Single Data Stream Research Articles

Multimodal measurements enhance insights into emotional responses to immediate feedback.

Adaptive learning technologies often provide students with immediate feedback on task performance. This feedback can elicit various emotional responses, which, in turn, influence learning. Most recent studies capture these emotions by single data streams, contradicting the multi-componential nature of emotion. Therefore, this study investigated 32 university students solving mathematical problems using an adaptive learning technology. Students received immediate feedback on every step in the solution process, after which their physiological, experiential and behavioral responses to this feedback were recorded. Physiological arousal was measured by electrodermal activity, valence was measured by self-reports (experiential), and emotion types were measured by observations of facial expressions (behavioral). Results showed more peaks in electrodermal activity after feedback than was expected based on chance. These responses were comparable in strength after feedback on failure and success. Students' experiential responses conveyed mostly positive valence after feedback on success and mostly negative valence after feedback on failure. Behavioral observations showed more negative than positive emotion types after feedback on failure and more positive than negative emotion types after feedback on success. These results show that physiological arousal is a valuable objective indicator of emotional responses after immediate feedback but should be accompanied by other data streams in order to understand students' emotional responses. Both valence and emotion types can be used for this purpose. These outcomes pave the way for designing adaptive learning technologies that take students' emotions into account.

Inferring community transmission of SARS-CoV-2 in the United Kingdom using the ONS COVID-19 Infection Survey

Key epidemiological parameters, including the effective reproduction number, R(t), and the instantaneous growth rate, r(t), generated from an ensemble of models, have been informing public health policy throughout the COVID-19 pandemic in the four nations of the United Kingdom of Great Britain and Northern Ireland (UK). However, estimation of these quantities became challenging with the scaling down of surveillance systems as part of the transition from the “emergency” to “endemic” phase of the pandemic.The Office for National Statistics (ONS) COVID-19 Infection Survey (CIS) provided an opportunity to continue estimating these parameters in the absence of other data streams. We used a penalised spline model fitted to the publicly-available ONS CIS test positivity estimates to produce a smoothed estimate of the prevalence of SARS-CoV-2 positivity over time. The resulting fitted curve was used to estimate the “ONS-based” R(t) and r(t) across the four nations of the UK. Estimates produced under this model are compared to government-published estimates with particular consideration given to the contribution that this single data stream can offer in the estimation of these parameters.Depending on the nation and parameter, we found that up to 77% of the variance in the government-published estimates can be explained by the ONS-based estimates, demonstrating the value of this singular data stream to track the epidemic in each of the four nations. We additionally find that the ONS-based estimates uncover epidemic trends earlier than the corresponding government-published estimates.Our work shows that the ONS CIS can be used to generate key COVID-19 epidemiological parameters across the four UK nations, further underlining the enormous value of such population-level studies of infection. This is not intended as an alternative to ensemble modelling, rather it is intended as a potential solution to the aforementioned challenge faced by public health officials in the UK in early 2022.

Multi-Stream Concept Drift Self-Adaptation Using Graph Neural Network

Concept drift is the phenomenon where the data distribution in a data stream changes over time. It is a ubiquitous problem in the real-world, for example, a traffic accident would cause a jam in a certain period, leading to a distribution change in traffic speed. Most research in the concept drift field focuses on single data stream, however, few of them consider multi-stream environments which are more in line with the application needs. To fill this gap, we propose a multi-stream prediction setting and a multi-stream concept drift self-adaptation framework using graph neural network, named SAGN. In SAGN, we reconsider the learning procedure of GNN-based predictors from an aspect of concept drift adaptation for multi-stream. By this design, the prediction task is converted into online streaming data tasks in sub-graphs. Each sub-graph corresponds to an adaptation target and will be updated over time. In this way, locally we can overcome drift in each sub-graph by a designed adaptation technique, and globally the correlation between different data streams is well-preserved as a graph structure. Therefore, whether drift occurs or not, in one or several streams, SAGNcan provide consistently accurate prediction results. We comprehensively tested SAGNon both synthetic and real-world, drift and non-drift data in the multi-step prediction task. The experiment results show that SAGNis able to achieve state-of-the-art performance in most cases.

Distributed interferometric fiber tip biosensors for a multi-channel and label-free biomolecular interaction analysis.

