the objective of this study is to investigate the advantages of parallel acceleration inter alia the differences to the execution on a CPU which are significant for all the function, the imposed overheads particular for a GPU application like memory transfer from the CPU to the GPU and vice versa as well as the challenges for real-time and concurrent execution. The study has been conducted with the aid of CUDA on three NVIDIA GPUs with different characteristics and with the aid of knowledge gained through extensive literature study about different depth estimation algo- rithms but also stereo vision and correspondence as well as CUDA in general. Using the full set of components of the algorithm and expecting (near) real-time execution is utopic in this setup and implementation, the slowing factors are in- ter alia the semi-global matching. Investigating alternatives shows that results for disparity maps of a certain accuracy are also achieved by local methods like the Hamming Distance alone and by a filter that refines the results. Further- more, it is demonstrated that the kernel launch configuration and the usage of GPU memory types like shared memory is crucial for GPU implementations and has an impact on the performance of the algorithm. Just concurrency proves to be a more complicated task, especially in the desired way of realization. For the future work and refinement of the algorithm it is therefore recommended to invest more time into further optimization possibilities in regards of shared memory and into integrating the algorithm into the actual pipeline.
As the market for low-power wide-area network (LPWAN) technologies expands and the number of connected devices increases, it is becoming important to investigate the performance of LPWAN candidate technologies in dense deployment scenarios. In dense deployments, where the networks usually exhibit the traits of an interference-limited system, a detailed intra- and inter-cell interference analysis of LPWANs is required. In this paper, we model and analyze the performance of uplink communication of a LoRa link in a multi-cell LoRa system. To such end, we use mathematical tools from stochastic geometry and geometric probability to model the spatial distribution of LoRa devices. The model captures the effects of the density of LoRa cells and the allocation of quasi-orthogonal spreading factors (SF) on the success probability of the LoRa transmissions. To account for practical deployment of LoRa gateways, we model the spatial distribution of the gateways with a Poisson point process (PPP) and Matèrn hard-core point process (MHC). Using our analytical formulation, we find the uplink performance in terms of success probability and potential throughput for each of the available SF in LoRa’s physical layer. Our results show that in dense multi-cell LoRa deployment with uplink traffic, the intercell interference noticeably degrades the system performance.
In todays society internet is used for communication between each other aroundthe world. The first video call was made around the year 1940 and it is time fora development, where 3D is something that can make video calls more real. Tomake this possible a system was constructed that would be able to get data fromdifferent time-of-flight cameras and color cameras and audio devices. That datashould later on be compressed and transmitted over internet to be able to play iton someone else’s 3D-display. To prevent the feeling of delay in the call, allparts together must happen in real time. The development methods that havebeen used is pair programming and a variation of test-driven development. Thesystem has been evaluated by time messurements, image quality and data sizeto find a good balance between time and quality. The system was constructedby five parts: capturing of images and audio, image upscaling, compression anddecompression, network streaming and also rendering. The result showed thatthe parts affected by data size and image quality could achieve a good balancebetween time and quality. However, all goals could not be achieved becausesome parts where too slow for the real time goal to be achieved and also someparts could not be constructed in time. Since the system was built up modularlythe parts that did not achieve the goals can be improved or replaced. Based onthe results, solution proposals was made to improve the results for a possiblefurther development.
This study investigates how to fine-tune the large language model GPTSW3 for a specific use case scenario. By using different training methods and techniques, the model has been adapted and evaluated based on its ability to generate correct answers. The research has identified the main challenges in the training process, which includes data quality, preprocessing and finding the optimal parameter settings. The study has also examined if the model’s ability to generate accurate answers is dependent on the size of the training data. The results showed that longer training periods, combined by supervised and unsupervised training, and optimization of the parameters are critical for improving the model’s ability to generate accurate answers. Future work should focus on increasing the datasets diversity and use a larger model to further improve the models ability to generate accurate answers.
