Figure 4 - uploaded by Carolina Fortuna
Content may be subject to copyright.
Source publication
Location based services, already popular with end users, are now inevitably becoming part of new wireless infras-tructures and emerging business processes. The increasingly popular Deep Learning (DL) artificial intelligence methods perform very well in wireless fingerprinting localization based on extensive indoor radio measurement data. However, w...
Context in source publication
Context 1
... select the best model for each of the four evaluation sets, we evaluated accuracy as a function of epochs, as shown Further insight into the quality of the proposed model is provided by the histograms in Figure 4, depicting the distribution of predictions as a function of MDE for different dataset splits. In the case of Random, the spread of MSE values is very narrow around very small values and shows high accuracy. ...
Citations
... However, openly available and well documented REFIT and UK-DALE datasets, recently used in many reference works as well as in this study, do not exhibit suitable performance with the classical machine learning and are used with DL models. To address the high computational complexity and energy consumption associated with DL models, we designed a novel DL architecture based on the principles established in our previous work [36]. Additionally, to ensure consistency with the latest research, we compared our approach to works by Langevin et al. [19] and Tanoni et al. [16], who also utilized the same datasets of REFIT and UK-DALE. ...
... Architectures from this family, adapted for time series data, have previously proved successful in NILM disaggregation tasks [43]. Additionally, the hyper-parameters of the architectures were determined empirically following the principles derived from our prior work [36], determining the ratio between prediction performance and computational complexvi This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and content may change prior to final publication. ...
... We estimate the complexity of the most energy-consuming layers, namely the convolutional, pooling, and fullyconnected layers, using the equations presented by Pirnat et al. in [36]. In addition, we calculate the complexity of the GRU layer using the equation proposed in [44]. ...
Non-intrusive load monitoring (NILM) is the process of obtaining appliance-level data from a single metering point, measuring total electricity consumption of a household or a business. Appliance-level data can be directly used for demand response applications and energy management systems as well as for awareness raising and motivation for improvements in energy efficiency. Recently, classical machine learning and deep learning (DL) techniques became very popular and proved as highly effective for NILM classification, but with the growing complexity these methods are faced with significant computational and energy demands during both their training and operation. In this paper, we introduce a novel DL model aimed at enhanced multi-label classification of NILM with improved computation and energy efficiency. We also propose an evaluation methodology for comparison of different models using data synthesized from the measurement datasets so as to better represent real-world scenarios. Compared to the state-of-the-art, the proposed model has its energy consumption reduced by more than 23% while providing on average approximately 8 percentage points in performance improvement when evaluating on data derived from REFIT and UK-DALE datasets. We also show a 12 percentage point performance advantage of the proposed DL based model over a random forest model and observe performance degradation with the increase of the number of devices in the household, namely with each additional 5 devices, the average performance degrades by approximately 7 percentage points.
... However, openly available and well documented REFIT and UK-DALE datasets, recently used in many reference works as well as in this study, do not exhibit suitable performance with the classical machine learning and are used with DL models. To address the high computational complexity and carbon footprint associated with DL models, we designed a novel DL architecture based on the principles established in our previous work [28]. Additionally, to ensure consistency with the latest research, we compared our approach to works by Langevin et al. [25] and Tanoni et al. [16], who also utilized the same datasets of REFIT and UK-DALE. ...
... In particular, the choice of employing the GRU layer was motivated by its usage in a study that closely aligns with our own research [16]. In order to enhance energy efficiency, we adhered to the principles derived from our prior work [28], we minimized the number of convolutional layers and replaced the fifth block of layers with a single transposed convolutional layer. ...
... We estimate the complexity of the most energy-consuming layers, namely the convolutional, pooling, and fully-connected layers, using the equations presented by Pirnat et al. in [28]. In addition, we calculate the complexity of the GRU layer using the equation proposed in [41]. ...
Non-intrusive load monitoring (NILM) is the process of obtaining appliance-level data from a single metering point, measuring total electricity consumption of a household or a business. Appliance-level data can be directly used for demand response applications and energy management systems as well as for awareness raising and motivation for improvements in energy efficiency and reduction in the carbon footprint. Recently, classical machine learning and deep learning (DL) techniques became very popular and proved as highly effective for NILM classification, but with the growing complexity these methods are faced with significant computational and energy demands during both their training and operation. In this paper, we introduce a novel DL model aimed at enhanced multi-label classification of NILM with improved computation and energy efficiency. We also propose a testing methodology for comparison of different models using data synthesized from the measurement datasets so as to better represent real-world scenarios. Compared to the state-of-the-art, the proposed model has its carbon footprint reduced by more than 23% while providing on average approximately 8 percentage points in performance improvement when testing on data derived from REFIT and UK-DALE datasets.
