Mini-Batch Alignment: A Deep-Learning Model for Domain Factor-Independent Feature Extraction for Wi-Fi-CSI Data.

Unobtrusive sensing (device-free sensing) aims to embed sensing into our daily lives. This is achievable by re-purposing communication technologies already used in our environments. Wireless Fidelity (Wi-Fi) sensing, using Channel State Information (CSI) measurement data, seems to be a perfect fit for this purpose since Wi-Fi networks are already omnipresent. However, a big challenge in this regard is CSI data being sensitive to 'domain factors' such as the position and orientation of a subject performing an activity or gesture. Due to these factors, CSI signal disturbances vary, causing domain shifts. Shifts lead to the lack of inference generalization, i.e., the model does not always perform well on unseen data during testing. We present a domain factor-independent feature-extraction pipeline called 'mini-batch alignment'. Mini-batch alignment steers a feature-extraction model's training process such that it is unable to separate intermediate feature-probability density functions of input data batches seen previously from the current input data batch. By means of this steering technique, we hypothesize that mini-batch alignment (i) absolves the need for providing a domain label, (ii) reduces pipeline re-building and re-training likelihood when encountering latent domain factors, and (iii) absolves the need for extra model storage and training time. We test this hypothesis via a vast number of performance-evaluation experiments. The experiments involve both one- and two-domain-factor leave-out cross-validation, two open-source gesture-recognition datasets called SignFi and Widar3, two pre-processed input types called Doppler Frequency Spectrum (DFS) and Gramian Angular Difference Field (GADF), and several existing domain-shift mitigation techniques. We show that mini-batch alignment performs on a par with other domain-shift mitigation techniques in both position and orientation one-domain leave-out cross-validation using the Widar3 dataset and DFS as input type. When considering a memory-complexity-reduced version of the GADF as input type, mini-batch alignment shows hints of recuperating performance regarding a standard baseline model to the extent that no additional performance due to weight steering is lost in both one-domain-factor leave-out and two-orientation-domain-factor leave-out cross-validation scenarios. However, this is not enough evidence that the mini-batch alignment hypothesis is valid. We identified pitfalls leading up to the hypothesis invalidation: (i) lack of good-quality benchmark datasets, (ii) invalid probability distribution assumptions, and (iii) non-linear distribution scaling issues.

Use of Domain Labels during Pre-Training for Domain-Independent WiFi-CSI Gesture Recognition.

To minimize dependency on the availability of data labels, some WiFi-CSI based-gesture recognition solutions utilize an unsupervised representation learning phase prior to fine-tuning downstream task classifiers. In this case, however, the overall performance of the solution is negatively affected by domain factors present in the WiFi-CSI data used by the pre-training models. To reduce this negative effect, we propose an integration of the adversarial domain classifier in the pre-training phase. We consider this as an effective step towards automatic domain discovery during pre-training. We also experiment with multi-class and label versions of domain classification to improve situations, in which integrating a multi-class and single label-based domain classifier during pre-training fails to reduce the negative impact domain factors have on overall solution performance. For our extensive random and leave-out domain factor cross-validation experiments, we utilise (i) an end-to-end and unsupervised representation learning baseline, (ii) integration of both single- and multi-label domain classification, and (iii) so-called domain-aware versions of the aformentioned unsupervised representation learning baseline in (i) with two different datasets, i.e., Widar3 and SignFi. We also consider an input sample type that generalizes, in terms of overall solution performance, to both aforementioned datasets. Experiment results with the Widar3 dataset indicate that multi-label domain classification reduces domain shift in position (1.2% mean metric improvement and 0.5% variance increase) and orientation (0.4% mean metric improvement and 1.0% variance decrease) in domain factor leave-out cross-validation experiments. The results also indicate that domain shift reduction, when considering single- or multi-label domain classification during pre-training, is negatively impacted when a large proportion of negative view combinations contain views that originate from different domains within a substantial amount of mini-batches considered during pre-training. This is caused by the view contrastive loss repelling the aforementioned negative view combinations, eventually causing more domain shift in the intermediate feature space of the overall solution.

A Codeword-Independent Localization Technique for Reconfigurable Intelligent Surface Enhanced Environments Using Adversarial Learning.