This paper describes fabrication and implementation of distributed optical fiber tip biosensor probes for simultaneously measuring label-free biomolecular interactions at multiple locations. Biosensor probes at the tip of a single-mode fiber are Fabry-Perot etalons that are functionalized with a capture layer for a specific biomolecule. A coherence multiplexing method is implemented to separate data collected from distributed biosensors in a single data stream. Multiplexing is achieved by using fiber tip biosensors of varying etalon lengths with the same or different capture layers for each biosensing channel. Experiments demonstrating simultaneous multi-channel recording of protein-to-protein interaction sensorgrams with fiber tip biosensor probes are presented.

Wireless Sensor Networks Fault Detection and Identification

We have developed and experimentally tested a set of models for the detection and identification of sensor faults that commonly occur in wireless sensor networks. Considered faults include outlier, spike, variance, high-frequency noise, offset, gain, and drift faults. These faults affect the system operations and can endanger operators, final users, and the general public. The fault detection models are divided into two classes: data-centric models, which only analyze a single data stream, and system-centric models, which consider the overall system. For data-centric models, we use the magnitude, the gradient, and the variance of raw sensor data to model faults. For system-centric models, we introduce variogram-based techniques that allow faults to be detected by comparing readings from multiple sensors that measure related phenomena. For data-centric and system-centric sensor fault detection, we show how a few model parameters affect the sensitivity of wireless sensor network fault models. We present simulation and experimental results that illustrate the fault detection and identification models. The system is intended for health monitoring applications of the NASA Stennis Space Center (SSC) test stands and widely distributed support systems, including pressurized gas lines, propellant delivery systems, and water coolant lines. The testbed consists of Coremicro® reconfigurable embedded smart sensor nodes [29] capable of wireless communication, a network-capable application processor, a wireless base station, the software that supports sensor and actuator health monitoring, a database server, and a smartphone running a health monitoring Android application.

A generic sketch for estimating super-spreaders and per-flow cardinality distribution in high-speed data streams

For a high-speed network, it is an important task to process the IP packet stream using limited memory and measure its statistical metrics of interest. While many algorithms have been proposed to estimate the cardinality of a single data stream (i.e., the number of distinct elements), it remains a great challenge when a stream contains numerous sub-streams, called flows. In this paper, we focus on a problem of designing a generic data structure to measure multiple types of per-flow statistics in a high-speed stream, including per-flow cardinality, top-K super-spreading flows with the greatest cardinalities, per-flow cardinality moments and per-flow cardinality distribution. Previous solutions for generic measurement mainly focus on the frequency-related statistics measurement, while this paper makes a step forward to support deduplication, i.e., cardinality-related measuring. To address this new problem, we propose a generic sketch named M2D. The challenge is that the per-flow cardinality distribution is often highly skewed with a small proportion of super-spreaders. To tame the skewness, we adopt the adjustable progressive sampling technique, which samples subsets of flows by an exponentially decreasing probability according to their cardinalities. Based on the sampled super-spreaders, we estimate the moments of per-flow cardinalities with different orders. We finally apply the method of moments to reconstruct the per-flow cardinality distribution with no priori knowledge about its formula. Our experiments show M2D’s high memory efficiency (average savings of 38%) and satisfactory distribution estimation accuracy (2% to 98% improvement) than other algorithms.

Beam-Space MIMO Radar for Joint Communication and Sensing With OTFS Modulation

Motivated by automotive applications, we consider joint radar sensing and data communication for a system operating at millimeter wave (mmWave) frequency bands, where a Base Station (BS) is equipped with a co-located radar receiver and sends data using the Orthogonal Time Frequency Space (OTFS) modulation format. We consider two distinct modes of operation. In <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Discovery</i> mode, a single common data stream is broadcast over a wide angular sector. The radar receiver must detect the presence of not yet acquired targets and perform coarse estimation of their parameters (angle of arrival, range, and velocity). In <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Tracking</i> mode, the BS transmits multiple individual data streams to already acquired users via beamforming, while the radar receiver performs accurate estimation of the aforementioned parameters. Due to hardware complexity and power consumption constraints, we consider a hybrid digital-analog architecture where the number of RF chains and A/D converters is significantly smaller than the number of antenna array elements. In this case, a direct application of the conventional MIMO radar approach is not possible. Consequently, we advocate a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">beam-space approach</i> where the vector observation at the radar receiver is obtained through a RF-domain beamforming matrix operating the dimensionality reduction from antennas to RF chains. Under this setup, we propose a likelihood function-based scheme to perform joint target detection and parameter estimation in Discovery, and high-resolution parameter estimation in Tracking mode, respectively. Our numerical results demonstrate that the proposed approach is able to reliably detect multiple targets while closely approaching the Cramér-Rao Lower Bound (CRLB) of the corresponding parameter estimation problem.