In this paper we first analyse the possibility for deniability under a strong adversary, who has an Internet-wide transcript of the communication. Secondly, we present a scheme which provides the desirable properties of previous messaging schemes, but with stronger deniability under the new adversary model. Our scheme requires physical meetings for exchanges of large amounts of random key-material via near-field communication and later uses this random data to key a one-time pad for text-messaging. We prove the correctness of the protocol and, finally, we evaluate the practical feasibility of the suggested scheme.
The assessment of perceived quality based on psychophysiological methods recently gained attraction as it potentially overcomes certain flaws of psychophysical approaches. Although studies report promising results, it is not possible to arrive at decisive and comparable conclusions that recommend the use of one or another method for a specific application or research question. The video quality expert group started a project on psychophysiological quality assessment to study these novel approaches and to develop a test plan that enables more systematic research. This test plan comprises of a specifically designed set of quality annotated video sequences, suggestions for psychophysiological methods to be studied in quality assessment, and recommendations for the documentation and publications of test results. The test plan is presented in this article.
In this article, we have investigated a VR simulator of a forestry crane used for loading logs onto a truck. We have mainly studied the Quality of Experience (QoE) aspects that may be relevant for task completion, and whether there are any discomfort related symptoms experienced during the task execution. QoE experiments were designed to capture the general subjective experience of using the simulator, and to study task performance. The focus was to study the effects of latency on the subjective experience, with regards to delays in the crane control interface. Subjective studies were performed with controlled delays added to the display update and hand controller (joystick) signals. The added delays ranged from 0 to 30 ms for the display update, and from 0 to 800 ms for the hand controller. We found a strong effect on latency in the display update and a significant negative effect for 800 ms added delay on latency in the hand controller (in total approx. 880 ms latency including the system delay). The Simulator Sickness Questionnaire (SSQ) gave significantly higher scores after the experiment compared to before the experiment, but a majority of the participants reported experiencing only minor symptoms. Some test subjects ceased the test before finishing due to their symptoms, particularly due to the added latency in the display update.
The MPEG 3DV project is working on the next generation video encoding standard and in this process a call for proposal of encoding algorithms was issued. To evaluate these algorithm a large scale subjective test was performed involving Laboratories all over the world. For the participating Labs it was optional to administer a slightly modified Simulator Sickness Questionnaire (SSQ) from Kennedy et al (1993) before and after the test. Here we report the results from one Lab (Acreo) located in Sweden. The videos were shown on a 46 inch film pattern retarder 3D TV, where the viewers were using polarized passive eye-glasses to view the stereoscopic 3D video content. There were 68 viewers participating in this investigation in ages ranges from 16 to 72, with one third females. The questionnaire was filled in before and after the test, with a viewing time ranging between 30 min to about one and half hour, which is comparable to a feature length movie. The SSQ consists of 16 different symptoms that have been identified as important for indicating simulator sickness. When analyzing the individual symptoms it was found that Fatigue, Eye-strain, Difficulty Focusing and Difficulty Concentrating were significantly worse after than before. SSQ was also analyzed according to the model suggested by Kennedy et al (1993). All in all this investigation shows a statistically significant increase in symptoms after viewing 3D video especially related to visual or Oculomotor system.
This project is made for creating a stabil and useful mobile application developed in Java. The purpose of this project is to make the access and communication easier between the student and the student portal. The product of this project is the unofficial app of Mid Sweden University, called Mittuniversitetets Android-app, that in first hand is limited to Android-devices. By doing research, both by letting students answer a survey online on Mid Sweden University's student portal and by asking student physically at Mid Sweden University campus Sundsvall, the investigation tells which parts of the student portal students would like to have in the mobile application and it is according to those answers that the direction of the application has been developed. The product, which is the application, is considered complete when the students are able to for exempel reach their information and get the information presented in a good way. Three different suggestions for a solution is given and by comparing the benefits of those only one solution is chosen. The solution chosen is ”direct-connection as solution”. To be able to give the reader better basical and understandable knowledge some parts are explained in more detail in this report. The report also shows that the access to the student portal is more effective now that only 5 taps are required by using the Mid Sweden University application to get a students information instead of 24 taps by using the web-browser on the cellphone. All goals in this thesis is considered accomplished and screenshots shows this in the report. Finally there is also suggestions for some future works given in this report.