Reconfigurable Intelligent Surfaces (RISs) not only enable software-defined radio in modern wireless communication networks but also have the potential to be utilized for localization. Most previous works used channel matrices to calculate locations, requiring extensive field measurements, which leads to rapidly growing complexity. Although a few studies have designed fingerprint-based systems, they are only feasible under an unrealistic assumption that the RIS will be deployed only for localization purposes. Additionally, all these methods utilize RIS codewords for location inference, inducing considerable communication burdens. In this paper, we propose a new localization technique for RIS-enhanced environments that does not require RIS codewords for online location inference. Our proposed approach extracts codeword-independent representations of fingerprints using a domain adversarial neural network. We evaluated our solution using the DeepMIMO dataset. Due to the lack of results from other studies, for fair comparisons, we define oracle and baseline cases, which are the theoretical upper and lower bounds of our system, respectively. In all experiments, our proposed solution performed much more similarly to the oracle cases than the baseline cases, demonstrating the effectiveness and robustness of our method.

Selfish Behavior in IEEE 802.11ah Networks: A Detection Algorithm and Mitigation Strategies.

One of the latest protocols developed for the Internet of Things networks is IEEE 802.11ah, proposed by the WiFi Alliance. The new channel access mechanism in IEEE 802.11ah, which is called Restricted Access Window, aims at reducing the contention between the stations by allowing only selected stations to transmit data at certain time slots. Stations may exhibit selfish behavior to maximize their own throughput. This will come at the cost of the overall network quality of service. In this paper, we first analyze the default behavior of the IEEE 802.11ah protocol in terms of fairness. We then introduce various percentages of selfish stations and observe how the network's quality of service in terms of fairness, throughput and packet-loss are affected. After establishing the inherent fairness of IEEE 802.11ah, we analyze applicability of two existing selfish behavior detection algorithms designed for IEEE 802.11 to the IEEE 802.11ah protocol. Due to their poor performance, we propose a new definition of 'selfish behavior' specifically for IEEE 802.11ah, based on which we present a new algorithm for detecting selfish behavior. To combat selfish behavior and to create a better fairness, throughput and lower packet loss, we consequently present a novel mitigation algorithm called Selfish Stations Quarantine Punishment Algorithm (SSQPA). The proposed algorithm takes advantage of the RAW grouping to isolate selfish stations from the honest stations, thus mitigating the effect of the selfish behavior. SSQPA comes in two variants: honest stations-centric and network-centric. Our experimental results show that both variants can successfully mitigate selfish behavior effects in IEEE 802.11ah networks and either one can be used depending on the goal of the network.

A Domain-Independent Generative Adversarial Network for Activity Recognition Using WiFi CSI Data.

Over the past years, device-free sensing has received considerable attention due to its unobtrusiveness. In this regard, context recognition using WiFi Channel State Information (CSI) data has gained popularity, and various techniques have been proposed that combine unobtrusive sensing and deep learning to accurately detect various contexts ranging from human activities to gestures. However, research has shown that the performance of these techniques significantly degrades due to change in various factors including sensing environment, data collection configuration, diversity of target subjects, and target learning task (e.g., activities, gestures, emotions, vital signs). This problem, generally known as the domain change problem, is typically addressed by collecting more data and learning the data distribution that covers multiple factors impacting the performance. However, activity recognition data collection is a very labor-intensive and time consuming task, and there are too many known and unknown factors impacting WiFi CSI signals. In this paper, we propose a domain-independent generative adversarial network for WiFi CSI based activity recognition in combination with a simplified data pre-processing module. Our evaluation results show superiority of our proposed approach compared to the state of the art in terms of increased robustness against domain change, higher accuracy of activity recognition, and reduced model complexity.

Threat Modeling-How to Visualize Attacks on IOTA?