Occupancy estimation with environmental sensors: The possibilities and limitations

Occupancy detection and estimation in buildings paves the way to improve the utilization of lighting and HVAC systems, induce energy savings and enhance the well-being of the occupants. This paper presents a comparative study of state-of-art machine learning techniques that solve two different occupancy monitoring problems using environmental sensor data. One is the regression problem that estimates the actual count of occupants while the other is the classification problem which estimates the level of occupancy (empty, sparse, full). The results of the best performing machine learning techniques that solve both problems for the open dataset from the University of Southern Denmark, Odense are presented to compare the accuracy of both approaches and the ease of implementation. The impact of CO2, temperature, and humidity features on the occupancy count/levels and detection accuracy (occupied versus unoccupied) are studied. Comprehensive analysis with different combinations of environmental features and other free features such as time-of-day along with different sampling techniques for training and testing are performed to understand how such models can be adapted for actual deployment. Our results indicate detection accuracy between 66% to 82% for different sampling schemes; with day-based sampling showing a better performance while random sampling generically showcasing lower accuracy (66.2%). The occupancy estimation (levels or counts) has accuracy in the range of 69% to 79% for random sampling and 71% to 80% for day-based sampling. Finally, results demonstrate that models based single environmental sensor data streams do not perform as well as the models with sensor fusion.

Double-Anonymous Sketch: Achieving Top-K-fairness for Finding Global Top-K Frequent Items

Finding top-K frequent items has been a hot topic in data stream processing in recent years, which has a wide range of applications. However, most of existing sketch algorithms focuses on finding local top-K in a single data stream. In this paper, we work on finding global top-K in multiple disjoint data streams. We find that directly deploying prior sketch algorithms is often unfair under global scenarios, which will degrade the accuracy of global top-K. We define top-K-fairness and show that it is important for finding global top-K. To achieve top-K-fairness, we propose a new sketch framework, called the Double-Anonymous sketch. The process of finding global top-K items is similar to that of paper reviewing and democratic elections. In these scenarios, double-anonymity is often an effective strategy to achieve top-K-fairness. We also propose two techniques, hot panning, and early freezing, to further improve the accuracy. We theoretically prove that the Double-Anonymous sketch achieves top-K-fairnesswhile keeping high accuracy. We perform extensive experiments to verify top-K-fairness in the scenario of disjoint data streams. The experimental results show that the Double-Anonymous sketch's error is up to 129 times (60 times on average) smaller than the state-of-the-art. All the related source code is open-sourced and available at Github.

Cross-stream contrastive learning for self-supervised skeleton-based action recognition

Self-supervised skeleton-based action recognition enjoys a rapid growth along with the development of contrastive learning. The existing methods rely on imposing invariance to augmentations of 3D skeleton within a single data stream, which merely leverages the easy positive pairs and limits the ability to explore the complicated movement patterns. In this paper, we advocate that the defect of single-stream contrast and the lack of necessary feature transformation are responsible for easy positives, and therefore propose a Cross-Stream Contrastive Learning framework for skeleton-based action Representation learning (CSCLR). Specifically, the proposed CSCLR not only utilizes intra-stream contrast pairs, but introduces inter-stream contrast pairs as hard samples to formulate a better representation learning. Besides, to further exploit the potential of positive pairs and increase the robustness of self-supervised representation learning, we propose a Positive Feature Transformation (PFT) strategy which adopts feature-level manipulation to increase the variance of positive pairs. To validate the effectiveness of our method, we conduct extensive experiments on three benchmark datasets NTU-RGB + D 60, NTU-RGB + D 120 and PKU-MMD. Experimental results show that our proposed CSCLR exceeds the state-of-the-art methods on a diverse range of evaluation protocols.