Interfacing the smart cities with cyber-physical systems (CPSs) improves cyber infrastructures while introducing security vulnerabilities that may lead to severe problems such as system failure, privacy violation, and/or issues related to data integrity if security and privacy are not addressed properly. In order for the CPSs of smart cities to be designed with proactive intelligence against such vulnerabilities, anomaly detection approaches need to be employed. This chapter will provide a brief overview of the security vulnerabilities in CPSs of smart cities. Following a thorough discussion on the applicability of conventional anomaly detection schemes in CPSs of smart cities, possible adoption of distributed anomaly detection systems by CPSs of smart cities will be discussed along with a comprehensive survey of the state of the art. The chapter will discuss challenges in tailoring appropriate anomaly detection schemes for CPSs of smart cities and provide insights into future directions for the researchers working in this field.
With the constant expansion of the industrial monitoring system, there is an urgent requirement to reduce investment and operating costs for the development of industrial communication technology. For industrial real-time monitoring systems, wireless technology can be used in a practical industrial production to take advantages of its flexibility and robustness. As wireless sensor networks have many advantages such as low investment costs, flexible structure and ease of transformation, it has become the focus with regards to industrial areas. THVRG is a routing algorithm that selects the routing path based on two-hop information. Since different information sensed by the sensors may have different requirements in order to reach the sink, a priority-based routing algorithm is required in order to adapt to this kind of situation. This thesis has proposed a priority routing algorithm based on the THVRG (Priority-based THVRG). In addition, a simulation of this algorithm was performed in OPNET. Finally, the report provides an evaluation of the proposed algorithm in industrial wireless sensor networks.
In modern applications, it is a big challenge that analyzing the order statistics about the most recent parts of the high-volume and high velocity stream data. There are some online quantile algorithms that can
keep the sketch of the data in the sliding window and they can answer the quantile or rank query in a very short time. But most of them take the
GK algorithm as the subroutine, which is not known to be mergeable. In this paper, we propose another algorithm to keep the sketch that maintains the order statistics over sliding windows. For the fixed-size window, the existing algorithms can’t maintain the correctness in the process of updating the sliding window. Our algorithm not only can maintain the correctness but also can achieve similar performance of the optimal algorithm. Under the basis of maintaining the correctness, the insert time and query time are close to the best results, while others can't maintain the correctness. In addition to the fixed-size window algorithm, we also provide the time-based window algorithm that the window size varies over time. Last but not least, we provide the window aggregation algorithm which can help extend our algorithm into the distributed system.
3D video conferencing is continuously evolving to make the visual experience realistic. The main advantage of 3D video conferencing deals with the addition of depth perception which enhances the user experience. The configuration of capturing and rendering equipments and the location of scene objects play an essential role in the quality of the user experience. An incorrect configuration of equipment parameters or an inconsistent distribution of scene components could cause an uncomfortable user experience, yielding in user sickness and dizziness.
The aim of this thesis is therefore to provide the tools and methods to assure a comfortable user experience when using Ericsson’s stereoscopic 3D video conferencing system. To achieve this goal, an investigation on the capturing and rendering systems has been performed to identify possible conflicts. This investigation has shown that accommodation-convergence rivalry, comfortable viewing range and stereo framing violation are the major sources of user discomfort in 3D video conferencing.
An algorithm for continuous analysis of produced stereoscopic content has been proposed. In particular, it detects wrong equipment configuration and problematic content in real time by means of automatic adjustments or user interaction, either at initiation phase or during the call session at capturing side. To validate and evaluate the efficiency of the implemented solution, a subjective test with participation of the 3D experts has been carried out. It has been shown that the proposed solution can detect targeted problems with high accuracy and apply corrective actions. It is important to note that manual solutions are not immune to problems. Nevertheless, it has also been shown that automatic solutions can considerably compensate manual methods inconsistency and provide a comfortable user experience.
