Search

Background: Impaired knee extensor neuromuscular function has been frequently observed in individuals with patellofemoral pain (PFP); however, few researchers have aimed to understand the influence of sex on knee extensor neuromuscular function in this pathological population. The authors aimed to determine whether there are differences in knee extensor neuromuscular function between sexes in individuals with and without PFP.Hypothesis: Women with PFP would exhibit greater deficits in knee extensor neuromuscular function than men with PFP, compared with sex-matched individuals without PFP.Study Design: Cross-sectional, case-control study.Level of Evidence: Level 4.Methods: A total of 110 individuals were classified into 4 groups: women with PFP (n = 25); men with PFP (n = 30); women without PFP (n = 25); and men without PFP (n = 30). Knee extensor strength (isometric peak torque [PT]), activation (central activation ratio), early, late, and total phase rate of torque development (RTD0-100, RTD100-200, and RTD20-80%), and endurance (isokinetic average PT) were assessed using an isokinetic dynamometer. Group differences were assessed using a 2-way multivariate analysis of variance (sex by PFP).Results: Both women and men with PFP exhibited lower knee extensor strength, activation, early, late, and total phase RTD, and endurance versus sex-matched individuals without PFP (P< 0.05 for all comparisons). Women with PFP exhibited lower early phase (7.91 ± 2.02 versus 9.78 ± 2.43 N·m/s/kg; P< 0.01; Cohen d= 0.83), late phase (5.34 ± 1.02 versus 7.28±2.28 N·m/s/kg; P< 0.01; Cohen d= 1.37), and total phase (7.40 ± 2.57 versus 8.72 ± 2.57 N·m/s/kg; P= 0.03; Cohen d= 0.51) RTD than men with PFP.Conclusion: Compared with sex-matched pain-free individuals, women with PFP displayed lower RTD than men with PFP. Clinicians should note that among individuals with PFP, women are more likely to experience a greater impairment in their knee extensor torque-generating capacity than men.Clinical Relevance: Additional treatment strategies that effectively improve the ability to rapidly generate torque should be developed and implemented, especially when treating women with PFP.

Recent studies on semantic communication commonly rely on neural network (NN) based transceivers such as deep joint source and channel coding (DeepJSCC). Unlike traditional transceivers, these neural transceivers are trainable using actual source data and channels, enabling them to extract and communicate semantics. On the flip side, each neural transceiver is inherently biased towards specific source data and channels, making different transceivers difficult to understand intended semantics, particularly upon their initial encounter. To align semantics over multiple neural transceivers, we propose a distributed learning based solution, which leverages split learning (SL) and partial NN fine-tuning techniques. In this method, referred to as SL with layer freezing (SLF), each encoder downloads a misaligned decoder, and locally fine-tunes a fraction of these encoder-decoder NN layers. By adjusting this fraction, SLF controls computing and communication costs. Simulation results confirm the effectiveness of SLF in aligning semantics under different source data and channel dissimilarities, in terms of classification accuracy, reconstruction errors, and recovery time for comprehending intended semantics from misalignment.

In the next generation of terrestrial systems (e.g., 6th generation (6G)), the Ku- and Ka-bands are expected to be utilized, much like their usage in low Earth orbit (LEO) satellite communications (SATCOM). Consequently, cognitive LEO-terrestrial systems have emerged as a key focus of recent advancements to address the spectrum scarcity problem in future wireless communications. This letter focuses on power control for cognitive LEO-terrestrial systems. Specifically, we study an optimal power control scheme for LEO ground users (GUs) to maximize the LEO system throughput while limiting the harmful interference to terrestrial base stations (BSs). Unlike previous literature that studied this problem in a sub-optimal manner, our work provides an optimal solution while considering more realistic channel models. Numerical results demonstrate the performance disparities between our optimal scheme and the existing sub-optimal scheme.

