6 results on '"Margot Deruyck"'
Search Results
2. A framework for energy-efficient equine activity recognition with leg accelerometers
- Author
-
Wout Joseph, Jaron Fontaine, Eli De Poorter, Margot Deruyck, Luc Martens, Anniek Eerdekens, and David Plets
- Subjects
Signal processing ,0106 biological sciences ,Technology and Engineering ,Behaviour classification ,Computer science ,Equines ,Horticulture ,Accelerometer ,01 natural sciences ,Convolutional neural network ,Activity recognition ,Energy-efficient method ,Machine learning ,business.industry ,Internet-of-animals ,Forestry ,Pattern recognition ,04 agricultural and veterinary sciences ,Computer Science Applications ,Random forest ,040103 agronomy & agriculture ,0401 agriculture, forestry, and fisheries ,Artificial intelligence ,Accelerometers ,business ,Agronomy and Crop Science ,Energy (signal processing) ,010606 plant biology & botany ,Global time ,Efficient energy use - Abstract
Automated behavioral detection and classification through sensors can enhance the horses’ health and welfare. Since monitoring needs to be carried out continuously, an energy-efficient method is needed. The number of logging axes, sampling rate, and selected features of accelerometer data not only have a significant impact on classification accuracy in activity recognition but also on the sensors’ energy needs. Three models are designed for detecting horses’ activities namely, a Random Forest classifier (RF), a Convolutional Neural Network (CNN) and a hybrid CNN, i.e. a CNN fused with statistical features that retain knowledge about the global time series form. The models are validated using an experimental dataset obtained from six different horses each performing seven different activities. The results indicate that using one leg accelerometer data is sufficient for high classification accuracies ( > 98.6%) for the three models. The hybrid CNN substantially improves over the RF and CNN at a sampling rate of 5 Hz with an increase in accuracy of 1.88% and 2.79%, respectively. The hybrid CNN is capable of excellent performance, detecting nearly 99.59% of the behaviours at 10 Hz. The experiments show that the CNN and hybrid CNN use as much as 17.2 and 13.5 times less energy respectively, than the RF-based method. The experimental results showed that, although the recognition rate of the proposed optimized hybrid CNN model is similar to that of the original hybrid CNN model, the former requires only 6% of the Multiply-Accumulate (MAC) operations. For automatic detection of the horses’ behavior those results suggest using one leg accelerometer data sampled at 10 Hz classified by an optimized hybrid CNN.
- Published
- 2021
- Full Text
- View/download PDF
3. Multi-objective optimization of cognitive radio networks
- Author
-
Rodney Martinez Alonso, Wout Joseph, David Plets, Luc Martens, Margot Deruyck, and Glauco Guillen Nieto
- Subjects
Technology and Engineering ,Computer Networks and Communications ,Computer science ,Wireless network ,business.industry ,Cognitive radio ,Real-time computing ,020206 networking & telecommunications ,Cloud computing ,02 engineering and technology ,Spectral efficiency ,Spectrum management ,Multi-objective optimization ,0202 electrical engineering, electronic engineering, information engineering ,020201 artificial intelligence & image processing ,Performance indicator ,Spectrum efficiency ,business ,Network optimization ,Energy (signal processing) - Abstract
New generation networks, based on Cognitive Radio technology, allow dynamic allocation of the spectrum, alleviating spectrum scarcity. These networks also have a resilient potential for dynamic operation for energy saving. In this paper, we present a novel wireless network optimization algorithm for cognitive radio networks based on a cloud sharing-decision mechanism. Three Key Performance Indicators (KPIs) were optimized: spectrum usage, power consumption, and exposure. For a realistic suburban scenario in Ghent city, Belgium, we determine the optimal trade-off between the KPIs. Compared to a traditional Cognitive Radio network design, our optimization algorithm for the cloud-based architecture reduced the network power consumption by 27.5%, the average global exposure by 34.3%, and spectrum usage by 34.5% at the same time. Even for the worst-case optimization (worst achieved result of a single KPI), our solution performs better than the traditional architecture by 4.8% in terms of network power consumption, 7.3% in terms of spectrum usage, and 4.3% in terms of global exposure.
- Published
- 2021
- Full Text
- View/download PDF
4. Automatic equine activity detection by convolutional neural networks using accelerometer data
- Author
-
Margot Deruyck, Wout Joseph, Jaron Fontaine, Eli De Poorter, Luc Martens, and Anniek Eerdekens
- Subjects
0106 biological sciences ,Technology and Engineering ,Computer science ,Equines ,Interval (mathematics) ,Horticulture ,HORSES ,Accelerometer ,01 natural sciences ,Convolutional neural network ,Cross-validation ,Reduction (complexity) ,biology.animal ,Behaviour ,SCALE ,biology ,Pony ,business.industry ,Deep learning ,Convolutional Neural Networks ,PAIN ,Forestry ,Pattern recognition ,04 agricultural and veterinary sciences ,Computer Science Applications ,classification ,040103 agronomy & agriculture ,0401 agriculture, forestry, and fisheries ,Artificial intelligence ,Scale (map) ,business ,Agronomy and Crop Science ,010606 plant biology & botany - Abstract
In recent years, with a widespread of sensors embedded in all kind of mobile devices, human activity analysis is occurring more often in several domains like healthcare monitoring and fitness tracking. This trend did also enter the equestrian world because monitoring behaviours can yield important information about the health and welfare of horses. In this research, a deep learning-based approach for activity detection of equines is proposed to classify seven activities based on accelerometer data. We propose using Convolutional Neural Networks (CNN) by which features are extracted automatically by using strong computing capabilities. Furthermore, we investigate the impact of the sampling frequency, the time series length and the type of underground on which the data is gathered on the recognition accuracy and evaluate the model on three types of experimental datasets that are compiled of labelled accelerometer data gathered from six different subjects performing seven different activities. Afterwards, a horse-wise cross validation is carried out to investigate the impact of the subjects themselves on the model recognition accuracy. Finally, a slightly adjusted model is validated on different amounts of 50 Hz sensor data. A 99% accuracy can be reached for detecting seven behaviours of a seen horse when the sampling rate is 25 Hz and the time interval is 2.1 s. Four behaviours of an unseen horse can be detected with the same accuracy when the sampling rate is 69 Hz and the time interval is 2.4 s. Moreover, the accuracy of the model for the three datasets decreased on average with about 4.75% when the sampling rate was decreased from 200 Hz to 25 Hz and with 5.27% when the time interval was decreased from 3 s to 0.6 s. In addition, the classification performance of the activity ”walk” was not influenced by the type of underground the horse was performing this movement on and even the model could conclude from which underground the data was gathered for three out of four undergrounds with accuracies above 93% at time intervals higher than 1.2 s. This ensures the evaluation of activity patterns in real world circumstances. The performance and ability of the model to generalise is validated on 50 Hz data from different horse types, using ten-fold cross validation, reaching a mean classification accuracy of 97.84% and 96.10% when validated on a lame horse and pony, respectively. Moreover, in this work we show that using data from one sensors is at the cost of only 0.24% reduction in accuracy (99.42% vs 99.66%).
