Hidden semi-Markov Model Research Articles

Artificial bee colony optimization based energy-efficient wireless network interface selection for industrial mobile devices

Management of energy is a crucial challenge for mobile devices, having the network activity frequently taking up a considerable part of the energy in the overall system. The important goal of this work is to offer energy efficiency for mobile handhelds. The industrial mobile devices in the recent times are fitted with several wireless network interfaces, these devices function with less battery power. In mobile handhelds, each user is able to run multiple applications with the purpose of have particular Quality of Service (QoS) requirements. Hence, it would be beneficial for a multi-interface terminal to simultaneously use two or more interfaces to gain in performance. So satisfying the QoS becomes very difficult task. An energy-effective Adaptive Wireless Network Interface Selection (AWNIS) with Artificial Bee Colony (ABC) is proposed in this paper. AWNIS-ABC algorithm chooses the best of the wireless network interface in regard to the energy consumed by taking the QoS of the link into consideration as well as following a hi-speed network interface-selection interval based on the scenario in the network. Then Hidden Semi Markov Model (HSMM) is introduced for searching the detouring paths if the energy state of a given relay node is very less to utilize the required interface, which ensures timely delivery. Also, it provides the diversification of the route paths in the network with the purpose of extending the lifetime of the network. An alternate path is proposed for reliable data delivery beforehand the route path encounters a breakage owing to energy exhaustion is found. A number of wireless network interfaces in mobile devices are utilized in improving the energy efficiency, satisfaction of QoS occurring during data transfer. The simulation of the network is implemented via the Network Simulator 2(NS2).

A New State Recognition and Prognosis Method Based on a Sparse Representation Feature and the Hidden Semi-Markov Model

Equipment degradation state recognition and prognosis are considered two significant parts of a prognostics and health management (PHM) system that help to reduce downtime and decrease economic losses. In this paper, a sparse representation (SR) feature is proposed as a new degradation feature, and the hidden semi-Markov model (HSMM) is established. The new method offers three significant advantages over the traditional HSMM. (1) Since the degradation information is incomplete, a Gaussian mixture model (GMM) is used here for degradation data clustering and state division. (2) A new degradation feature based on the wavelet packet transform (WPT) and SR can better extract the structural information of the collected signal and reflect the degradation characteristics. (3) To conduct remaining useful life (RUL) predictions, an improved model is proposed, which adds a control variable that can dynamically adjust the state duration. The effectiveness of the proposed method is demonstrated using 8 groups of bearing data from the Center for Intelligent Maintenance Systems (IMS). The results show that the HSMM with the SR feature achieves the best recognition accuracy, of 85.28%. Moreover, the improved prediction model achieves a prediction accuracy of 86.11% on average for 8 bearings.

Visual tracking enhancement by trajectory simulation based on hidden semi‐Markov model

In this Letter, the authors present a novel tracking system, in which tracking accuracy can be enhanced by trajectory simulation. They generate synthetic trajectories based on observed trajectories by adopting the hidden semi-Markov model (HSMM). In the course of trajectory simulation, HSMM can encode representative states and speeds of the targets. The simulated trajectories enforce the proposed tracker to focus on the areas where targets will move at following frames. Experimental results demonstrate that it is easy to integrate the proposed trajectory simulation into traditional visual tracking methods and the trajectory simulation can considerably improve the accuracy of visual trackers.

Hidden Markov Model of System Elements Technical Maintenance by Age

Technical maintenance is between the methods of operation reliability and effectiveness increasing for systems of different purposes including power systems. In the paper the hidden semi-Markov model of technical maintenance is built basing on the semi-Markov model of two-component system elements technical maintenance by age. The hidden Markov model is used to solve the problems of dynamics analyzing, predicting the states of a system modelled based on the vector of signals obtained during its operation.