This paper investigates the utilization of unmanned aerial vehicles (UAVs) to offer communication support to ground users in remote areas where terrestrial networks are unavailable. We assume these UAVs are equipped with transceivers, allowing them to establish connections with low-earth orbit (LEO) satellites. Serving as relay nodes, these UAVs are responsible for forwarding data packets received from ground users to LEO satellites with buffers with allocated channel resources. For resource allocation, we introduce a latency-optimized approach that aims to equalize the queue lengths of the UAVs as a convex optimization problem. However, this approach is not only myopic, as it centers on minimizing the maximum of the one-step-ahead queue lengths, but also relies on statistics related to arrival rates of data packets from ground users and LEO channel conditions. To address these issues, we also employ reinforcement learning for resource allocation. This strategy capitalizes on the quasi-periodic behaviors exhibited by channel conditions to the serving LEO satellite, which can be learned over time. Through simulations, we validate the effectiveness of this strategy. Furthermore, to analyze the proposed resource allocation approach, which equalizes the queue lengths, we examine the effective bandwidth with equalized quality-of-service (QoS) exponents to reflect the equalized queue lengths. This analysis reveals that the behavior of the proposed approach can be accurately predicted using the QoS exponent.

The efficient utilization of limited spectral resources in space presents a significant challenge given the rapidly increasing number of satellites (SATs) in various orbits. To address this challenge, it is paramount to enhance spectral efficiency by facilitating inter-operation between SATs in different orbits. In this letter, we investigate the impact of dynamic spectrum access (DSA) on the spectral efficiency of a SAT network that integrates geostationary (GEO) and low Earth orbit (LEO) SATs. Specifically, in the context of DSA, the secondary LEO SATs can opportunistically reuse the frequency bands originally assigned to primary GEO SATs. To characterize the resultant spectral efficiency in a realistic time-varying SAT network topology, we analyze the signal-to-interference-plus-noise ratio (SINR) at the nadir user of a randomly selected LEO SAT and an upper bound on average throughput of the LEO SAT. Simulation results demonstrate the significant advantages of DSA in enhancing the spectral efficiency of the integrated GEO-LEO SAT network.

Federated learning (FL) is a key enabler for efficient communication and computing, leveraging devices’ distributed computing capabilities. However, applying FL in practice is challenging due to the local devices’ heterogeneous energy, wireless channel conditions, and non-independently and identically distributed (non-IID) data distributions. To cope with these issues, this paper proposes a novel learning framework by integrating FL and width-adjustable slimmable neural networks (SNN). Integrating FL with SNNs is challenging due to time-varying channel conditions and data distributions. In addition, existing multi-width SNN training algorithms are sensitive to the data distributions across devices, which makes SNN ill-suited for FL. Motivated by this, we propose a communication and energy-efficient SNN-based FL (named SlimFL) that jointly utilizes superposition coding (SC) for global model aggregation and superposition training (ST) for updating local models. By applying SC, SlimFL exchanges the superposition of multiple-width configurations decoded as many times as possible for a given communication throughput. Leveraging ST, SlimFL aligns the forward propagation of different width configurations while avoiding inter-width interference during backpropagation. We formally prove the convergence of SlimFL. The result reveals that SlimFL is not only communication-efficient but also deals with non-IID data distributions and poor channel conditions, which is also corroborated by data-intensive simulations.

Classical medium access control (MAC) protocols are interpretable, yet their task-agnostic control signaling messages (CMs) are ill-suited for emerging mission-critical applications. By contrast, neural network (NN) based protocol models (NPMs) learn to generate task-specific CMs, but their rationale and impact lack interpretability. To fill this void, in this article we propose, for the first time, a semantic protocol model (SPM) constructed by transforming an NPM into an interpretable symbolic graph written in the probabilistic logic programming language (ProbLog). This transformation is viable by extracting and merging common CMs and their connections, while treating the NPM as a CM generator. By extensive simulations, we corroborate that the SPM tightly approximates its original NPM while occupying only 0.02% memory. By leveraging its interpretability and memory-efficiency, we demonstrate several SPM-enabled applications such as SPM reconfiguration for collision-avoidance, as well as comparing different SPMs via semantic entropy calculation and storing multiple SPMs to cope with non-stationary environments.

Although deep learning (DL) has already become a state-of-the-art technology for various data processing tasks, data security and computational overload problems often arise due to their high data and computational power dependency. To solve this problem, quantum deep learning (QDL) and distributed deep learning (DDL) has emerged to complement existing DL methods. Furthermore, a quantum distributed deep learning (QDDL) technique that combines and maximizes these advantages is getting attention. This paper compares several model structures for QDDL and discusses their possibilities and limitations to leverage QDDL for some representative application scenarios.

