Data Mining Practical Machine Learning Tools and Techniques

Human activity recognition systems are developed as part of a framework to enable continuous monitoring of human behaviours in the area of ambient assisted living, sports injury detection, elderly care, rehabilitation, and entertainment and surveillance in smart home environments. The extraction of relevant features is the most challenging part of the mobile and wearable sensor-based human activity recognition pipeline. Feature extraction influences the algorithm performance and reduces computation time and complexity. However, current human activity recognition relies on handcrafted features that are incapable of handling complex activities especially with the current influx of multimodal and high dimensional sensor data. With the emergence of deep learning and increased computation powers, deep learning and artificial intelligence methods are being adopted for automatic feature learning in diverse areas like health, image classification, and recently, for feature extraction and classification of simple and complex human activity recognition in mobile and wearable sensors. Furthermore, the fusion of mobile or wearable sensors and deep learning methods for feature learning provide diversity, offers higher generalisation, and tackles challenging issues in human activity recognition. The focus of this review is to provide in-depth summaries of deep learning methods for mobile and wearable sensor-based human activity recognition. The review presents the methods, uniqueness, advantages and their limitations. We not only categorise the studies into generative, discriminative and hybrid methods but also highlight their important advantages. Furthermore, the review presents classification and evaluation procedures and discusses publicly available datasets for mobile sensor human activity recognition. Finally, we outline and explain some challenges to open research problems that require further research and improvements.

Deep learning algorithms for human activity recognition using mobile and wearable sensor networks: State of the art and research challenges

The Internet of Things (IoT) is one of the most promising technologies for the near future. Healthcare and well-being will receive great benefits with the evolution of this technology. This paper presents a review of techniques based on IoT for healthcare and ambient-assisted living, defined as the Internet of Health Things (IoHT), based on the most recent publications and products available in the market from industry for this segment. Also, this paper identifies the technological advances made so far, analyzing the challenges to be overcome and provides an approach of future trends. Through selected works, it is possible to notice that further studies are important to improve current techniques and that novel concept and technologies of IoHT are needed to overcome the identified challenges. The presented results aim to serve as a source of information for healthcare providers, researchers, technology specialists, and the general population to improve the IoHT.

Enabling Technologies for the Internet of Health Things

Sensor-based and vision-based human activity recognition: A comprehensive survey

Data fusion and multiple classifier systems for human activity detection and health monitoring: Review and open research directions

This chapter elaborates on radio resource management issues for an optimum frequency portion exploitation and maximum energy conservation in a centralised cognitive radio network architecture. Two optimisation algorithms are provided that enable the efficient use of TV White Spaces (TVWS) through a broker entity that coordinates the spectrum trading process among a number of wireless secondary systems. TVWS can be a mixed economy with licence-exempt use, or light-licensed or licensed approaches. The resource management can facilitate different mixes according to alternative context and needs. In this framework, this chapter is mainly focused on the effective use of TVWS as a part of the mobile operators' spectrum portfolio. This is one strand of TVWS administration, while the chapter does not cover other use-cases for the exploitation of this spectrum part by microphone users or Wireless Internet Service Providers (WISPs). The proposed approach follows a real-time spectrum trading scheme based on either a fixed-price or a combinatorial auction-based method, depending on the requirements for spectrum by the secondary systems. The validity of the two energy-efficient algorithms is tested via several simulations, carried out under controlled conditions. The simulations evaluate the performance of the algorithms in terms of energy consumption levels, the spectrum broker benefit, the end-to-end latency and the throughput response.

TV White Space Spectrum Administration

The immense network of interconnected objects, called the Internet of Things (IoT), will eventually become a reality thanks to the MP2P (Mobile Peer-to-Peer) nature of the upcoming 5G and it will have the potential of growing into a valuable asset and revolutionize a plethora of fields. Due to its omnipresent character, the IoT can be successfully deployed in the area of healthcare, allowing this way the efficient management of equipment in medical facilities, as well as the immediate and remote monitoring and treatment of patients. In addition to IoT, cloud computing can also be integrated in such systems, opening this way new doors for the field of Ambient Assisted Living (AAL) by taking advantage of the benefits the cloud brings to the table when properly utilized. This chapter briefly presents IoT and cloud computing, as well as how their combination can affect the area of AAL. It also provides information on how cloud services can be used for remotely monitoring the physiological condition of a patient, along with how IoT can be used in gathering the desired data. Additionally, Electronic Health Records (EHR) are presented, followed by how wearable technology, used in AAL, can benefit from an IoT cloud system. Finally, challenges and issues that appear from the utilization of cloud-based IoT for AAL are presented, such as security and privacy requirements, the difficulties in handling the massive amount of data generated, as well as device interoperability issues.

Rossitza Goleva

Papers

TV White Space Spectrum Administration

Cloud-Oriented Domain for AAL