An Improved Artificial Bees Colony Algorithm to Solve Minimal Exposure Problem in Wireless Sensor Networks

The problem of exposure, which is associated with the quality of coverage, is a basic issue faced in wireless sensor networks. Search for the Minimum Exposure Path (MEP) is one among the crucial problems in Wireless Sensor Networks (WSNs). The exposure with respect to the MEP is one of the indicators used for measuring the coverage quality of a WSN, depicting the efficiency with which the monitoring of a mobile target is done all along the sensing field. The available hybrid technique does not have sufficient accuracy to get the MEP, which is too complicated, and is not suitable to networks having heterogeneous sensor nodes, or in large number, or an all-sensor field intensity function. In order to get over these problems, an Improved Artificial Bee Colony (IABC) model is introduced for the MEP problem associated with getting the optimal shortest path in this research work. The IABC model exhibits an extensive use of energy and data congestion will happen. In order to have an efficient resolution to the data congestion problem, Prim’s algorithm has introduced the link level congestion scenario. Also for effective solving of this issue, in accordance with the features of the node, a semi Hidden Markov Model is developed to solve the problem of energy efficiency in WSN model. A number of experiments were carried out, and the results show that the proposed model and the designed IABC can help in increasing the solution accuracy and can be useful to not just the heterogeneous sensors’ case but also the scenario in where there are a big number of sensor nodes and also an all-sensor field intensity function.

HVGH: Unsupervised Segmentation for High-Dimensional Time Series Using Deep Neural Compression and Statistical Generative Model.

Humans perceive continuous high-dimensional information by dividing it into meaningful segments, such as words and units of motion. We believe that such unsupervised segmentation is also important for robots to learn topics such as language and motion. To this end, we previously proposed a hierarchical Dirichlet process–Gaussian process–hidden semi-Markov model (HDP-GP-HSMM). However, an important drawback of this model is that it cannot divide high-dimensional time-series data. Furthermore, low-dimensional features must be extracted in advance. Segmentation largely depends on the design of features, and it is difficult to design effective features, especially in the case of high-dimensional data. To overcome this problem, this study proposes a hierarchical Dirichlet process–variational autoencoder–Gaussian process–hidden semi-Markov model (HVGH). The parameters of the proposed HVGH are estimated through a mutual learning loop of the variational autoencoder and our previously proposed HDP-GP-HSMM. Hence, HVGH can extract features from high-dimensional time-series data while simultaneously dividing it into segments in an unsupervised manner. In an experiment, we used various motion-capture data to demonstrate that our proposed model estimates the correct number of classes and more accurate segments than baseline methods. Moreover, we show that the proposed method can learn latent space suitable for segmentation.

Analysis and Comparison of Bitcoin and S and P 500 Market Features Using HMMs and HSMMs †

We implement hidden Markov models (HMMs) and hidden semi-Markov models (HSMMs) on Bitcoin/US dollar (BTC/USD) with the aim of market phase detection. We make analogous comparisons to Standard and Poor’s 500 (S and P 500), a benchmark traditional stock index and a protagonist of several studies in finance. Popular labels given to market phases are “bull”, “bear”, “correction”, and “rally”. In the first part, we fit HMMs and HSMMs and look at the evolution of hidden state parameters and state persistence parameters over time to ensure that states are correctly classified in terms of market phase labels. We conclude that our modelling approaches yield positive results in both BTC/USD and the S and P 500, and both are best modelled via four-state HSMMs. However, the two assets show different regime volatility and persistence patterns—BTC/USD has volatile bull and bear states and generally weak state persistence, while the S and P 500 shows lower volatility on the bull states and stronger state persistence. In the second part, we put our models to the test of detecting different market phases by devising investment strategies that aim to be more profitable on unseen data in comparison to a buy-and-hold approach. In both cases, for select investment strategies, four-state HSMMs are also the most profitable and significantly outperform the buy-and-hold strategy.

Haptics Electromyography Perception and Learning Enhanced Intelligence for Teleoperated Robot

Due to the lack of transparent and friendly human-robot interaction (HRI) interface, as well as various uncertainties, it is usually a challenge to remotely manipulate a robot to accomplish a complicated task. To improve the teleoperation performance, we propose a new perception mechanism by integrating a novel learning method to operate the robots in the distance. In order to enhance the perception of the teleoperation system, we utilize a surface electromyogram signal to extract the human operator's muscle activation. As a response to the changes in the external environment, as sensed through haptic and visual feedback, a human operator naturally reacts with various muscle activations. By imitating the human behaviors in task execution, not only motion trajectory but also arm stiffness adjusted by muscle activation, it is expected that the robot would be able to carry out the repetitive tasks autonomously or uncertain tasks with improved intelligence. To this end, we develop a robot learning algorithm based on probability statistics under an integrated framework of the hidden semi-Markov model (HSMM) and the Gaussian mixture method. This method is employed to obtain a generative task model based on the robot's trajectory. Then, Gaussian mixture regression based on HSMM is applied to correct the robot trajectory with the reproduced results from the learned task model. The execution procedures consist of a learning phase and a reproduction phase. To guarantee the stability, immersion, and maneuverability of the teleoperation system, a variable gain control method that involves electromyography (EMG) is introduced. Experimental results have demonstrated the effectiveness of the proposed method.