Integrating low-altitude earth orbit (LEO) satellites (SATs) and unmanned aerial vehicles (UAVs) within a non-terrestrial network (NTN), we investigate the problem of forwarding packets between two faraway ground terminals through SAT and UAV relays using either radio-frequency (RF) or free-space optical (FSO) link. Towards maximizing the communication efficiency, the associations with orbiting SATs and the trajectories of UAVs should be optimized, which is challenging due to the time-varying network topology and a huge number of possible control actions. To overcome the difficulty, we lift this problem to multi-agent deep reinforcement learning with a novel action dimensionality reduction technique. Simulation results corroborate that our proposed SAT-UAV integrated scheme achieves 1.99x higher end-to-end sum throughput compared to a benchmark scheme with fixed ground relays. While improving the throughput, our proposed scheme also aims to reduce the UAV control energy, yielding 2.25x higher energy efficiency than a baseline method only maximizing the throughput. Lastly, thanks to utilizing hybrid FSO/RF links, the proposed scheme achieves up to 62.56x higher peak throughput and 21.09x higher worst-case throughput than the cases utilizing either RF or FSO links, highlighting the importance of co-designing SAT-UAV associations, UAV trajectories, and hybrid FSO/RF links in beyond-5 G NTNs.

A mega-constellation of low-altitude earth orbit (LEO) satellites (SATs) are envisaged to provide a global coverage SAT network in beyond fifth-generation (5G) cellular systems. LEO SAT networks exhibit extremely long link distances of many users under time-varying SAT network topology. This makes existing multiple access protocols, such as random access channel (RACH) based cellular protocol designed for fixed terrestrial network topology, ill-suited. To overcome this issue, in this paper, we propose a novel contention-based random access solution for LEO SAT networks, dubbed emergent random access channel protocol (eRACH). In stark contrast to existing model-based and standardized protocols, eRACH is a model-free approach that emerges through interaction with the non-stationary network environment, using multi-agent deep reinforcement learning (MADRL). Furthermore, by exploiting known SAT orbiting patterns, eRACH does not require central coordination or additional communication across users, while training convergence is stabilized through the regular orbiting patterns. Compared to RACH, we show from various simulations that our proposed eRACH yields 54.6% higher average network throughput with around two times lower average access delay while achieving 0.989 Jain’s fairness index.

Large-scale wireless sensor networks are instrumental for several Internet of Things (IoT) applications involving data analytics and machine learning. The huge data volume generated by such networks imposes a change of paradigm from centralized machine learning to decentralized. Federated Learning (FL) is a well-known type of decentralized machine learning, whose efficiency heavily depends on the use of wireless communication resources. Random Access (RA) protocols, such as ALOHA, despite their simplicity, can improve the convergence time of FL systems if multiple orthogonal channels are used. This letter considers that devices are involved in the optimization of more than one model in a FL system, and then proposes an optimum method to allocate wireless resources in a multi-channel ALOHA setup. The proposed method outperforms uniform and fully-shared channel allocations in terms of convergence time.

This letter studies deep learning methods for beam selection in multiuser beamforming with limited feedback. We construct a set of orthogonal random beams and allocate the beams to users to maximize the sum rate, based on limited feedback regarding the channel power on the orthogonal beams. We formulate the beam allocation problem as a classification or a regression task using a deep neural network (DNN). The results demonstrate that the DNN-based methods achieve higher sum rates than a conventional limited feedback solution in the low signal-to-noise ratio regime under Rician fading, thanks to their robustness to noisy limited feedback.

Inspired by the multi-head attention (MHA) mechanism in natural language processing, this letter proposes an iterative single-head attention (ISHA) mechanism for multi-UAV path planning. The ISHA mechanism is run by a communication helper collecting the state embeddings of UAVs and distributing an attention score vector to each UAV. The attention scores computed by ISHA identify how many interactions with other UAVs should be considered in each UAV’s control decision-making. Simulation results corroborate that the ISHA-based communication and control framework achieves faster travel with lower inter-UAV collision risks than an MHA-aided baseline, particularly under limited communication resources.

Radiofrequency ablation provided long-term effectiveness in management of colorectal liver metastases, and a prediction model identified three independent risk factors for local tumor progression: subcapsular location, tumor size, and minimal ablative margin.