In today’s society the need for more hardware efficient software since some people think that the doubling of computer power for the same price that Moore’s law predicted is no more. Reactive programming can be a
step in the right direction, this has led to an increase in interest in reactive programming. This object of this thesis is to evaluate the possibility of using reactive programming and R2DBC in Java to communicate with a relation database. This has been done by creating two Spring applications
one using the standards JDBC and servlet stack and one using R2DBC and the reactive stack. Then connecting them to a MySQL database and selecting and inserting values in to and from it and measuring the CPU usage, memory usage and execution time. In addition to this the possibilities to handle BLOBs in a good enough way were researched. The study shows that there are both advantages and disadvantages with using R2DBC it has basic support and it is based on good idea but at the time of this thesis it still needs more development before it can be used fully.
Recently, representations and methods analysing decision problems where probabilities and values (utilities) are associated with belief distributions over them (second order representations) have been suggested. In this paper we present an approach to how imprecise information can be modelled by means of second-order distributions and how a risk evaluation process can be elaborated by integrating procedures for numerically impreciseprobabilities and utilities. We discuss some shortcomings in the use of the principle of maximising the expectedutility and of utility theory in general, and offer remedies by the introduction of supplementary decision rules based on a concept of risk constraints taking advantage of second-order distributions.
The limited amount of good tools for supporting elicitation of preference information in multi-criteria decision analysis (MCDA) causes practical problem. In our experiences, this can be remedied by allowing more relaxed input statements from decision-makers, causing the elicitation process to be less cognitively demanding. Furthermore, it should not be too time consuming and must be able to actually use of the information the decision-maker is able to supply. In this paper, we propose a useful weight elicitation method for MAVT/MAUT decision making, which builds on the ideas of rank-order methods, but increases the precision by adding numerically imprecise cardinal information as well.
Face recognition is often described as the process of identifying and verifying people in a photograph by their face. Researchers have recently given this field increased attention, continuously improving the underlying models. The objective of this study is to implement a real-time face recognition system using one-shot learning. “One shot” means learning from one or few training samples. This paper evaluates different methods to solve this problem. Convolutional neural networks are known to require large datasets to reach an acceptable accuracy. This project proposes a method to solve this problem by reducing the number of training instances to one and still achieving an accuracy close to 100%, utilizing the concept of transfer learning.
The reason for performing this project work is to develop a Web application for the Student Union of Mid Sweden University applying the modern and comprehensive Microsoft .NET framework platform architecture. At present, the existing web application is divided into several modules which are built of server‐side scripting language technique and an open source database. The customer would like to develop the entire web applications using the Microsoft development tools and technologies in order to determine the possible benefit which could be obtained in terms of cost, maintenance, flexibility and the security perspective issues and also in terms of user friendly interactions options for all the involving partners in an effective way. The primary aim for the project is to start building a bookstore module for the Students Union that is responsible for selling literature to the students at the University. The module will also be integrated into a database system into which an administrator, a member of staff working in the Student Union, will be able to add a new book when it arrives and also update or delete if necessary later on. In addition to this module application all the book’s details belong to a certain category viewable to the students. The other part of this project work is aiming at finding a pattern similar to the bookstore module in which ordinary users can authenticate them towards a database and be able to add their curriculum vitae data entry and update it at a later stage as required.
Citizen quality of life can be improved through facilities and services that have been thought to ease citizen interaction with municipal authorities, offices and structures. All technologies and devices, used for developing these facilities, are the pillars of the Smart City idea: a City that adapts itself, at least in part, to citizens’ needs. Advanced Metering Infrastructure (AMI) could become the backbone of all the smart city projects. Other public services can be loaded on AMI’s to be smart and thus helping to find the affordability of investments. The paper deals with this topic by describing devices and results of a pilot project, which has been carried out in an Italian middle city (Salerno), to experience the use of RF 169MHz wM-bus based AMI. Experimental results regarding a set of about 2500 installed devices for gas and water metering, car parking management and elder tele-assistance, will be reported in detail to show convenience and problems of this approach.
