Rinat Khusainov

Real-time human ambulation, activity, and physiological monitoring: taxonomy of issues, techniques, applications, challenges and limitations.

Automated methods of real-time, unobtrusive, human ambulation, activity, and wellness monitoring and data analysis using various algorithmic techniques have been subjects of intense research. The general aim is to devise effective means of addressing the demands of assisted living, rehabilitation, and clinical observation and assessment through sensor-based monitoring. The research studies have resulted in a large amount of literature. This paper presents a holistic articulation of the research studies and offers comprehensive insights along four main axes: distribution of existing studies; monitoring device framework and sensor types; data collection, processing and analysis; and applications, limitations and challenges. The aim is to present a systematic and most complete study of literature in the area in order to identify research gaps and prioritize future research directions.

Sensor data acquisition and processing parameters for human activity classification.

It is known that parameter selection for data sampling frequency and segmentation techniques (including different methods and window sizes) has an impact on the classification accuracy. For Ambient Assisted Living (AAL), no clear information to select these parameters exists, hence a wide variety and inconsistency across todays literature is observed. This paper presents the empirical investigation of different data sampling rates, segmentation techniques and segmentation window sizes and their effect on the accuracy of Activity of Daily Living (ADL) event classification and computational load for two different accelerometer sensor datasets. The study is conducted using an ANalysis Of VAriance (ANOVA) based on 32 different window sizes, three different segmentation algorithm (with and without overlap, totaling in six different parameters) and six sampling frequencies for nine common classification algorithms. The classification accuracy is based on a feature vector consisting of Root Mean Square (RMS), Mean, Signal Magnitude Area (SMA), Signal Vector Magnitude (here SMV), Energy, Entropy, FFTPeak, Standard Deviation (STD). The results are presented alongside recommendations for the parameter selection on the basis of the best performing parameter combinations that are identified by means of the corresponding Pareto curve.

Artificial Immune Systems for Anomaly Detection in Ambient Assisted Living Applications

This paper makes a case for the use of Artificial Immune Systems AIS in the area of Ambient Assisted Living AAL for anomaly detection and long term monitoring. A brief literature review of some of the solutions developed for AAL and the use of AIS in other fields of research is presented. The authors advocate the use of AIS in AAL based on their unique features and their ability to address problems specific to the long term monitoring of people. An improved method for the optimisation of detector generation for AIS, which uses a novel intelligent seeding technique, is presented. The new seeding technique is compared with two other detector seeding methods. The simulation results are presented showing an improvement in the classification accuracy and warranting current and future work.

A two year case study: Technology Assisted Project Supervision (TAPaS)

Abstract During their final year undergraduate project a student may feel under-supported, stressed or isolated. In an internally funded project we set out to investigate the benefits of using a diverse blend of collaboration and communication tools alongside traditional methods of final year project supervision. We established separate formal and informal communication channels between the supervisor and their project students and a community of practice of students and supervisors was set up using twitter or a web forum. Using a wiki as a collaborative workspace and repository, student project pages were created and virtual supervision was blended with face-to-face supervision using electronic logs. At first some students were dubious about our approach and disapproved of at least one of the tools used. The supervisors were also initially sceptical of an increase in workload due to the multiplicity of tools used. In this paper we present how the staff and students benefited precisely because of the diverse range of tools used. The methods used resulted in transparency of students’ and supervisors’ actions, however, lessons were learnt about how to address student concerns about plagiarism in such an open environment.

On time series sensor data segmentation for fall and activity classification

The vast amount of literature on human ambulation and Activities of Daily Living (ADL) events classification has highlighted significant details on most aspects of the research area including: monitoring techniques, Wearable Sensor-based Monitoring Device (WSMD) placement on human body parts, and ambulation and ADL data collection methods, among others. However literature has failed to highlight meaningful details on one of the most important aspects of such studies, sensor data segmentation for feature extraction. The choice of segmentation techniques is in general very important, because inappropriate segmentation will most likely result in features without discriminant power. No classifier of whatever sophistication will give meaningful results with features that have no discriminant power. The optimal segmentation technique has been empirically investigated using sensor data from a bi-axial accelerometer. Results of the empirical investigation are presented.

Telecare:legal, ethical and socioeconomic factors

Legal, ethical and socio-economic factors in community telecare differ from those pertaining to telemedicine and are examined with reference to older persons’ care. Issues discussed include equipment liability, service malpractice, technical and service standards, consent (including the Mental Capacity Act), research, trials, human factors, dependence, privacy, security, accessibility, quality, affordability, social inequalities and community factors.

Rain granularity effects on bandwidth demand for faded DVB-RCS systems

Broadband satellite communication networks, operating at Ka band and above, play a vital role in today’s worldwide telecommunication infrastructure. The problem, however, is that rain can be the most dominant impairment factor for radio propagation above 10 GHz. This paper studies bandwidth and time slot allocation problem for rain faded DVB-RCS satellite networks. We investigate how using finer rain granularity can improve bandwidth utilization in DVB-RCS return links. The paper presents a mathematical model to calculate the bandwidth on demand. We formulate the radio resource allocation as an optimization problem and propose a novel algorithm for dynamic carrier bandwidth and time slots allocation, which works with constant bit rate type of traffic. We provide theoretical analysis for the time slot allocation problem and show that the proposed algorithm achieves optimal results. The algorithm is evaluated using a MATLAB simulation with historical rain data for the UK.

Email Reply Prediction: A Machine Learning Approach

Email has now become the most-used communication tool in the world and has also become the primary business productivity applications for most organizations and individuals. With the ever increasing popularity of emails, email over-load and prioritization becomes a major problem for many email users. Users spend a lot of time reading, replying and organizing their emails. To help users organize and prioritize their email messages, we propose a new framework; email reply prediction with unsupervised learning. The goal is to provide concise, highly structured and prioritized emails, thus saving the user from browsing through each email one by one and help to save time. In this paper, we discuss the features used to differentiate emails, show promising initial results with unsupervised machine learning model, and outline future directions for this work.

Evolving email clustering method for email grouping: A machine learning approach

This paper presents the design and implementation of a new system to manage email messages using email evolving clustering method with unsupervised learning approach to group emails base on activities found in the email messages, namely email grouping. Users spend a lot of time reading, replying and organizing their emails. To help users organize their email messages, we propose a new framework to help organise and prioritize email better. The goal is to provide highly structured and prioritized emails, thus saving the user from browsing through each email one by one and help to save time.

Activity Recognition from Video Data Using Spatial and Temporal Features

A method to monitor elderly people in an indoor environment using conventional cameras is presented. The method can be used to identify peoples activities and initiate suitable actions as needed. The originality of our approach is in combining spatial and temporal contexts with the position and orientation for the detected person. Preliminary evaluation, based only on the first two features (spatial and temporal), achieved the accuracy over 60% in a realistic residential environment. Although the results are based on using only two out of the four proposed input features, they already demonstrate a promising improvement over using a single feature in isolation.