ALGORITHMIC DESIGN AND SOFTWARE FOR A MICROCONTROLLER-BASED WEARABLE BIOMEDICAL SENSOR MONITORING SYSTEM VIA ESP-NOW PROTOCOL

IntroductionPhysics simulation has emerged as a promising approach to generate virtual Inertial Measurement Unit (IMU) data, offering a solution to reduce the extensive cost and effort of real-world data collection. However, the fidelity of virtual IMU depends heavily on the quality of the source motion data, which varies with motion capture setups. We hypothesize that improving virtual IMU fidelity is crucial to fully harness the potential of physics simulation for virtual IMU data generation in training Human Activity Recognition (HAR) models.MethodTo investigate this, we introduce WIMUSim, a 6-axis wearable IMU simulation framework designed to accurately parameterize real IMU properties when deployed on people. WIMUSim models IMUs in wearable sensing using four key parameters: Body (skeletal model), Dynamics (movement patterns), Placement (device positioning), and Hardware (IMU characteristics). Using these parameters, WIMUSim simulates virtual IMU through differentiable vector manipulations and quaternion rotations. A key novelty enabled by this approach is the identification of WIMUSim parameters using recorded real IMU data through gradient descent-based optimization, starting from an initial estimate. This process enhances the fidelity of the virtual IMU by optimizing the parameters to closely mimic the recorded IMU data. Adjusting these identified parameters allows us to introduce physically plausible variabilities.ResultsOur fidelity assessment demonstrates that WIMUSim accurately replicates real IMU data with optimized parameters and realistically simulates changes in sensor placement. Evaluations using exercise and locomotion activity datasets confirm that models trained with optimized virtual IMU data perform comparably to those trained with real IMU data. Moreover, we demonstrate the use of WIMUSim for data augmentation through two approaches: Comprehensive Parameter Mixing, which enhances data diversity by varying parameter combinations across subjects, outperforming models trained with real and non-optimized virtual IMU data by 4–10 percentage points (pp); and Personalized Dataset Generation, which customizes augmented datasets to individual user profiles, resulting in average accuracy improvements of 4 pp, with gains exceeding 10 pp for certain subjects.DiscussionThese results underscore the benefit of high-fidelity virtual IMU data and WIMUSim's utility in developing effective data generation strategies, alleviating the challenge of data scarcity in sensor-based HAR.

Background:Quantification of spinal mobility in chronic axial spondyloarthritis (axSpA) is an important component of disease assessment and is currently assessed using the clinical Bath Ankylosing Spondylitis Metrology Index (BASMI). Functional...

Many methods have been proposed for human activity classification, which rely either on Inertial Measurement Unit (IMU) data or data from static cameras watching subjects. There have been relatively less work using egocentric videos, and even fewer approaches combining egocentric video and IMU data. Systems relying only on IMU data are limited in the complexity of the activities that they can detect. In this paper, we present a robust and autonomous method, for fine-grained activity classification, that leverages data from multiple wearable sensor modalities to differentiate between activities, which are similar in nature, with a level of accuracy that would be impossible by each sensor alone. We use both egocentric videos and IMU sensors on the body. We employ Capsule Networks together with Convolutional Long Short Term Memory (LSTM) to analyze egocentric videos, and an LSTM framework to analyze IMU data, and capture temporal aspect of actions. We performed experiments on the CMU-MMAC dataset achieving overall recall and precision rates of 85.8% and 86.2%, respectively. We also present results of using each sensor modality alone, which show that the proposed approach provides 19.47% and 39.34% increase in accuracy compared to using only ego-vision data and only IMU data, respectively.

Multi-layer satellite network has become a hot spot for its wider coverage and higher bandwidth level. However, due to the frequent link changes and complexity of network, it is hard to find out a mechanism to handle well on long delay and high packet loss level. This paper proposes an optimized OSPF protocol called OOWLP to eliminate unnecessary routing convergence to optimize the packet loss level and delay ultimately. Link plan table, which records link contacting plan, will be used to update the link state database periodically so that we can eliminate the flooding procedure caused by scheduled link changes. On the other hand, Constrained Shortest Path First (CSPF) will be used to get business differentiated routes in multi-layer satellite network to optimized the throughput capacity in congestion scenario. We divide the sending packets into different businesses and get the routes for each business with longer duration limited by remaining bandwidth. Simulation results show that in normal scenario, average packet loss rate and delay performance are improved 17.42%, 51.44ms respectively, average packet loss rate and throughput capacity performance are optimized 79.05%, 9.81Mbps respectively in congestion scenario compared to standard OSPF. As a result, the proposed mechanism is able to shorten the average delay and lower the packet loss level in multi-layer satellite network.