... Bast et al (2020) took a different approach and replaced the convolutional layers for feature extraction with building blocks inspired by ResNet (He et al, 2016). Recent publications by Chin et al (2020); Cerar et al (2021); Pirnat et al (2022) focused on modifications where they emphasize the size of the neural network architecture (i.e., number of weights) in addition to the accuracy of position prediction. ...
... where F l refers to the l th layer of the architecture, which for some existing architectures proposed for localization, such as Pirnat et al (2022), can be either F f c from Eq. 1, F c from Eq. 4, or F p from Eq. 5. The final number of FLOPs can then be used to estimate the complexity of the proposed architecture, while directly correlating with the amount of energy required for the training and production phases of the DL model. ...
... Therefore we attempted to explore how to take an existing DL architecture and adapt it in order to perform comparably to the state of the art while significantly reducing the computational complexity and carbon footprint. Our findings, including the resulting PirnatEco architecture, were published in Pirnat et al (2022) while in this section we elaborate on the specifics of the data made available in the respective challenge and the design decisions made for developing PirnatEco. (2019). ...
Location based services, already popular with end users, are now inevitably becoming part of new wireless infrastructures and emerging business processes. The increasingly popular Deep Learning (DL) artificial intelligence methods perform very well in wireless fingerprinting localization based on extensive indoor radio measurement data. However, with the increasing complexity these methods become computationally very intensive and energy hungry, both for their training and subsequent operation. Considering only mobile users, estimated to exceed 7.4 billion by the end of 2025, and assuming that the networks serving these users will need to perform only one localization per user per hour on average, the machine learning models used for the calculation would need to perform predictions per year. Add to this equation tens of billions of other connected devices and applications that rely heavily on more frequent location updates, and it becomes apparent that localization will contribute significantly to carbon emissions unless more energy-efficient models are developed and used. In this Chapter, we discuss the latest results and trends in wireless localization and look at paths towards achieving more sustainable AI. We then elaborate on a methodology for computing DL model complexity, energy consumption and carbon footprint and show on a concrete example how to develop a more resource-aware model for fingerprinting. We finally compare relevant works in terms of complexity and training CO footprint.
The emerging B5G and 6G applications have brought forth the need for high-precision indoor localization. However, the complexity of indoor environments poses significant challenges to this goal, particularly due to the presence of non-line-of-sight (NLOS) conditions and multipath effects. This letter proposes an attention-based positioning network (ABPN) that exploits fine-grained features from MIMO channel state information (CSI) by spatial attention to combat the limited receptive field of traditional convolutional neural networks (CNNs) as well as channel attention to discriminate the importance of different wireless channels. Extensive experiments, conducted on two real-world datasets, demonstrate that the proposed ABPN outperforms the popular PirnatEco, AAresCNN, MIMOnet and CLnet with an average localization accuracy improvement of over 50%.
Location-based services have become an indispensable component of wireless networks, but high-precision positioning is challenging. With the application of multiple-input multiple-output (MIMO) in 5G, accurate channel state information (CSI) can be obtained and leveraged for high-precision positioning. Solving the MIMO positioning problem by deep learning has demonstrated better accuracy than traditional methods. To further improve the positioning accuracy, we propose a novel deep learning model named ACPNet, which incorporates two types of attention mechanisms and an improved training scheme. Experiment results show that compared to the state-of-the-art work, ACPNet exhibits more than 20% positioning accuracy improvement, and also maintains a relatively low computation complexity.
Location-based services, already popular with end users, are now inevitably becoming part of new wireless infrastructures and emerging business processes. The increasingly popular deep learning (DL) artificial intelligence methods perform very well in wireless fingerprinting localization based on extensive indoor radio measurement data. However, with the increasing complexity, these methods become computationally very intensive and energy hungry, both for their training and subsequent operation. Considering only mobile users, estimated to exceed 7.4 billion by the end of 2025, and assuming that the networks serving these users will need to perform only one localization per user per hour on average, the machine learning models used for the calculation would need to perform 65 × 1012 predictions per year. Add to this equation tens of billions of other connected devices and applications that rely heavily on more frequent location updates, and it becomes apparent that localization will contribute significantly to carbon emissions unless more energy-efficient models are developed and used. In this chapter, we discuss the latest results and trends in wireless localization and look at paths toward achieving more sustainable AI. We then elaborate on a methodology for computing DL model complexity, energy consumption, and carbon footprint and show on a concrete example how to develop a more resource-aware model for fingerprinting. We finally compare relevant works in terms of complexity and training CO2 footprint.KeywordsLocalizationFingerprintingSustainable AIDeep learning
