Search

In this paper, we propose a scheme for distributed source coding of correlated sources using a single systematic LDPC code. In particular, since we are interested in wireless sensor network applications, we consider LDPC codes with short to moderate lengths that achieve every arbitrary coding rate on the Slepian-WoIf rate region. We simplify the distributed source coding problem to the rate-compatible LDPC code design with an unequal error protection property. The decoders communicate to each other to exchange information bits prior to decoding. However, thereafter, each performs the decoding independently. Therefore, errors in one decoder do not affect the other one. The simulation results confirm that the gap from the theoretical limit remains almost the same for different rates on the Slepian- Wolf rate region. First, we consider two correlated sources. We show that our proposed scheme improves the performance of distributed source coding of two sources considerably. This benefit is more stressed for application with short to moderate length sequences. Then, we study distributed source coding of three sources. As a special case, we investigate three sources that are pairwise correlated with the same correlation probability. We show that the gap from the theoretical limit is smaller than that of previous work. We also investigate the distributed source coding of correlated sources when there is no prior knowledge of the correlation parameter at the time of code design. We note that although the proposed distributed source coding is well suited for sensor networks (where sequences with less than 10000 bits are used), the method can be generalized to other distributed source coding applications. [ABSTRACT FROM AUTHOR]

This correspondence introduces two-dimensional (2-D) filter bank codes (TDFBCs). The synthesis bank of a two-channel two-dimensional filter bank over a finite field is used to design the 2-D code, and the corresponding analysis bank is used to generate the syndrome of the code. First, we study the encoder of half-rate TDFBCs and show that these linear codes are lattice-cyclic. It is proven that any 2-D lattice-cyclic code can also be generated by a 2-D filter bank. Second, we introduce a methodology to design TDFBCs over binary erasure channels. These codes have a simple and efficient maximum-likelihood (ML) decoding algorithm for burst erasures. We show that half-rate TDFBCs of dimensions N1 x N2 can recover burst erasures of size up to N1 x N2 /2 and N1 /2 x N2 using the proposed decoding technique. Finally, we present examples of TDFBCs that satisfy the Reiger bound with equality, i.e., they are capable of correcting any burst of size N1 N2 /2. [ABSTRACT FROM AUTHOR]

This paper introduces a procedure for constructing convolutional codes using finite-field wavelets. This provides novel insight into the study of convolutional codes and permits the design of the new convolutional codes that is not possible by conventional methods. Exploiting algebraic properties of the wavelet codes, we show that a rate K/L wavelet convolutional encoder is a basic encoder that is noncatastrophic. In addition, we prove that any rate 1/L wavelet convolutional encoder is minimal-basic and that every L-band orthogonal wavelet system generates a rate 1/L self-orthogonal code. As an application of wavelet convolutional codes, we construct time-varying convolutional codes. These codes have unique trellis structures that result in fast and low computational complexity decoding algorithms. As examples, we present some time-varying wavelet convolutional codes that can be decoded faster than comparable lime-invariant convolutional codes. We construct 16- and 32-state time-varying wavelet convolutional codes with minimum-free distances of seven and eight, respectively. These codes have the same minimum-free distance as the best time-invariant codes of the same rate and state complexity, but they can be decoded almost twice as fast. We show that a 32-state time-varying wavelet convolutional code is superior to the Lauer code in performance while having almost the same decoding complexity. Although the 32-state wavelet code is inferior to the 16-state Golay convolutional code as far as computational complexity, it outperforms this code in the performance. We also prove that orthogonal filterbanks generate self-dual time-varying codes. We give a design for doubly even self-dual time-varying convolutional codes by imposing some constraints on the filters that define the L-band orthogonal wavelets. As another application of wavelet convolutional codes, we propose a new scheme for generating rate-adaptive codes. These codes have the property that multiple rates of the code can be decoded on one trellis and its subtrellises. [ABSTRACT FROM AUTHOR]