Freezing of gait (FOG) is a debilitating symptom of Parkinson's disease (PD), characterized by an absence or reduction in forward movement of the legs despite the intention to walk. Detecting FOG during free-living conditions presents significant challenges, particularly when using only inertial measurement unit (IMU) data, as it must be distinguished from voluntary stopping events that also feature reduced forward movement. Influences from stress and anxiety, measurable through galvanic skin response (GSR) and electrocardiogram (ECG), may assist in distinguishing FOG from normal gait and stopping. However, no study has investigated the fusion of IMU, GSR, and ECG for FOG detection. Therefore, this study introduced two methods: a two-step approach that first identified reduced forward movement segments using a Transformer-based model with IMU data, followed by an XGBoost model classifying these segments as FOG or stopping using IMU, GSR, and ECG features; and an end-to-end approach employing a multi-stage temporal convolutional network to directly classify FOG and stopping segments from IMU, GSR, and ECG data. Results showed that the two-step approach with all data modalities achieved an average F1 score of 0.728 and F1@50 of 0.725, while the end-to-end approach scored 0.771 and 0.759, respectively. However, no significant difference was found compared to using only IMU data in both approaches (p-values: 0.466 to 0.887). In conclusion, adding physiological data did not provide a statistically significant benefit in distinguishing between FOG and stopping. The limitations may be specific to GSR and ECG data, and may not generalize to other physiological modalities.

Video streaming over wireless networks have many challenges due to the high error rate and burst packet error characteristic. Forward error correction (FEC) is a method commonly used to handle losses in real-time communication. Conventional FEC mechanisms provide redundancy by an averaged packet loss rate and performance decreases by burst packet losses. However, the average packet loss rate cannot give any indication of burst packet loss. Hence, the conventional FEC mechanisms cannot recover original source data over wireless networks. In this paper, we propose a burst-aware adaptive FEC (BAFEC) control mechanism to overcome burst packet losses. We will therefore be able to take account of average burst packet loss length. The sender can rely on this information in order to adjust the FEC redundancy. The experimental result shows that compared to the conventional FEC mechanisms, our proposed mechanism achieved better recovery performance in terms of packet loss rate and PSNR.

Video communications over wireless networks suffer various patterns of losses, including both random packet loss and burst losses. Previous error resilient techniques simply consider the average packet loss rate to enhance error robustness for video transmission. However, loss patterns, specifically burst losses, have great impact on video quality. In this paper, we propose a method that can take account of both random and burst losses to further improve the error resilience of video coding. Our method estimates the end-to-end distortion based on recursive optimal per-pixel estimate (ROPE) including both random and burst losses, and applies it for rate-distortion (RD)-based optimal mode selection. We apply our method in two cases: For single description video coding, we estimate the reconstructed pixel values for random packet loss and burst losses, and calculate the overall distortion. For multiple description video coding, we estimate the end-to-end distortion for multiple state video coding (MSVC) by considering the network conditions and multiple state recovery to reduce the error propagation due to packet loss in both descriptions for MSVC. Simulation results show that our proposed method achieves better performance than MSVC and original ROPE (only considering average packet loss rate) over wireless networks with burst losses.