Development, testing, parameterisation, and calibration of a human PBPK model for the plasticiser, di-(2-ethylhexyl) terephthalate (DEHTP) using in silico, in vitro and human biomonitoring data.

A physiologically based pharmacokinetic model for di-(2-ethylhexyl) terephthalate (DEHTP) based on a refined model for di-(2-propylheptyl) phthalate (DPHP) was developed to interpret the metabolism and biokinetics of DEHTP following a single oral dose of 50mg to three male volunteers. In vitro and in silico methods were used to generate parameters for the model. For example, measured intrinsic hepatic clearance scaled from in vitro to in vivo and plasma unbound fraction and tissue:blood partition coefficients (PCs) were predicted algorithmically. Whereas the development and calibration of the DPHP model was based upon two data streams, blood concentrations of parent chemical and first metabolite and the urinary excretion of metabolites, the model for DEHTP was calibrated against a single data stream, the urinary excretion of metabolites. Despite the model form and structure being identical significant quantitative differences in lymphatic uptake between the models were observed. In contrast to DPHP the fraction of ingested DEHTP entering lymphatic circulation was much greater and of a similar magnitude to that entering the liver with evidence for the dual uptake mechanisms discernible in the urinary excretion data. Further, the absolute amounts absorbed by the study participants, were much higher for DEHTP relative to DPHP. The in silico algorithm for predicting protein binding performed poorly with an error of more than two orders of magnitude. The extent of plasma protein binding has important implications for the persistence of parent chemical in venous blood-inferences on the behaviour of this class of highly lipophilic chemicals, based on calculations of chemical properties, should be made with extreme caution. Attempting read across for this class of highly lipophilic chemicals should be undertaken with caution since basic adjustments to PCs and metabolism parameters would be insufficient, even when the structure of the model itself is appropriate. Therefore, validation of a model parameterized entirely with in vitro and in silico derived parameters would need to be calibrated against several human biomonitoring data streams to constitute a data rich source chemical to afford confidence for future evaluations of other similar chemicals using the read-across approach.

Fully-Integrated SPAD-Based Receiver With Nanosecond Dead Time for Optical Wireless Communication

This paper presents the design and characterization of a fully-integrated receiver based on single-photon avalanche diodes (SPADs) with nanosecond dead time for high-speed high-sensitivity optical wireless communication (OWC). The receiver consists of a 4x4 SPAD array that is based on a p-well/deep n-well (DNW) structure, and each SPAD is integrated with a tunable front-end circuit to perform quench and reset. In addition, an OR tree is designed to combine the 16 channels of output from the front-end circuits to generate a single data stream for signal processing, and an output buffer is implemented as the interface to drive 50-Ω loading for testing purpose. Fabricated in a standard 180 nm CMOS process, the receiver achieves a minimum dead time of about 2.5 ns. Bit error rate (BER) measurement of the implemented receiver indicates a sensitivity of −31.6 dBm at 100 Mb/s for a BER of 2 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−3</sup> and a wavelength of 520 nm, where on-off keying (OOK) modulation and a 2 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">15</sup> -1 pseudorandom binary sequence (PRBS-15) are employed. To recover the transmitted data stream from the received signal, a signal processing flow specific for SPAD-based receivers is proposed and implemented.

Detecting group concept drift from multiple data streams

Concept drift may lead to a sharp downturn in the performance of streaming in data-based algorithms, caused by unforeseeable changes in the underlying distribution of data. In this paper, we are mainly concerned with concept drift across multiple data streams, and in situations where the drift of each data stream cannot be detected in time, due to slight underlying distribution drifts. We call this group concept drift. When compared to the detection of concept drift for a single data stream, the challenges of detecting group concept drift arise from three aspects: first, the training data become more complex; second, the underlying distribution becomes more complex; and third, the correlations between data streams become more complex. To address these challenges, the key idea of our method is to construct a distribution free test statistic, free from any underlying distribution in multiple data streams. Then, for streaming data, we design an online learning algorithm to obtain this test statistic, thereby determining the concept drift caused by the hypothesis test. The experiment evaluations with both synthetic and real-world datasets prove that our method can accurately detect concept drift from multiple data streams.