Internet of Things (IoT) has been deployed in a vast number of smart applications with the aim to bring ease and comfort into our lives. However, with the expansion of IoT applications, the number of security and privacy breaches has also increased, which brings into question the resilience of existing security and trust mechanisms. Furthermore, the contemporaneous centralized technology is posing significant challenges viz scalability, transparency and efficiency to wide range of IoT applications such as smart logistics, where millions of IoT devices need to be connected simultaneously. Alternatively, IOTA is a distributed ledger technology that offers resilient security and trust mechanisms and a decentralized architecture to overcome IoT impediments. IOTA has already been implemented in many applications and has clearly demonstrated its significance in real-world applications. Like any other technology, IOTA unfortunately also encounters security vulnerabilities. The purpose of this study is to explore and highlight security vulnerabilities of IOTA and simultaneously demonstrate the value of threat modeling in evaluating security vulnerabilities of distributed ledger technology. IOTA vulnerabilities are scrutinized in terms of feasibility and impact and we have also presented prevention techniques where applicable. To identify IOTA vulnerabilities, we have examined existing literature and online blogs. Literature available on this topic is very limited so far. As far as we know IOTA has barely been addressed in the traditional journals, conferences and books. In total we have identified six vulnerabilities. We used Common Vulnerability Scoring System (CVSS v3.0) to further categorize these vulnerabilities on the basis of their feasibility and impact.

Coalition Formation Based Compressive Sensing in Wireless Sensor Networks.

Compressive sensing originates in the field of signal processing and has recently become a topic of energy-efficient data gathering in wireless sensor networks. In this paper, we propose an energy efficient distributed compressive sensing solution for sensor networks. The proposed solution utilizes sparsity distribution of signals to group sensor nodes into several coalitions and then implements localized compressive sensing inside coalitions. This solution improves data-gathering performance in terms of both data accuracy and energy consumption. The approach curbs both data-transmission costs and number of measurements. Coalition-based data gathering cuts transmission costs, and the number of measurements is reduced by scheduling sensor nodes and adjusting their sampling frequency. Our simulation showed that our approach enhances network performance by minimizing energy cost and improving data accuracy.

Towards an Online Seizure Advisory System-An Adaptive Seizure Prediction Framework Using Active Learning Heuristics.

In the last decade, seizure prediction systems have gained a lot of attention because of their enormous potential to largely improve the quality-of-life of the epileptic patients. The accuracy of the prediction algorithms to detect seizure in real-world applications is largely limited because the brain signals are inherently uncertain and affected by various factors, such as environment, age, drug intake, etc., in addition to the internal artefacts that occur during the process of recording the brain signals. To deal with such ambiguity, researchers transitionally use active learning, which selects the ambiguous data to be annotated by an expert and updates the classification model dynamically. However, selecting the particular data from a pool of large ambiguous datasets to be labelled by an expert is still a challenging problem. In this paper, we propose an active learning-based prediction framework that aims to improve the accuracy of the prediction with a minimum number of labelled data. The core technique of our framework is employing the Bernoulli-Gaussian Mixture model (BGMM) to determine the feature samples that have the most ambiguity to be annotated by an expert. By doing so, our approach facilitates expert intervention as well as increasing medical reliability. We evaluate seven different classifiers in terms of the classification time and memory required. An active learning framework built on top of the best performing classifier is evaluated in terms of required annotation effort to achieve a high level of prediction accuracy. The results show that our approach can achieve the same accuracy as a Support Vector Machine (SVM) classifier using only 20 % of the labelled data and also improve the prediction accuracy even under the noisy condition.

Poaching Detection Technologies-A Survey.

Between 1960 and 1990, 95% of the black rhino population in the world was killed. In South Africa, a rhino was killed every 8 h for its horn throughout 2016. Wild animals, rhinos and elephants, in particular, are facing an ever increasing poaching crisis. In this paper, we review poaching detection technologies that aim to save endangered species from extinction. We present requirements for effective poacher detection and identify research challenges through the survey. We describe poaching detection technologies in four domains: perimeter based, ground based, aerial based, and animal tagging based technologies. Moreover, we discuss the different types of sensor technologies that are used in intruder detection systems such as: radar, magnetic, acoustic, optic, infrared and thermal, radio frequency, motion, seismic, chemical, and animal sentinels. The ultimate long-term solution for the poaching crisis is to remove the drivers of demand by educating people in demanding countries and raising awareness of the poaching crisis. Until prevention of poaching takes effect, there will be a continuous urgent need for new (combined) approaches that take up the research challenges and provide better protection against poaching in wildlife areas.

Potential of Wake-Up Radio-Based MAC Protocols for Implantable Body Sensor Networks (IBSN)-A Survey.