Cross-layer design is a generic designation for a set of efficient adaptive transmission schemes, across multiple layers of the protocol stack, that are aimed at enhancing the spectral efficiency and increasing the transmission reliability of wireless communication systems. In this paper, one such cross-layer design scheme that combines physical layer adaptive modulation and coding (AMC) with link layer truncated automatic repeat request (T-ARQ) is proposed for multiple-input multiple-output (MIMO) systems employing orthogonal space--time block coding (OSTBC). The performance of the proposed cross-layer design is evaluated in terms of achievable average spectral efficiency (ASE), average packet loss rate (PLR) and outage probability, for which analytical expressions are derived, considering transmission over two types of MIMO fading channels, namely, spatially correlated Nakagami-m fading channels and keyhole Nakagami-m fading channels. Furthermore, the effects of the maximum number of ARQ retransmissions, numbers of transmit and receive antennas, Nakagami fading parameter and spatial correlation parameters, are studied and discussed based on numerical results and comparisons. Copyright © 2009 John Wiley & Sons, Ltd. A cross-layer design combining physical-layer adaptive modulation and link-layer truncated automatic repeat request protocol is proposed for multiple-input multiple-output (MIMO) systems, the performance of which is evaluated in terms of average spectral efficiency, average packet loss rate, and outage probability, in Nakagami-m fading channels that are either spatially correlated or exhibit the keyhole phenomenon. Owing to this framework, the impact of important design parameters, such as maximum number of retransmissions, MIMO configuration, Nakagami-m index, and spatial fading correlation, are revealed. Part of the material in this paper has been presented at IEEE Globecom'04, Dallas, TX, U.S.A., November–December 2004.

There has been significant amount of research work on human activity classification relying either on Inertial Measurement Unit (IMU) data or data from static cameras providing a third-person view. There has been relatively less work using wearable cameras, providing first-person or egocentric view, and even fewer approaches combining egocentric video with IMU data. Using only IMU data limits the variety and complexity of the activities that can be detected. For instance, the sitting activity can be detected by IMU data, but it cannot be determined whether the subject has sat on a chair or a sofa, or where the subject is. To perform fine-grained activity classification, and to distinguish between activities that cannot be differentiated by only IMU data, we present an autonomous and robust method using data from both wearable cameras and IMUs. In contrast to convolutional neural network-based approaches, we propose to employ capsule networks to obtain features from egocentric video data. Moreover, Convolutional Long Short Term Memory framework is employed both on egocentric videos and IMU data to capture the temporal aspect of actions. We also propose a genetic algorithm-based approach to autonomously and systematically set various network parameters, rather than using manual settings. Experiments have been conducted to perform 9- and 26-label activity classification, and the proposed method, using autonomously set network parameters, has provided very promising results, achieving overall accuracies of 86.6% and 77.2%, respectively. The proposed approach, combining both modalities, also provides increased accuracy compared to using only egovision data and only IMU data.

Inverse Perspective Mapping (IPM) is a technique used in Intelligent Transportation Systems to generate virtual bird’s eye view (BEV) images, enabling accurate obstacle detection and free space estimation. The IPM approach used before exclusively prioritized straight and flat road surface. If the IPM approach is used on an uneven road, this becomes exceedingly challenging. As a result, the IPM approach does not fully use the Inertial Measurement Unit (IMU) sensor. As an outcome, there will be a temporal delay in the data when image and IMU data are combined. The goal of this study is to present an IPM-based technique for stabilizing the exact position of the vehicle using angular orientation parameters derived from images and IMU data that are synchronized and filtered using the Kalman and complementary filters. Experimental results show that IPM can determine the actual angle when passing the uneven roads at 94% accuracy, surpassing previous IPM techniques. It is obvious that the proposed method can generate reliable and accurate IPM images. The image data and IMU data that have been recorded can be utilized for analysis to choose the parameters that are most appropriate to be employed throughout the IPM process, despite the fact that the road structure is continually changing.

Multi-hop relay can effectively improve the average packet loss rate (PLR) of vehicular ad hoc networks within a particular zone of interest. Challenges arise from analyzing the average PLR affected by distributed relay selections, which depend on the mobility of vehicles, the wireless channel conditions, and media access control (MAC). In this letter, we propose an average PLR analysis model taking into account the above three factors. However, the closed-form expression for the average PLR is intractable mainly due to the multiple integral of the joint distance distribution integrating with the channel conditions and vehicle mobility. An explicit expression for the upper bound of the average PLR is obtained by using Taylor series expansion, Holder's inequality, and the relay probability relaxation, which can facilitate the selection of the parameters at the physical and MAC layers for a better PLR. Simulation results validate our analyses.