H-DAC: discriminative associative classification in data streams

In this paper, we propose an efficient and highly accurate method for data stream classification, called discriminative associative classification. We define class discriminative association rules (CDARs) as the class association rules (CARs) in one data stream that have higher support compared with the same rules in the rest of the data streams. Compared to associative classification mining in a single data stream, there are additional challenges in the discriminative associative classification mining in multiple data streams, as the Apriori property of the subset is not applicable. The proposed single-pass H-DAC algorithm is designed based on distinguishing features of the rules to improve classification accuracy and efficiency. Continuously arriving transactions are inserted at fast speed and large volume, and CDARs are discovered in the tilted-time window model. The data structures are dynamically adjusted in offline time intervals to reflect each rule supported in different periods. Empirical analysis shows the effectiveness of the proposed method in the large fast speed data streams. Good efficiency is achieved for batch processing of small and large datasets, plus 0–2% improvements in classification accuracy using the tilted-time window model (i.e., almost with zero overhead). These improvements are seen only for the first 32 incoming batches in the scale of our experiments and we expect better results as the data streams grow.

Double-coupling learning for multi-task data stream classification

Data stream classification methods exploiting cohesion in a single data stream have demonstrated promising performance. However, scenarios involving multiple data streams are indeed common in practice. They involve several correlated data streams and can be viewed as multi-task data streams. Instead of considering them separately as individual data streams, it is beneficial to leverage the correlations among the multi-task data streams in data stream modeling. In this regard, a novel classification method called double-coupling support vector machines (DC-SVM) is proposed to classify multiple data streams simultaneously, where both the external correlations between multiple data streams and the internal relationships within each individual data stream are considered. Experimental results on synthetic and real-world multi-task data streams show that the proposed method outperforms traditional data stream classification methods.

Transformer-Based Model for Electrical Load Forecasting

Amongst energy-related CO2 emissions, electricity is the largest single contributor, and with the proliferation of electric vehicles and other developments, energy use is expected to increase. Load forecasting is essential for combating these issues as it balances demand and production and contributes to energy management. Current state-of-the-art solutions such as recurrent neural networks (RNNs) and sequence-to-sequence algorithms (Seq2Seq) are highly accurate, but most studies examine them on a single data stream. On the other hand, in natural language processing (NLP), transformer architecture has become the dominant technique, outperforming RNN and Seq2Seq algorithms while also allowing parallelization. Consequently, this paper proposes a transformer-based architecture for load forecasting by modifying the NLP transformer workflow, adding N-space transformation, and designing a novel technique for handling contextual features. Moreover, in contrast to most load forecasting studies, we evaluate the proposed solution on different data streams under various forecasting horizons and input window lengths in order to ensure result reproducibility. Results show that the proposed approach successfully handles time series with contextual data and outperforms the state-of-the-art Seq2Seq models.

Process-and-Forward two-way relay using multiple space-time line codes

In this paper, we consider a two-way relay (TWR) network that two source nodes (SNs) with multiple antennas exchange data to each other via a relay node (RN) with multiple antennas, when full channel state information (CSI) is available at RN and no CSIs are available at SNs. In case that both SNs transmit space-time block coded multiple data streams to the RN, a new precoding method is proposed for the RN by combing block-wise processing and multiple space-time line codes (STLCs) requiring no CSI at the receiver except the channel norm. An optimal precoding matrix is derived that maximized the minimum achievable rate using a gradient-based iterative algorithm. Moreover, to reduce the design complexity, we derive the closed-form precoding matrix minimizing the total mean square error (MSE), and its theoretical performance is analyzed in terms of the per-stream MSE and detection signal-to-interference-plus-noise ratio. Through numerical simulations, the theoretical MSE is verified by comparing with an experimental MSE, and it is demonstrated that the proposed precoding method is more beneficial than existing retransmission methods for the RN when the SN transfers both single and multiple data stream(s).