- Published
- 2020
- Full Text
- View/download PDF
5. Energy efficiency of femtocell deployment in combined wireless/optical access networks
- Author
-
Margot Deruyck, Mario Pickavet, Luc Martens, Slavisa Aleksic, Wout Joseph, and Willem Vereecken
- Subjects
Wi-Fi array ,business.product_category ,Computer Networks and Communications ,Computer science ,Access networks ,law.invention ,law ,Broadband ,Internet access ,Femtocell ,Wireless ,Wi-Fi ,Radio resource management ,Wired communication ,Fixed wireless ,Optical/wireless convergence ,Access network ,business.industry ,Wireless network ,ComputerSystemsOrganization_COMPUTER-COMMUNICATIONNETWORKS ,Wireless WAN ,Science General ,Telecommunications network ,Communication networks ,Wireless site survey ,Energy efficiency ,10G-PON ,IBCN ,business ,Telecommunications ,Municipal wireless network ,Efficient energy use ,Computer network - Abstract
Optical/wireless convergence has become of particular interest recently because a combined radio wireless and optical wired network has the potential to provide both mobility and high bandwidth in an efficient way. Recent developments of new radio access technologies such as the Long Term Evolution (LTE) and introduction of femtocell base stations open new perspectives in providing broadband services and applications to everyone and everywhere, but the instantaneous quality of radio channel varies in time, space and frequency and radio communication is inherently energy inefficient and susceptible to reflections and interference. On the other hand, optical fiber-based networks do not provide mobility, but they are robust, energy efficient, and able to provide both an almost unlimited bandwidth and high availability. In this paper, we analyze the energy efficiency of combined wireless/optical access networks, in which LTE technology provides ubiquitous broadband Internet access, while optical fiber-based technologies serve as wireless backhaul and offer high-bandwidth wired Internet access to business and residential customers. In this contest, we pay a particular attention to femtocell deployment for increasing both access data rates and area coverage. The paper presents a novel model for evaluating the energy efficiency of combined optical/wireless networks that takes into account the main architectural and implementational aspects of both RF wireless and optical parts of the access network. Several hypothetical network deployment scenarios are defined and used to study effects of femtocell deployment and power saving techniques on network's energy efficiency in urban, suburban and rural areas and for different traffic conditions.
- Published
- 2013
- Full Text
- View/download PDF
6. Modelling and optimization of power consumption in wireless access networks
- Author
-
Luc Martens, Wout Joseph, Margot Deruyck, and Emmeric Tanghe
- Subjects
Technology and Engineering ,Coverage ,Computer Networks and Communications ,Computer science ,Access networks ,MIMO ,Access stratum ,Base station ,Hardware_GENERAL ,Femtocell ,Wireless ,Radio resource management ,Access network ,business.industry ,IMT Advanced ,ComputerSystemsOrganization_COMPUTER-COMMUNICATIONNETWORKS ,Wireless WAN ,WiMAX ,Power consumption ,business ,UMTS frequency bands ,Efficient energy use ,Computer network ,Communication channel - Abstract
The power consumption of wireless access networks will become an important issue in the coming years. In this paper, the power consumption of base stations for mobile WiMAX, fixed WiMAX, UMTS, HSPA, and LTE is modelled and related to the coverage. A new metric, the power consumption per covered area PC"a"r"e"a, is introduced, to compare the energy efficiency of the considered technologies for a range of bit rates. Assuming the model parameters are correct, the conclusions are then as follows. For a 5MHz channel, UMTS is the most energy-efficient technology until a bit rate of 2.8Mbps, LTE between 2.8Mbps and 8.2Mbps, fixed WiMAX between 8.2Mbps and 13.8Mbps and finally mobile WiMAX for bit rates higher than 13.8Mbps. Furthermore, the influence of MIMO is investigated. For a 2x2 MIMO system, PC"a"r"e"a decreases by 36% for mobile WiMAX and by 23% for HSPA and LTE compared to the SISO system, resulting in a higher energy efficiency. The power consumption model for base stations is used in the deployment tool GRAND (Green Radio Access Network Design) for green wireless access networks. GRAND uses a genetic based algorithm and is applied on an actual case for the Brussels Capital Region, showing the possibilities of energy-efficient planning.
- Published
- 2011
- Full Text
- View/download PDF
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.