Average packet loss rate (PLR) or average mean opinion score (MOS) are often used performance indicators for voice communications. However, even for a fixed average PLR, the delivered voice quality depends on the location of the packet losses as well as the distribution of packet losses due to the differences in voice codecs, their packet loss concealment schemes, the difficulty of concealing packet losses in unvoiced- to-voiced and voiced-to-unvoiced transition regions, and the difficulty of concealing successive packet losses. The result is that there is a distribution of achieved MOS values for a fixed average PLR and that the average MOS value may not capture the voice communications performance. Using the PESQ-MOS, we study the distribution of MOS values for wireless voice communications over additive white Gaussian noise (AWGN) and multipath fading channels using G.711 and G.729 voice codecs and their packet loss concealment schemes. Based on the distribution of PESQ-MOS, we define a quality indicator referred to as the MOSx, which is the MOS value that a user can expect to achieve or exceed x% of the time, where MOS and x are chosen to correspond to an acceptable voice outage rate. The MOSx is then used to evaluate the performance of G.711 and G.729 codecs over frequency selective fading and AWGN channels for voice over IEEE 802.11 a wireless LANs.

There has been a proliferation of smartphones, smart watches, and wearable sensors, making them ubiquitous in our daily lives. Mobile sensors have found widespread use due to their ever-decreasing cost, ease of deployment, and ability to provide continuous monitoring as opposed to sensors installed at fixed locations. Various techniques have been proposed for fall detection, gait analysis, activity monitoring, and heart rate and sleep sensing by wearable sensors and mobile phones. Compared to works that use inertial measurement unit (IMU) data or static cameras installed in the environment, there has been relatively less work using egocentric videos, meaning providing the first-person view from wearable cameras. Moreover, most of the existing studies on egocentric videos are based on only one sensor modality, namely the camera. There have been even fewer approaches that combine egocentric video data with IMU data. In this chapter, we will describe three different applications using wearable cameras together with IMU data. First, we will present an overview of a fall detection system using wearable devices, e.g., smartphones and tablets, equipped with cameras and accelerometers. Since the portable device is worn by the subject, monitoring is not limited to confined areas, and extends to wherever the subject may travel, as opposed to static sensors installed in certain rooms. Second, we will present an autonomous and robust method for counting footsteps, and tracking and calculating stride length by using both accelerometer and camera data from smartphones or Google™ glass. To provide higher precision, instead of using a preset stride length, the proposed method calculates the distance traveled with each step by using the camera data. This method is compared with the commercially available accelerometer-based step counter apps. The results show that the proposed method provides a significant increase in accuracy, and has the lowest average error rate both in number of steps taken and the distance traveled. Finally, we will provide an overview of a robust and autonomous method to detect activities with more details and context by using accelerometer and egocentric video data obtained from a smartphone.

The simultaneous localization and mapping (SLAM) method based on the RGB-D sensor is widely researched in recent years. However, the accuracy of the RGB-D SLAM relies heavily on correspondence feature points, and the position would be lost in case of scenes with sparse textures. Therefore, plenty of fusion methods using the RGB-D information and inertial measurement unit (IMU) data have investigated to improve the accuracy of SLAM system. However, these fusion methods usually do not take into account the size of matched feature points. The pose estimation calculated by RGB-D information may not be accurate while the number of correct matches is too few. Thus, considering the impact of matches in SLAM system and the problem of missing position in scenes with few textures, a loose fusion method combining RGB-D with IMU is proposed in this paper. In the proposed method, we design a loose fusion strategy based on the RGB-D camera information and IMU data, which is to utilize the IMU data for position estimation when the corresponding point matches are quite few. While there are a lot of matches, the RGB-D information is still used to estimate position. The final pose would be optimized by General Graph Optimization (g2o) framework to reduce error. The experimental results show that the proposed method is better than the RGB-D camera's method. And this method can continue working stably for indoor environment with sparse textures in the SLAM system.

Once a voice buffer is full, it remains full for a certain period, during which many packets are possibly blocked, resulting in consecutive clippings in voice. The packet loss rate during this period changes slowly and has large fluctuations. It is shown that the temporal behavior of packet loss, especially at high rate, is inherently determined by voice correlation and system capacity and is independent of buffer size. Buffering may reduce the occurrence of short blocking periods associated with low rates packet loss but does not affect long ones associated with high packet loss rates. In fact, increasing the buffer size merely extends nonblocking periods, and thereby reduces the overall average packet loss rate, but packet-loss performance within existing blocking periods is not significantly improved. A simple tool is developed for calculating the boundary performance. It is found that it is possible to design a packet-switched voice system without buffering only at the expense of supporting a fewer number of calls. The issue of voice delay allocation between source and network is discussed, and it is shown that it is more effective to keep the network delay short while extending the source delay.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>