Public transport organizations, such as Din Tur, require affordable modern solutions to improve their public image and passenger satisfaction. An easy way to create positive associations in people is to provide either useful services or, preferably, entertainment. This report covers the design, development and evaluation of an entertainment system – consisting of a smartphone game, a cloud-hosted backend, and a supporting on-bus hardware system – with the objective of making Din Tur's bus service seem more modern and appealing. The smartphone game, “Håll Platsen”, is developed in Unity game engine, focuses on providing brief entertainment during bus commutes, and incorporates gamification design elements. The Python-based back-end resides in Google's App Engine and Datastore platforms, and provides a unified virtual game environment enabling player cooperation and competition. The prototype on-bus hardware system uses the Raspberry Pi as a Light-Emitting Diode control system to supply real-world feedback of the game's virtual environment. The systems incorporate real-world busstop positioning, player location, online mapping services, team location-control mechanics, reflex-based minigames, player progression mechanics, and mobile-focused design. The resulting system can be useful in estimating public response to non-standard “smart” promotion methods, the use of games to improve everyday routines (i.e. commuting), and serve as a basis for further research in human & smart-technology interaction.
Wireless sensor networks have been a key driver of innovation and societal progressover the last three decades. They allow for simplicity because they eliminate ca-bling complexity while increasing the flexibility of extending or adjusting networksto changing demands. Wireless sensor networks are a powerful means of fillingthe technological gap for ever-larger industrial sites of growing interconnection andbroader integration. Nonetheless, the management of wireless networks is difficultin situations wherein communication requires application-specific, network-widequality of service guarantees. A minimum end-to-end reliability for packet arrivalclose to 100% in combination with latency bounds in the millisecond range must befulfilled in many mission-critical applications.The problem addressed in this thesis is the demand for algorithmic support forend-to-end quality of service guarantees in mission-critical wireless sensor networks.Wireless sensors have traditionally been used to collect non-critical periodic read-ings; however, the intriguing advantages of wireless technologies in terms of theirflexibility and cost effectiveness justify the exploration of their potential for controland mission-critical applications, subject to the requirements of ultra-reliable com-munication, in harsh and dynamically changing environments such as manufactur-ing factories, oil rigs, and power plants.This thesis provides three main contributions in the scope of wireless sensor net-works. First, it presents a scalable algorithm that guarantees end-to-end reliabilitythrough scheduling. Second, it presents a cross-layer optimization/configurationframework that can be customized to meet multiple end-to-end quality of servicecriteria simultaneously. Third, it proposes an extension of the framework used toenable service differentiation and priority handling. Adaptive, scalable, and fast al-gorithms are proposed. The cross-layer framework is based on a genetic algorithmthat assesses the quality of service of the network as a whole and integrates the phys-ical layer, medium access control layer, network layer, and transport layer.Algorithm performance and scalability are verified through numerous simula-tions on hundreds of convergecast topologies by comparing the proposed algorithmswith other recently proposed algorithms for ensuring reliable packet delivery. Theresults show that the proposed SchedEx scheduling algorithm is both significantlymore scalable and better performing than are the competing slot-based schedulingalgorithms. The integrated solving of routing and scheduling using a genetic al-vvigorithm further improves on the original results by more than 30% in terms of la-tency. The proposed framework provides live graphical feedback about potentialbottlenecks and may be used for analysis and debugging as well as the planning ofgreen-field networks.SchedEx is found to be an adaptive, scalable, and fast algorithm that is capa-ble of ensuring the end-to-end reliability of packet arrival throughout the network.SchedEx-GA successfully identifies network configurations, thus integrating the rout-ing and scheduling decisions for networks with diverse traffic priority levels. Fur-ther, directions for future research are presented, including the extension of simula-tions to experimental work and the consideration of alternative network topologies.
Computer experiments are part of the daily business formany researchers within the area of computational intelligence. However, there is no standard for either human orcomputer readable documentation of computer experiments.Such a standard could considerably improve the collaboration between experimental researchers, given it is intuitiveto use. In response to this deficiency the Intelligent Param-eter Utilization Tool ( InPUT ) is introduced. InPUT offers ageneral and programming language independent format forthe definition of parameters and their ranges. It providesservices to simplify the implementation of algorithms andcan be used as a substitute for input mechanisms of existing frameworks. InPUT reduces code-complexity and increases the reusability of algorithm designs as well as the reproducibility of experiments. InPUT is available as open-source for Java and this will soon also be extended to C++, two ofthe predominant languages of choice for the development ofevolutionary algorithms.