High‐angle cameras are commonly used for trajectory data collection in transportation research. However, without refinement and validation, trajectory data obtained through video processing software may be unreliable, inaccurate, or incomplete. This paper focuses on a critical issue in the field of trajectory data acquisition and analysis—there is still no reliable and fully vetted trajectory dataset in the research community. The current practice for validating video‐based trajectory can be classified as indirect methods and direct methods. Indirect methods of trajectory validation use algorithms to efficiently correct data anomalies without human intervention but may overlook detailed driving behaviors, whereas direct methods involve meticulous manual verification to preserve data fidelity but are labor‐intensive and less scalable. The spatial‐temporal maps (STMaps) method offers an additional layer of verification to affirm the accuracy and reliability of trajectory data. To enhance the performance, the deep spatial‐temporal embedding model is proposed for trajectory instance segmentation on STMaps using the contrastive learning framework. The parity constraints at both pixel and instance levels guide the deep neural network to learn the embedding spaces that can be built on different backbone networks. The reconstructed Next Generation Simulation (NGSIM) highway dataset trajectory dataset is thoroughly validated against manually processed ground truth, and the error‐free NGSIM data are refined to be a reliable resource for transportation research based on car‐following behaviors, lane‐change frequency, consistency, and jerk value measurements. [ABSTRACT FROM AUTHOR]

Machine learning (ML) methods, particularly Reinforcement Learning (RL), have gained widespread attention for optimizing traffic signal control in intelligent transportation systems. However, existing ML approaches often exhibit limitations in scalability and adaptability, particularly within large traffic networks. This paper introduces an innovative solution by integrating decentralized graph-based multi-agent reinforcement learning (DGMARL) with a Digital Twin to enhance traffic signal optimization, targeting the reduction of traffic congestion and network-wide fuel consumption associated with vehicle stops and stop delays. In this approach, DGMARL agents are employed to learn traffic state patterns and make informed decisions regarding traffic signal control. The integration with a Digital Twin module further facilitates this process by simulating and replicating the real-time asymmetric traffic behaviors of a complex traffic network. The evaluation of this proposed methodology utilized PTV-Vissim, a traffic simulation software, which also serves as the simulation engine for the Digital Twin. The study focused on the Martin Luther King (MLK) Smart Corridor in Chattanooga, Tennessee, USA, by considering symmetric and asymmetric road layouts and traffic conditions. Comparative analysis against an actuated signal control baseline approach revealed significant improvements. Experiment results demonstrate a remarkable 55.38% reduction in Eco_PI, a developed performance measure capturing the cumulative impact of stops and penalized stop delays on fuel consumption, over a 24 h scenario. In a PM-peak-hour scenario, the average reduction in Eco_PI reached 38.94%, indicating the substantial improvement achieved in optimizing traffic flow and reducing fuel consumption during high-demand periods. These findings underscore the effectiveness of the integrated DGMARL and Digital Twin approach in optimizing traffic signals, contributing to a more sustainable and efficient traffic management system. [ABSTRACT FROM AUTHOR]

After short-term, acute-care hospitalization for stroke, patients may be discharged home or other facilities for continued medical or rehabilitative management. The site of postacute care affects overall mortality and functional outcomes. Determining discharge disposition is a complex decision by the healthcare team. Early prediction of discharge destination can optimize poststroke care and improve outcomes. Previous attempts to predict discharge disposition outcome after stroke have limited clinical validations. In this study, readmission status was used as a measure of the clinical significance and effectiveness of a discharge disposition prediction. Low readmission rates indicate proper and thorough care with appropriate discharge disposition. We used Medicare beneficiary data taken from a subset of base claims in the years of 2014 and 2015 in our analyses. A predictive tool was created to determine discharge disposition based on risk scores derived from the coefficients of multivariable logistic regression related to an adjusted odds ratio. The top five risk scores were admission from a skilled nursing facility, acute heart attack, intracerebral hemorrhage, admission from "other" source, and an age of 75 or older. Validation of the predictive tool was accomplished using the readmission rates. A 75% probability for facility discharge corresponded with a risk score of greater than 9. The prediction was then compared to actual discharge disposition. Each cohort was further analyzed to determine how many readmissions occurred in each group. Of the actual home discharges, 95.7% were predicted to be there. However, only 47.8% of predictions for home discharge were actually discharged home. Predicted discharge to facility had 15.9% match to the actual facility discharge. The scenario of actual discharge home and predicted discharge to facility showed that 186 patients were readmitted. Following the algorithm in this scenario would have recommended continued medical management of these patients, potentially preventing these readmissions. [ABSTRACT FROM AUTHOR]