This letter proposes a neural network (NN) aided remote unmanned aerial vehicle (UAV) online control algorithm, coined oHJB. By downloading a UAV’s state, a base station (BS) trains an HJB NN that solves the Hamilton-Jacobi-Bellman equation (HJB) in real time, yielding a sub-optimal control action. Initially, the BS uploads this control action to the UAV. If the HJB NN is sufficiently trained and the UAV is far away, the BS uploads the HJB NN model, enabling to locally carry out control decisions even when the connection is lost. Simulations corroborate the effectiveness of oHJB in reducing the UAV’s travel time and energy by utilizing the trade-off between uploading delays and control robustness in poor channel conditions.

This paper proposes an intelligent sensing framework for Internet-of-Things platforms, where sensor measurements stem from multiple causes. Sensors are selectively chosen for data collection to identify the cause with partial measurements. We employ variational deep embedding, a generative model capable of clustering and generation, to identify causes, cluster measurements accordingly, and determine causes for estimating complete measurements from partial data. These estimates aid in efficient sensor selection for data collection. Results demonstrate early and reliable cause sensing and complete measurement estimation using the proposed framework.

Immersive social interactions of mobile users are soon to be enabled within a virtual space, by means of virtual reality (VR) technologies and wireless cellular systems. In a VR mobile social network, the states of all interacting users should be updated synchronously and with low latency via two-way communications with edge computing servers. The resulting end-to-end latency depends on the relationship between the virtual and physical locations of the wireless VR users and of the edge servers. In this letter, the problem of analyzing and optimizing the end-to-end latency is investigated for a simple network topology, yielding important insights into the interplay between physical and virtual geometries.

Real-time distributed control is a promising application of 5G in which communication links should satisfy certain reliability guarantees. In this letter, we derive closed-form maximum average rate when a device (e.g., industrial machine) downloads a sequence of $ n$ operational commands through cellular connection, while guaranteeing a certain signal-to-interference ratio ( $ {\mathsf {SIR}}$ ) coverage for all $n$ messages. The result is based on novel closed-form $n$ -successive $ {\mathsf {SIR}}$ coverage bounds. The proposed bounds provide simple approximations that are increasingly accurate in the high reliability region.

Energy efficiency and traffic load balancing are the important factors for wireless sensor network (WSN) designer. To achieve the efficient energy consumption and well-distributed load balancing, the clustering techniques can be an efficient solution. Furthermore, the several applications like real-time surveillance require the guaranteed delay in order to satisfy their delay requirements. The guaranteed delay can be one of the critical design issues for these WSN applications as well. In this paper, we propose a novel clustering algorithm to solve the energy and delay issues in WSN. Our algorithm uses an entropy-based cluster cost function, which takes into account the traffic load and the traffic variation. The simulation results show that our algorithm has longer lifetime and lower delay than low-energy adaptive clustering hierarchy (LEACH) and hybrid energy-efficient distributed clustering (HEED).

In an uplink underlaid device-to-device (D2D) cellular network, this paper considers its two aspects of throughput improvement. The two-fold gain comprises the throughput increase by offloading downlink cellular traffic to D2D communications, duplexing gain, and the increase by reusing uplink resources of D2D transmissions, capacity gain. Both impacts are investigated by exploiting stochastic geometry. On the basis of the analysis, a throughput optimal D2D operation guideline is provided for different network congestion environments.