Analysis of Wave Propagation Models With Radio Network Planning Using Dual Polarized MIMO Antenna for 5G Base Station Applications

Dual polarized printed multiple input multiple output (MIMO) antenna for Band 42 (3.4 - 3.6 GHz) with wave propagation models is presented. Polarization and spatial diversity are achieved by utilizing two printed bow-tie antennas in orthogonal orientation. The designed dual polarized antenna element with <inline-formula> <tex-math notation="LaTeX">$2\times 2$ </tex-math></inline-formula>, <inline-formula> <tex-math notation="LaTeX">$4\times 4$ </tex-math></inline-formula> and <inline-formula> <tex-math notation="LaTeX">$8\times 8$ </tex-math></inline-formula> massive MIMO antenna configuration radiation patterns are deployed in selected geographical situation for detailed radio network planning using FEKO-WinProp platform. Knife edge diffraction, extended walfisch-ikegami and dominant path wave propagation models are implemented with designed MIMO antenna configurations. Modulation schemes of QPSK and QAM with corresponding data rates and throughput for all propagation models are presented. The signal strength and quality reflecting parameters reference signal received power (RSRP), received signal strength indicator (RSSI), reference signal received quality (RSRQ), and signal to noise plus interference ratio (SNIR) are also evaluated for each model. From the simulation results dominant path model provides data rate and throughput of 3.827, 995 MBit/s and 3.577, 930.1 MBit/s for single stream of data in uplink and downlink respectively. The maximum data rate of 1.37 GBits/s is achieved for deployed base stations with <inline-formula> <tex-math notation="LaTeX">$8\times 8$ </tex-math></inline-formula> massive MIMO antenna configuration effectively covering the entire geographical site.

A hybrid model for multipoint real time potency observation in continuous direct compression manufacturing operations

The ongoing transition from batch to continuous manufacturing offers both challenges and opportunities in the field of oral solid dosage form production. In turn, Process Analytical Technology (PAT) offers a path towards the successful deployment of continuous tablet manufacturing in rotary tablet presses. One promising PAT tool for this endeavour is the NIR-derived potency measurement. However, the high degree of noise in the data may hamper the extraction of useful information. For this reason, this work focused on the implementation of an adaptive Kalman filter algorithm that incorporates and reconciles the potency prediction given by one or more NIR probes with those of a semi-mechanistic compartmental model developed for the application at hand. This approach allowed for more robust concentration estimations. Furthermore, it was observed that potency levels in multiple locations in the studied tablet press (including those in the finished tablets) could be appropriately inferred using a single in-line measurement data stream. This methodology thus opens the door to advanced process control applications.

Identifying latent behavioural states in animal movement with M4, a nonparametric Bayesian method

Abstract Understanding animal movement often relies upon telemetry and biologging devices. These data are frequently used to estimate latent behavioural states to help understand why animals move across the landscape. While there are a variety of methods that make behavioural inferences from biotelemetry data, some features of these methods (e.g. analysis of a single data stream, use of parametric distributions) may limit their generality to reliably discriminate among behavioural states. To address some of the limitations of existing behavioural state estimation models, we introduce a nonparametric Bayesian framework called the mixed‐membership method for movement (M4), which is available within the open‐source bayesmove R package. This framework can analyse multiple data streams (e.g. step length, turning angle, acceleration) without relying on parametric distributions, which may capture complex behaviours more successfully than current methods. We tested our Bayesian framework using simulated trajectories and compared model performance against two segmentation methods (behavioural change point analysis (BCPA) and segclust2d), one machine learning method [expectation‐maximization binary clustering (EMbC)] and one type of state‐space model [hidden Markov model (HMM)]. We also illustrated this Bayesian framework using movements of juvenile snail kites Rostrhamus sociabilis in Florida, USA. The Bayesian framework estimated breakpoints more accurately than the other segmentation methods for tracks of different lengths. Likewise, the Bayesian framework provided more accurate estimates of behaviour than the other state estimation methods when simulations were generated from less frequently considered distributions (e.g. truncated normal, beta, uniform). Three behavioural states were estimated from snail kite movements, which were labelled as ‘encamped’, ‘area‐restricted search’ and ‘transit’. Changes in these behaviours over time were associated with known dispersal events from the nest site, as well as movements to and from possible breeding locations. Our nonparametric Bayesian framework estimated behavioural states with comparable or superior accuracy compared to the other methods when step lengths and turning angles of simulations were generated from less frequently considered distributions. Since the most appropriate parametric distributions may not be obvious a priori, methods (such as M4) that are agnostic to the underlying distributions can provide powerful alternatives to address questions in movement ecology.