Robot complex motion learning based on unsupervised trajectory segmentation and movement primitives

This paper presents a robot skill acquisition framework for learning and reproducing humanoid trajectories with complex forms. A new unsupervised segmentation method is proposed to detect motion units in the demonstrated kinematic data using the concept of key points. To find the consistent features of trajectories, a Hidden Semi-Markov Model (HSMM) is used to identify key points common to all the demonstrations. Generalizing the motion units is achieved via a Probability-based Movement Primitive (PbMP), which encapsulates multiple trajectories into one model. Such a framework can generate trajectories suitable for robot execution with arbitrary shape and complexity from a small number of demonstrations, which greatly expands the application scenarios of robot programming by demonstration. The automatic segmentation process does not rely on a priori knowledge or models for specific tasks, and the generalized trajectory retains more consistent features than those produced by other algorithms. We demonstrate the effectiveness of the proposed framework through simulations and experiments.

A Continuous-Time Model-Based Approach for Activity Recognition in Pervasive Environments

We present a model-based approach to Activity Recognition (AR) in Ambient Assisted Living (AAL). The approach leverages an a priori stochastic model termed Continuous-Time Hidden Semi-Markov Model (CT-HSMM), capturing the continuous-time durations of activities and inter-event times. The model is enhanced according to the observed statistics, associating the events with an occurrence probability, and the sojourn time and the inter-event time in each activity with a continuous-time probability density function, allowing effective fitting of observed durations through non-Markovian distributions. The model is updated at run time according to a sequence of time-stamped observations, exploiting the method of stochastic state classes to perform transient analysis and derive a measure of likelihood that an activity is currently performed. The approach supports both online AR, predicting the activity performed at time $t$ using only the events observed until that time, and offline AR, applying a forward–backward procedure that exploits all the events observed before and after time $t$ . The approach is experimented on a real dataset of the literature, providing performance measures that can be compared with those of offline Hidden Markov Models (HMMs) and offline Hidden Semi-Markov Models (HSMMs).

Driving Style Analysis Using Primitive Driving Patterns With Bayesian Nonparametric Approaches

Analysis and recognition of driving styles are profoundly important to intelligent transportation and vehicle calibration. This paper presents a novel driving style analysis framework using the primitive driving patterns learned from naturalistic driving data. In order to achieve this, first, a Bayesian nonparametric learning method based on a hidden semi-Markov model (HSMM) is introduced to extract primitive driving patterns from time series driving data without prior knowledge of the number of these patterns. In the Bayesian nonparametric approach, we utilize a hierarchical Dirichlet process (HDP) instead of learning the unknown number of smooth dynamical modes of HSMM, thus generating the primitive driving patterns. Each primitive pattern is clustered and then labeled using behavioral semantics according to drivers' physical and psychological perception thresholds. For each driver, 75 primitive driving patterns in car-following scenarios are learned and semantically labeled. In order to show the HDP-HSMM's utility to learn primitive driving patterns, other two Bayesian nonparametric approaches, HDP-HMM and sticky HDP-HMM, are compared. The naturalistic driving data of 18 drivers were collected from the University of Michigan Safety Pilot Model Deployment (SPDM) database. The individual driving styles are discussed according to distribution characteristics of the learned primitive driving patterns and also the difference in driving styles among drivers are evaluated using the Kullback-Leibler divergence. The experiment results demonstrate that the proposed primitive pattern-based method can allow one to semantically understand driver behaviors and driving styles.

A hidden semi-Markov model for indoor radio source localization using received signal strength

Multipath propagation makes the use of received signal strength (RSS) unreliable as a signal propagation model for localization of a radio source based on RSS data. An approach to mitigation of this problem is the use of a Hidden Markov model (HMM) to represent the relationship between RSS and the radio source location by incorporating an environment prior and RSS source dynamics. The HMM structure forces a geometric form for the distribution for the sojourn time. This, combined with missing data problems, reduces confidence in location estimation. It is found, in this paper, that Hidden semi-Markov Models (HsMMs), with a more flexible sojourn time distribution are more able to represent source dynamics while retaining the advantages of HMMs for environmental constraints and improving resilience to missing measurements. The performance of the proposed HsMM algorithm is compared with a standard HMM via experiments and simulations involving indoor radio source localization using RSS data. Simulations use a ray tracing software-based simulator and the experiments for transmitter localization with RSS data are collected by software defined radios.