The neuromechanical influence of anterior knee pain on sensorimotor function, independent of other knee pathology factors, is unclear. The purpose of this study was to determine whether experimental anterior knee pain independently alters static and dynamic postural control. Fifteen healthy participants performed single-leg quiet stance, landing, and walking tasks under 3 experimental conditions (pain, sham, control), at 3 different times (preinjection, injection, and postinjection) for each experimental condition. The dependent variables were (1) participant-perceived pain, (2) anterior–posterior and medial–lateral center of pressure velocity during single-leg quiet stance, (3) anterior–posterior, medial–lateral, and vertical time to stabilization during landing, (4) and medial–lateral center of pressure position during the stance phase of walking. Participant-perceived pain was influenced by time and experimental condition (The neuromechanical influence of anterior knee pain on sensorimotor function, independent of other knee pathology factors, is unclear. The purpose of this study was to determine whether experimental anterior knee pain independently alters static and dynamic postural control. Fifteen healthy participants performed single-leg quiet stance, landing, and walking tasks under 3 experimental conditions (pain, sham, control), at 3 different times (preinjection, injection, and postinjection) for each experimental condition. The dependent variables were (1) participant-perceived pain, (2) anterior–posterior and medial–lateral center of pressure velocity during single-leg quiet stance, (3) anterior–posterior, medial–lateral, and vertical time to stabilization during landing, (4) and medial–lateral center of pressure position during the stance phase of walking. Participant-perceived pain was influenced by time and experimental condition (P< .05); however, the other dependent variables were not significantly influenced by time or experimental condition, indicating that anterior knee pain does not independently affect static and dynamic postural control. [Athletic Training & Sports Health Care.2014;6(1):7–14.]

Context:Knee injuries are prevalent, and the associated knee pain is linked to disability. The influence of knee pain on movement biomechanics, independent of other factors related to knee injuries, is difficult to study and unclear.Objective:(1) To evaluate a novel experimental knee-pain model and (2) better understand the independent effects of knee pain on walking and running biomechanics.Design:Crossover study.Setting:Biomechanics laboratory.Patients or Other Participants:Twelve able-bodied volunteers (age 23 ± 3 years, height 1.73 ± 0.09 m, mass 75 ± 14 kg).Intervention(s):Participants walked and ran at 3 time intervals (preinfusion, infusion, and postinfusion) for 3 experimental conditions (control, sham, and pain). During the infusion time interval for the pain and sham conditions, hypertonic or isotonic saline, respectively, was continuously infused into the right infrapatellar fat pad for 22 minutes.Main Outcome Measure(s):We used repeated-measures analyses of variance to evaluate the effects of time and condition on (1) perceived knee pain and (2) key biomechanical characteristics (ground reaction forces, and joint kinematics and kinetics) of walking and running (P< .05).Results:The hypertonic saline infusion (1) increased perceived knee pain throughout the infusion and (2) reduced discrete characteristics of each component of the walking ground reaction force, walking peak plantar-flexion angle (range 62°–67°), walking peak plantar-flexion moment (range 95–104 N·m), walking peak knee-extension moment (range 36–49 N·m), walking peak hip-abduction moment (range 62–73 N·m), walking peak support moment (range 178–207 N·m), running peak plantar-flexion angle (range 38°–77°), and running peak hip-adduction angle (range 5–21°).Conclusions:This novel experimental knee pain model consistently increased perceived pain during various human movements and produced altered running and walking biomechanics that may cause abnormal knee joint-loading patterns.

Objective The authors asked the following research questions: will an anterior knee infusion model induce constant pain? will perceived pain alter motoneuron pool (MNP) excitability? and will treatments alter perceived pain and/or MNP excitability?Methods Thirty-six neurologically healthy volunteers participated in this randomised controlled laboratory study. To induce anterior knee pain (AKP), 5% hypertonic saline (0.12 ml/min with a total volume of 8.5 ml over 70 min) was injected into the infrapatellar fat pad of the dominant leg. One of four 30-min treatments was randomly assigned to each subject after pain was induced (acupuncture, cryotherapy, sham cryotherapy and no treatment). Five acupuncture needles (SP9, SP10, ST36, GB34 and an ah shi point) were inserted to a depth of 1 cm. Vastus medialis (VM) maximum Hoffmann reflexes normalised by maximum motor response were recorded from each subject at baseline, 20 min post-injection, 50 min post-injection and 70 min post-injection. To record pain perception, a visual analogue scale was used every 5 min after injection.Results An anterior knee infusion pain model increased perceived pain (p<0.0001). No change was found in VM MNP excitability among the four treatments (p<0.19) or at any of the time intervals (p<0.52). Cryotherapy reduced perceived pain compared with acupuncture (p=0.0003) and sham treatment (p=0.0002).Conclusions A pain model may be used in other neurophysiological intervention studies related to AKP. AKP alone may not directly alter quadriceps activation. None of the treatments altered VM MNP excitability. Cryotherapy reduced pain while a single session of acupuncture and sham treatments did not.

No Abstract.