Objectives: Early determination of hospital discharge disposition status at an acute admission is extremely important for stroke management and the eventual outcomes of patients with stroke. We investigated the hospital discharge disposition of patients with stroke residing in Tennessee and developed a predictive tool for clinical adoption. Our investigational aims were to evaluate the association of selected patient characteristics with hospital discharge disposition status and predict such status at the time of an acute stroke admission.Methods: We analyzed 127,581 records of patients with stroke hospitalized between 2010 and 2014. Logistic regression was used to generate odds ratios with 95% confidence intervals to examine the factor outcome association. An easy-to-use clinical predictive tool was built by using integer-based risk scores derived from coefficients of multivariable logistic regression.Results: Among the 127,581 records of patients with stroke, 86,114 (67.5%) indicated home discharge and 41,467 (32.5%) corresponded to facility discharge. All considered patient characteristics had significant correlations with hospital discharge disposition status. Patients were at greater odds of being discharged to another facility if they were women; older; black; patients with a subarachnoid or intracerebral hemorrhage; those with the comorbidities of diabetes mellitus, heart disease, hypertension, chronic kidney disease, arrhythmia, or depression; those transferred from another hospital; or patients with Medicare as the primary payer. A predictive tool had a discriminatory capability with area under the curve estimates of 0.737 and 0.724 for derivation and validation cohorts, respectively.Conclusions: Our investigation revealed that the hospital discharge disposition pattern of patients with stroke in Tennessee was associated with the key patient characteristics of selected demographics, clinical indicators, and insurance status. These analyses resulted in the development of an easy-to-use predictive tool for early determination of hospital discharge disposition status. [ABSTRACT FROM AUTHOR]

Falls are common and dangerous for survivors of stroke at all stages of recovery. The widespread need to assess fall risk in real time for individuals after stroke has generated emerging requests for a reliable, inexpensive, quantifiable, and remote clinical measure/tool. In order to meet these requests, we explore the Functional Reach Test (FRT) for real-time fall risk assessment and implement the FRT function in mStroke, a real-time and automatic mobile health system for poststroke recovery and rehabilitation. mStroke is designed, developed, and delivered as an Application (App) running on a hardware platform consisting of an iPad and one or two wireless body motion sensors based on different mobile health functions. The FRT function in mStroke is extensively tested on healthy human subjects to verify its concept and feasibility. Preliminary performance will be presented to justify the further exploration of the FRT function in mStroke through clinical trials on individuals after stroke, which may guide its ubiquitous exploitation in the near future. [ABSTRACT FROM AUTHOR]

• Traffic accidents are a common danger everyone faces. • Different ML models with different creation methods are used to predict accidents. • Generic model with good performance and easy data collection. • Emergency responders benefit from shorter response times. • Local governments benefit from prescribed roadway changes. Given the ever present threat of vehicular accident occurrence endangering the lives of most people, preventative measures need to be taken to combat vehicle accident occurrence. From dangerous weather to hazardous roadway conditions, there are a high number of factors to consider when studying accident occurrence. To combat this issue, we propose a method using a multilayer perceptron model to predict where accident hotspots are for any given day in the city of Chattanooga, TN. This model analyzes accidents and their associated weather and roadway geometrics to understand the causes of accident occurrence. The model is offered as a live service to local law enforcement and emergency response services to better allocate resources and reduce response times for accident occurrence. Multiple models were made, each having different variables present, and each yielding varying results. [ABSTRACT FROM AUTHOR]