The provision of quality of service for Wireless Sensor Networks is more relevant than ever now where wireless solutions with their flexibility advantages are considered for the extension/substitution of wired networks for a multitude of industrial applications. Scheduling algorithms that give end-to-end guarantees for both reliability and latency exist, but according to recent investigations is the achieved quality of service insufficient for most control applications. Data aggregation is an effective tool to significantly improve on end-to-end contention and energy efficiency compared to single packet transmissions. In practice, though, it is not extensively used for process data processing on the MAC layer. In this paper, we outline the challenges for the use of data aggregation in Industrial Wireless Sensor Networks. We further extend SchedEx, a reliability-aware scheduling algorithm extension, for packet aggregation. Our simulations for scheduling algorithms from the literature show its great potential for industrial applications. Features for the inclusion of data aggregation into industrial standards such as WirelessHART are suggested, and remaining open issues for future work are presented and discussed.
We propose novel strategies for end-to-end reliability-aware scheduling in Industrial Wireless Sensor Networks (IWSNs). Becauseof stringent reliability requirements in industrial applications where missed packets may have disastrous or lethal consequences,all IWSN communication standards are based on Time Division Multiple Access (TDMA), allowing for deterministic channelaccess on the MAC layer. We therefore extend an existing generic and scalable reliability-aware scheduling approach by the name ofSchedEx. SchedEx has proven to quickly produce TDMA schedules that guarantee a user-defined end-to-end reliability level 𝜌 for allmultihop communication in a WSN. Moreover, SchedEx executes orders of magnitude faster than recent algorithms in the literaturewhile producing schedules with competitive latencies. We generalize the original problem formulation from single-channel tomultichannel scheduling and propose a scalable integration into the existing SchedEx approach. We further introduce a noveloptimal bound that produces TDMA schedules with latencies around 20% shorter than the original SchedEx algorithm. Combiningthe novel strategies with multiple sinks, multiple channels, and the introduced optimal bound, we could through simulations verifylatency improvements by almost an order of magnitude, reducing the TDMA superframe execution times from tens of seconds toseconds only, which allows for a utilization of SchedEx for many time-critical control applications.
In programming, concurrency allows threads to share processing units interleaving and seemingly simultaneous to improve resource utilization and performance. Previous research has found that concurrency faults are hard to avoid, hard to find, often leading to undesired and unpredictable behavior. Further, with the growing availability of multi-core devices and adaptation of concurrency features in high-level languages, concurrency faults occur reportedly often, which is why countermeasures must be investigated to limit harm. Reactive programming provides an abstraction to simplify complex concurrent and asynchronous tasks through reactive language extensions such as the RxJava and Project Reactor libraries for Java. Still, blocking violations are possibly resulting in concurrency faults with no Java compiler warnings. BlockHound is a tool that detects incorrect blocking by wrapping the original code and intercepting blocking calls to provide appropriate runtime errors. In this study, we seek an understanding of how common blocking violations are and whether a tool such as BlockHound can give us insight into the root-causes to highlight them as pitfalls to developers. The investigated Softwares are Java-based open-source projects using reactive frameworks selected based on high star ratings and large fork quantities that indicate high adoption. We activated BlockHound in the project’s test-suites and analyzed log files for common patterns to reveal blocking violations in 7/29 investigated open-source projects with 5024 stars and 1437 forks. A small number of system calls could be identified as root-causes. We here present countermeasures that successfully removed the uncertainty of blocking violations. The code’s intentional logic was retained in all validated projects through passing unit-tests.