With the advent of nano-technology, medical sensors and devices are becoming highly miniaturized. Consequently, the number of sensors and medical devices being implanted to accurately monitor and diagnose a disease is increasing. By measuring the symptoms and controlling a medical device as close as possible to the source, these implantable devices are able to save lives. A wireless link between medical sensors and implantable medical devices is essential in the case of closed-loop medical devices, in which symptoms of the diseases are monitored by sensors that are not placed in close proximity of the therapeutic device. Medium Access Control (MAC) is crucial to make it possible for several medical devices to communicate using a shared wireless medium in such a way that minimum delay, maximum throughput, and increased network life-time are guaranteed. To guarantee this Quality of Service (QoS), the MAC protocols control the main sources of limited resource wastage, namely the idle-listening, packet collisions, over-hearing, and packet loss. Traditional MAC protocols designed for body sensor networks are not directly applicable to Implantable Body Sensor Networks (IBSN) because of the dynamic nature of the radio channel within the human body and the strict QoS requirements of IBSN applications. Although numerous MAC protocols are available in the literature, the majority of them are designed for Body Sensor Network (BSN) and Wireless Sensor Network (WSN). To the best of our knowledge, there is so far no research paper that explores the impact of these MAC protocols specifically for IBSN. MAC protocols designed for implantable devices are still in their infancy and one of their most challenging objectives is to be ultra-low-power. One of the technological solutions to achieve this objective so is to integrate the concept of Wake-up radio (WuR) into the MAC design. In this survey, we present a taxonomy of MAC protocols based on their use of WuR technology and identify their bottlenecks to be used in IBSN applications. Furthermore, we present a number of open research challenges and requirements for designing an energy-efficient and reliable wireless communication protocol for IBSN.

Underwater acoustic wireless sensor networks: advances and future trends in physical, MAC and routing layers.

This survey aims to provide a comprehensive overview of the current research on underwater wireless sensor networks, focusing on the lower layers of the communication stack, and envisions future trends and challenges. It analyzes the current state-of-the-art on the physical, medium access control and routing layers. It summarizes their security threads and surveys the currently proposed studies. Current envisioned niches for further advances in underwater networks research range from efficient, low-power algorithms and modulations to intelligent, energy-aware routing and medium access control protocols.

Sensing solutions for collecting spatio-temporal data for wildlife monitoring applications: a review.

Movement ecology is a field which places movement as a basis for understanding animal behavior. To realize this concept, ecologists rely on data collection technologies providing spatio-temporal data in order to analyze movement. Recently, wireless sensor networks have offered new opportunities for data collection from remote places through multi-hop communication and collaborative capability of the nodes. Several technologies can be used in such networks for sensing purposes and for collecting spatio-temporal data from animals. In this paper, we investigate and review technological solutions which can be used for collecting data for wildlife monitoring. Our aim is to provide an overview of different sensing technologies used for wildlife monitoring and to review their capabilities in terms of data they provide for modeling movement behavior of animals.

Underwater sensor networks: a new energy efficient and robust architecture.

The specific characteristics of underwater environments introduce new challenges for networking protocols. In this paper, a specialized architecture for underwater sensor networks (UWSNs) is proposed and evaluated. Experiments are conducted in order to analyze the suitability of this protocol for the subaquatic transmission medium. Moreover, different scheduling techniques are applied to the architecture in order to study their performance. In addition, given the harsh conditions of the underwater medium, different retransmission methods are combined with the scheduling techniques. Finally, simulation results illustrate the performance achievements of the proposed protocol in end-to-end delay, packet delivery ratio and energy consumption, showing that this protocol can be very suitable for the underwater medium.

Sensor networks in the low lands.

This paper provides an overview of scientific and industrial developments of the last decade in the area of sensor networks in The Netherlands (Low Lands). The goal is to highlight areas in which the Netherlands has made most contributions and is currently a dominant player in the field of sensor networks. On the one hand, motivations, addressed topics, and initiatives taken in this period are presented, while on the other hand, special emphasis is given to identifying current and future trends and formulating a vision for the coming five to ten years. The presented overview and trend analysis clearly show that Dutch research and industrial efforts, in line with recent worldwide developments in the field of sensor technology, present a clear shift from sensor node platforms, operating systems, communication, networking, and data management aspects of the sensor networks to reasoning/cognition, control, and actuation.