A hidden semi-Markov model for estimating burst suppression EEG.

Burst suppression is an electroencephalogram (EEG) pattern associated with profoundly inactivated brain states characterized by cerebral metabolic depression. This pattern is distinguished by short-duration band-limited electrical activity (bursts) interspersed between relatively near-isoelectric periods (suppressions). Prior work in neurophysiology suggests that burst and suppression segments are respectively associated with consumption and regeneration of adenosine triphosphate resource in cortical networks. This indicates that once a suppression (or, burst) segment begins, the propensity to switch out of the state gradually increases with duration spent in the state. Prior EEG monitoring frameworks that track the brain state during burst suppression by tracking the estimated fraction of time spent in suppression, relative to bursts, do not incorporate this information. In this work, we incorporate this information within a hidden semi-Markov model (HSMM) wherein two states (burst & suppression) stochastically switch between each other using sojourn-time dependent transition probabilities. We demonstrate the HSMM's utility in analyzing clinical data by estimating the state probabilities, the optimal state sequence, and the brain's metabolic activation level characterized by parameters governing sojourn-time dependence in transition probabilities. The HSMM-based approach proposed here provides a novel statistical framework that advances the state-of-the-art in analyzing burst suppression EEG.

A Clinically Evaluated Interferometric Continuous-Wave Radar System for the Contactless Measurement of Human Vital Parameters.

Vital parameters are key indicators for the assessment of health. Conventional methods rely on direct contact with the patients’ skin and can hence cause discomfort and reduce autonomy. This article presents a bistatic 24 GHz radar system based on an interferometric six-port architecture and features a precision of 1 µm in distance measurements. Placed at a distance of 40 cm in front of the human chest, it detects vibrations containing respiratory movements, pulse waves and heart sounds. For the extraction of the respiration rate, time-domain approaches like autocorrelation, peaksearch and zero crossing rate are compared to the Fourier transform, while template matching and a hidden semi-Markov model are utilized for the detection of the heart rate from sphygmograms and heart sounds. A medical study with 30 healthy volunteers was conducted to collect 5.5 h of data, where impedance cardiogram and electrocardiogram were used as gold standard for synchronously recording respiration and heart rate, respectively. A low root mean square error for the breathing rate (0.828 BrPM) and a high overall F1 score for heartbeat detection (93.14%) could be achieved using the proposed radar system and signal processing.

A Subject-Driven Unsupervised Hidden Semi-Markov Model and Gaussian Mixture Model for Heart Sound Segmentation

The analysis of heart sounds is a challenging task, due to the quick temporal onset between successive events and the fact that an important fraction of the information carried by phonocardiogram (PCG) signals lies in the inaudible part of the human spectrum. For these reasons, computer-aided analysis of the PCG can dramatically improve the quantity of information recovered from such signals. In this paper, a hidden semi-Markov model (HSMM) is used to automatically segment PCG signals. In the proposed models, the emission probability distributions are approximated via Gaussian mixture model (GMM) priors. The choice of GMM emission probability distributions allow to apply re-estimation routines to automatically adjust the HSMM emission probability distributions to each subject. Building on the proposed method for fine tuning emission distributions, a novel subject-driven unsupervised heart sound segmentation algorithm is proposed and validated over the publicly available PhysioNet dataset. Perhaps surprisingly, the proposed unsupervised method achieved results in line with state-of-the-art supervised approaches, when applied to long heart sounds.

Returns spillovers between tourism ETFs

This paper investigates the relationship between tourism-related markets and equity, consumer staples equities and consumer discretionary equities by means of spillovers and volatility transmission. Relying on the recently introduced Exchange Traded Founds (ETFs), this study is the first to analyze return spillovers derived from an E-GARCH model and to take into account frequency dynamics to understand changes in connectedness across periods of time. The results uncover numerous channels of return transmission across the selected ETF markets over the last 13 years and highlight the role of equity ETFs as the most influential market in the sample. Furthermore, the study provides insights into the characteristics of tourism-related markets using a hidden semi-Markov model. Finally, we prove that tourism-related markets (except casino and gaming) have gained importance as investment assets over the last few years.