One of the biggest obstacles for a broad deploymentof Wireless Sensor Networks for industrial applications is the dif-ficulty to ensure end-to-end reliability guarantees while providingas tight latency guarantees as possible. In response, we proposea novel centralized optimization framework for Wireless SensorNetworks that identifies TDMA schedules and routing combi-nations in an integrated manner. The framework is shown toguarantee end-to-end reliability for all events send in a schedulingframe while minimizing the delay of all packet transmissions. Itcan further be applied using alternative Quality of Service ob-jectives and constraints including energy efficiency and fairness.We consider network settings with multiple channels, multiplesinks, and stringent reliability constraints for data collectingflows. We compare the results to those achieved by the onlyscalable reliability-aware TDMA scheduling algorithm to ourknowledge, SchedEx, which conducts scheduling only. By makingrouting part of the problem and by introducing the conceptof source-aware routing, we achieve latency improvements forall topologies, with a notable average improvement of up to31percent.
Wireless Sensor Networks (WSN) are gaining popularity as a flexible and economical alternative to field-bus installations for monitoring and control applications. For missioncritical applications, communication networks must provide endto- end reliability guarantees, posing substantial challenges for WSN. Reliability can be improved by redundancy, and is often addressed on the MAC layer by re-submission of lost packets, usually applying slotted scheduling. Recently, researchers have proposed a strategy to optimally improve the reliability of a given schedule by repeating the most rewarding slots in a schedule incrementally until a deadline. This Incrementer can be used with most scheduling algorithms but has scalability issues which narrows its usability to offline calculations of schedules, for networks that are rather static. In this paper, we introduce SchedEx, a generic heuristic scheduling algorithm extension which guarantees a user-defined end-to-end reliability. SchedEx produces competitive schedules to the existing approach, and it does that consistently more than an order of magnitude faster. The harsher the end-to-end reliability demand of the network, the better SchedEx performs compared to the Incrementer. We further show that SchedEx has a more evenly distributed improvement impact on the scheduling algorithms, whereas the Incrementer favors schedules created by certain scheduling algorithms.
Wireless sensor networks (WSN) must ensure worst-case end-to-end delay and reliability guarantees for mission-critical applications.TDMA-based scheduling offers delay guarantees, thus it is used in industrial monitoring and automation. We propose to evolve pairs of TDMA schedule and routing-tree in a cross-layer in order to fulfill multiple conflicting QoS requirements,exemplified by latency and reliability.The genetic algorithm we utilize can be used as an analytical tool for both the feasibility and expected QoS in production. Near-optimal cross-layer solutions are found within seconds and can be directly enforced into the network.
Security concerns became high with the rapid technology advancement andwith the open nature of the internet. BankID is the leading electronic identificationsystem in Sweden which is used by around 5 million people in a variety ofpublic and private services. BankID allows users to securely authenticate themselvesand digitally sign important documents and transactions over the internet.In 2011, BankID Security App was launched to be used in mobile smartphones and tablet computers. In this paper, different components of the PublicKey Infrastructure (PKI) which is a cryptographic technique that enables usersto safely communicate over the insecure internet has been studied in detail. Furthermore,a test BankID-integrated PhoneGap based app on the Android platformis implemented and a performance evaluation and security analysis wereperformed. The test implementation of the BankID-integrated app on theAndroid platform provides user authentication and digital signing functions.The implemented backend system consists of a server with digital certificateand a database. The performance test emphasizes on the measurement of the accesstime between the components of the system and usability of the application.Access time measurement includes a reasonable amount of time in whichthe user is able to perform different activities in the system. In usability assessmentnumber of actions to perform a certain task and the ease of the user interfacehas been taken into consideration. The security analysis aims to identifypotential security flaws in the system and discuss possible solutions. The potentialsecurity risks we identified during the implementation of the system are theman-in-the-middle-attack, the Heartbleed bug, losing the mobile device andphysical access to the backend system. The potential security risks in the systemwere examined with regard to severity and probability of occurrence. Finally,the thesis project has been discussed in terms of the future work and system expansions.The result of the thesis will be used as a base in production developmentby Dewire, the company for which the thesis work has been conducted.
Recently, representations and methods aimed at analysing decision problems where probabilities and values (utilities) are associated with distributions over them (second-order representations) have been suggested. In this paper we present an approach to how imprecise information can be modelled by means of second-order distributions and how a risk evaluation process can be elaborated by integrating procedures for numerically imprecise probabilities and utilities. We discuss some shortcomings of the use of the principle of maximising the expected utility and of utility theory in general, and offer remedies by the introduction of supplementary decision rules based on a concept of risk constraints taking advantage of second-order distributions.
