Search results for "A2"
showing 10 items of 1101 documents
Convolutional neural networks in skin cancer detection using spatial and spectral domain
2019
Skin cancers are world wide deathly health problem, where significant life and cost savings could be achieved if detection of cancer can be done in early phase. Hypespectral imaging is prominent tool for non-invasive screening. In this study we compare how use of both spectral and spatial domain increase classification performance of convolutional neural networks. We compare five different neural network architectures for real patient data. Our models gain same or slightly better positive predictive value as clinicians. Towards more general and reliable model more data is needed and collection of training data should be systematic. peerReviewed
Cognitive self-healing system for future mobile networks
2015
This paper introduces a framework and implementation of a cognitive self-healing system for fault detection and compensation in future mobile networks. Performance monitoring for failure identification is based on anomaly analysis, which is a combination of the nearest neighbor anomaly scoring and statistical profiling. Case-based reasoning algorithm is used for cognitive self-healing of the detected faulty cells. Validation environment is Long Term Evolution (LTE) mobile system simulated with Network Simulator 3 (ns-3) [1, 2]. Results demonstrate that cognitive approach is efficient for compensation of cell outages and is capable to improve network coverage. Anomaly analysis can be used fo…
Context-aware data caching for 5G heterogeneous small cells networks
2016
In this work, we investigate the problem of context-aware data caching in the heterogeneous small cell networks (HSCNs) to provide satisfactory to the end-users in reducing the service latency. In particular, we explore the storage capability of base stations (BSs) in HSCNs and propose a data caching model consists of edge caching elements (CAEs), small cell base stations (SBSs), and macro cell BS (MBS). Then, we concentrate on how to efficiently match the data contents to the different cache entities in order to minimize the overall system service latency. We model it as a distributed college admission (CA) stable matching problem and tackle this issue by utilizing contextual information t…
UAV-based hyperspectral monitoring of small freshwater area
2014
Recent development in compact, lightweight hyperspectral imagers have enabled UAV-based remote sensing with reasonable costs. We used small hyperspectral imager based on Fabry-Perot interferometer for monitoring small freshwater area in southern Finland. In this study we shortly describe the utilized technology and the field studies performed. We explain processing pipeline for gathered spectral data and introduce target detection-based algorithm for estimating levels of algae, aquatic chlorophyll and turbidity in freshwater. Certain challenges we faced are pointed out.
Scalable implementation of dependence clustering in Apache Spark
2017
This article proposes a scalable version of the Dependence Clustering algorithm which belongs to the class of spectral clustering methods. The method is implemented in Apache Spark using GraphX API primitives. Moreover, a fast approximate diffusion procedure that enables algorithms of spectral clustering type in Spark environment is introduced. In addition, the proposed algorithm is benchmarked against Spectral clustering. Results of applying the method to real-life data allow concluding that the implementation scales well, yet demonstrating good performance for densely connected graphs. peerReviewed
The influence of dataset size on the performance of cell outage detection approach in LTE-A networks
2015
The configuration and maintenance of constantly evolving mobile cellular networks are getting more and more complex and hence expensive. Self-Organizing Networks (SON) concept is an umbrella term for the set of automated solutions for network operations proposed by 3rd Generation Partnership Project (3GPP) group. Automated cell outage detection is one of the components of SON functionality. In early studies our research group developed data-driven approach for the detection of malfunctioning cells. In this paper we investigate the performance of the proposed solution as a function of the density of active users and the size of observation interval. The evaluation is conducted in Long Term E…
Effects of Temperature and Humidity on Radio Signal Strength in Outdoor Wireless Sensor Networks
2015
Many wireless sensor networks operating outdoors are exposed to changing weather conditions, which may cause severe degradation in system performance. Therefore, it is essential to explore the factors affecting radio link quality in order to mitigate their impact and to adapt to varying conditions. In this paper, we study the effects of temperature and humidity on radio signal strength in outdoor wireless sensor networks. Experimental measurements were performed using Atmel ZigBit 2.4GHz wireless modules, both in summer and wintertime. We employed all the radio channels specified by IEEE 802.15.4 for 2.4GHz ISM frequency band with two transmit power levels. The results show that changes in …
Higher-order Nonnegative CANDECOMP/PARAFAC Tensor Decomposition Using Proximal Algorithm
2019
Tensor decomposition is a powerful tool for analyzing multiway data. Nowadays, with the fast development of multisensor technology, more and more data appear in higherorder (order > 4) and nonnegative form. However, the decomposition of higher-order nonnegative tensor suffers from poor convergence and low speed. In this study, we propose a new nonnegative CANDECOM/PARAFAC (NCP) model using proximal algorithm. The block principal pivoting method in alternating nonnegative least squares (ANLS) framework is employed to minimize the objective function. Our method can guarantee the convergence and accelerate the computation. The results of experiments on both synthetic and real data demonstrate …
Detecting explosive substances by the IR spectrography
2014
Fast and safe detection methods of explosive substances are needed both before and after actualized explosions. This article presents an experiment of the detection of three selected explosives by the ATR FTIR spectrometer and by three different IR hyperspectral imaging devices. The IR spectrometers give accurate analyzing results, whereas hyperspectral imagers can detect and analyze desired samples without touching the unidentified target at all. In the controlled explosion experiment TNT, dynamite and PENO were at first analyzed as pure substances with the ATR FTIR spectrometer and with VNIR, SWIR and MWIR cameras. After three controlled explosions also the residues of TNT, dynamite and P…
The challenges of analysing blood stains with hyperspectral imaging
2014
Hyperspectral imaging is a potential noninvasive technology for detecting, separating and identifying various substances. In the forensic and military medicine and other CBRNE related use it could be a potential method for analyzing blood and for scanning other human based fluids. For example, it would be valuable to easily detect whether some traces of blood are from one or more persons or if there are some irrelevant substances or anomalies in the blood. This article represents an experiment of separating four persons' blood stains on a white cotton fabric with a SWIR hyperspectral camera and FT-NIR spectrometer. Each tested sample includes standardized 75 _l of 100 % blood. The results s…