Modelling noise and pulse width modulation interference in indoor visible light communication channels

Visible light communication (VLC) has the potential to supplement the growing demand for wireless connectivity. In order to achieve the full potential of VLC, channel models are required. Discrete channel models based on semi-hidden Markov models (Fritchman model) for indoor VLC using low data rate LEDs are presented. Each channel considered includes background noise and differing types of interference from fluorescent lights and pulse-width modulated (PWM) LEDs, which could be part of an indoor smart lighting system. Models were developed based on experimental error sequences from a VLC system using an on-off keying (OOK) modulation scheme. The error sequences were input into the Baum-Welch algorithm to determine the model parameters by expectation maximisation. Simulated error sequences generated by the models are compared to and, in most cases, perform better than simpler models with a single bit error rate. The models closely approximate the experimental errors sequences in terms of error distribution. The models performed better in channels where there is less interference. It was also found that periodic errors were introduced as a results of the PWM modulated smart lighting LEDs. These models have use for designing error control codes and simulating indoor VLC environments with different types of interference.

An application of a semi-hidden Markov model in wireless communication systems

Stochastic processes are approved presentation of real systems which its development in space or time can be supposed as random. A semi-hidden Markov model as a type of stochastic processes is a modification of hidden Markov models with states that are no longer totally unobservable and are less hidden. This mathematical model is employed for modeling data sequences with long runs, memory and statistical inertia. In this article, we investigate the theory of the semi-hidden Markov model along with its parameter estimation and order estimation methods. Moreover, the proposed model is applied to model the error traces generated by the wireless channels. A new Markov-based trace analysis algorithm is suggested to divide a non-stationary network error trace into stationary parts. By means of the best semi-hidden Markov model and fitting probability distribution, we would be able to model these parts accurately. Calculating the information measure criteria and the autocorrelation function by running the modified Baum–Welch algorithm several times help us to find the optimal order of the semi-hidden Markov model.

Improved state change estimation in dynamic functional connectivity using hidden semi-Markov models

The study of functional brain networks has grown rapidly over the past decade. While most functional connectivity (FC) analyses estimate one static network structure for the entire length of the functional magnetic resonance imaging (fMRI) time series, recently there has been increased interest in studying time-varying changes in FC. Hidden Markov models (HMMs) have proven to be a useful modeling approach for discovering repeating graphs of interacting brain regions (brain states). However, a limitation lies in HMMs assuming that the sojourn time, the number of consecutive time points in a state, is geometrically distributed. This may encourage inaccurate estimation of the time spent in a state before switching to another state. We propose a hidden semi-Markov model (HSMM) approach for inferring time-varying brain networks from fMRI data, which explicitly models the sojourn distribution. Specifically, we propose using HSMMs to find each subject's most probable series of network states and the graphs associated with each state, while properly estimating and modeling the sojourn distribution for each state. We perform a simulation study, as well as an analysis on both task-based fMRI data from an anxiety-inducing experiment and resting-state fMRI data from the Human Connectome Project. Our results demonstrate the importance of model choice when estimating sojourn times and reveal their potential for understanding healthy and diseased brain mechanisms.

Detection of abnormal behaviour for dementia sufferers using Convolutional Neural Networks.

In recent years, there is a rapid increase in the population of elderly people. However, elderly people may suffer from the consequences of cognitive decline, which is a mental health disorder that primarily affects cognitive abilities such as learning, memory, etc. As a result, the elderly people may get dependent on caregivers to complete daily life tasks. Detecting the early indicators of dementia before it gets worsen and warning the caregivers and medical doctors would be helpful for further diagnosis. In this paper, the problem of activity recognition and abnormal behaviour detection is investigated for elderly people with dementia. First of all, the paper presents a methodology for generating synthetic data reflecting on some behavioural difficulties of people with dementia given the difficulty of obtaining real-world data. Secondly, the paper explores Convolutional Neural Networks (CNNs) to model patterns in activity sequences and detect abnormal behaviour related to dementia. Activity recognition is considered as a sequence labelling problem, while abnormal behaviour is flagged based on the deviation from normal patterns. Moreover, the performance of CNNs is compared against the state-of-art methods such as Naïve Bayes (NB), Hidden Markov Models (HMMs), Hidden Semi-Markov Models (HSMM), Conditional Random Fields (CRFs). The results obtained indicate that CNNs are competitive with those state-of-art methods.