Videoconferencing is one of the most common telepresence methods today and educational videos is rising in popularity among distance learners. Traditional videoconferencing is unable to convey gestures and mutual eye contact between participants. This study aim to propose a Virtual Reality telepresence solution using game engines. A literature study confirmed the effectiveness achieved in VR is comparable to the effectiveness in face-to-face meetings. The suggested solution implements whiteboard functionality from a real-life perspective, confirming it is possible to include new functionality and directly transfer old functionality to the VR system from the communication systems today. The system was evaluated based on the response time, packet loss, bandwidth, frame rate and through user tests. The evaluation shows it is possible to design a telepresence system with VR capable of passing the Turing Test for Telepresence. The participants of the user tests did not experience discomfort and they were positively inclined to the telepresence system. Though, discomfort may emerge if the VR character is used with a common office workstation. Future studies in this topic would involve modifications of the third person camera, making the head's rotation follow the direction of the camera view and implementing movable eye pupils on the VR character using the upcoming eye-tracking accessory.
Java is one of the more recent programming languages that in runtime free applications from manual memory management by using automatic Garbage collector (GC) threads. Although, at the cost of stop-the-world pauses that pauses the whole application. Since the initial GC algorithms new collectors has been developed to improve the performance of Java applications. Still, memory related errors occurs and developers struggle to pick the correct GC for each specific case. Since the concept of microservices were established the benefits of using it over a monolith system has been brought to attention but there are still problems to solve, some associated to garbage collectors.
In this study the performance of garbage collectors are evaluated and compared in a microservice environment. The measurements were conducted in a Java SpringBoot application using Docker and a docker compose file to simulate a microservice environment. The application outputted log files that were parsed into reports which were used as a basis for the analysis. The tests were conducted both with and without a database connection. Final evaluations show that one GC does not fit all application environments. ZGC and Shenandoah GC was proven to perform very good regarding lowering latency, although not being able to handle the a microservice environment as good as CMS. ZGC were not able to handle the database connection tests at all while CMS performed unexpectedly well. Finally, the study enlightens the importance of balancing between memory and hardware usage when choosing what GC to use for each specific case.
This study explores the development of a robust Single Sample FaceRecognition (SSFR) system leveraging deep learning techniques. The focus is on overcoming the challenges associated with limited data, where only one image per person is available for training. Various convolutional neural network (CNN) models, including VGGFace,VGG-16 ResNet-50, and MobileNet, have been evaluated for their accuracy and scalability. Advanced preprocessing techniques, such as multi-scale, multicrop,
flipping, and Principal Component Analysis (PCA), have been used to increase model performance. The final model, using VGGFace for feature extraction and K-Nearest Neighbors (KNN) for classification, achieved an accuracy of 88.13% on the Labeled Faces in the Wild (LFW) dataset.
Scalability tests indicated that the model’s training time increased linearly with the number of training samples, demonstrating practical applicability for large-scale implementations. The study shows the importance
of applying suitable preprocessing and modular approaches in developing scalable and accurate SSFR systems.
Recent advancements in neuromorphic computing hardware have primarily centered around the development of integrated circuits, along with the analysis and simulation of analog circuits using the SPICE program. However, a critical shortfall of SPICE is its incapacity to accommodate algorithms and circuits based on microcontrollers, posing challenges for implementing on-chip training necessary for in-memory computing. To bridge this gap, our paper explores the implementation of SPICE-compatible circuit designs in the Proteus circuit simulator. We integrate a memristor component and a neuron model featuring a sigmoid activation function into Proteus. Subsequently, we construct a neuromorphic accelerator comprising 30 memristors and 4 neurons to demonstrate on-chip training and inference processes using an Arduino microcontroller. Additionally, our study introduces an optimal learning algorithm tailored for training memristors and adjusting synaptic weights during the training phase. The designed peripheral circuits govern all memristors throughout both training and inferencing and formulate an algorithm to apply test data for evaluating network accuracy.