Wearable technology and mobile healthcare systems are both increasingly popular solutions to traditional healthcare due to their ease of implementation and cost-effectiveness for remote health monitoring. Recent advances in research, especially the miniaturization of sensors, have significantly contributed to commercializing the wearable technology. Most of the traditional commercially available sensors are either mechanical or optical, but nowadays transdermal microneedles are also being used for micro-sensing such as continuous glucose monitoring. However, there remain certain challenges that need to be addressed before the possibility of large-scale deployment. The biggest challenge faced by all these wearable sensors is our skin, which has an inherent property to resist and protect the body from the outside world. On the other hand, biosensing is not possible without overcoming this resistance. Consequently, understanding the skin structure and its response to different types of sensing is necessary to remove the scientific barriers that are hindering our ability to design more efficient and robust skin sensors. In this article, we review research reports related to three different biosensing modalities that are commonly used along with the challenges faced in their implementation for detection. We believe this review will be of significant use to researchers looking to solve existing problems within the ongoing research in wearable sensors.
This paper aims to establish a new optimization paradigm to efficiently execute distributed learning tasks on wireless edge nodes with heterogeneous computing and communication capacities. We will refer to this new paradigm as “Mobile Edge Learning (MEL)”. The problem of adaptive task allocation for MEL is considered in this paper with the aim to maximize the learning accuracy, while guaranteeing that the total times of data distribution/aggregation over heterogeneous channels, and local computation on heterogeneous nodes, are bounded by a preset duration. The problem is first formulated as a quadratically-constrained integer linear problem. Being NP-hard, the paper relaxes it into a non-convex problem over real variables. We then propose a solution based on deriving analytical upper bounds on the optimal solution of this relaxed problem using KKT conditions. The merits of this proposed solution is exhibited by comparing its performances to both numerical approaches and the equal task allocation approach.
This paper proposes a framework to support federated learning or mobile edge learning (MEL) when there are both, delay constraints on the learning process and constraints on the energy consumed by each device. The aim is to maximize learning accuracy while guaranteeing that the total time taken and energy consumed by each learner in the system are bounded by a preset duration and maximum energy, respectively, while taking into account heterogeneous communication and communication capabilities of the channels and nodes. The problem of interest is shown to be NP-hard and a suggest-and-improve (SAI) approach is proposed based on the solution of the Lagrangian Dual problem (suggest) followed by a local optimizer based on coordinate descent (the improve step). The merits of this proposed solution, which is heterogeneity aware (HA), are exhibited by comparing its performances to both numerical approaches and the heterogeneity unaware (HU) approach.
21 December 2021 in IEEE Transactions on Mobile Computing
This paper introduces the new paradigm of Mobile Edge Learning "MEL" that enables the implementation of realistic distributed machine learning (DML) tasks on wireless edge nodes while taking into consideration the heterogeneous computing and networking environments. Therefore, a heterogeneity aware (HA) scheme is designed to solve the problem of dynamic task allocation for MEL in a way that maximizes the DML accuracy over wireless heterogeneous nodes or 'learners' while respecting the time constraints. The problem is first formulated as a quadratically-constrained integer linear program (QCILP). Being NP-hard, it is relaxed into a non-convex problem over real variables which can be solved using commercially available numerical solvers. The relaxation also allows us to propose a solution based on deriving the analytical upper bounds of the optimal solution using Lagrangian analysis and Karush-Kuhn-Tucker (KKT) conditions. The merits of the proposed analytical solution are demonstrated by comparing its performance to the numerical approaches and comparing the validation accuracy of the proposed HA scheme to the baseline heterogeneity unaware (HU) equal task allocation approach. Simulation results show that the HA schemes decrease convergence time up-to 56% and increase the final validation accuracy up-to 8%.
Predicting the prices of stocks is a very important field of research nowadays. Many different types of models have already been implemented. Two types of techniques include ARMA models and Neural Networks. In this work, ARMA models, along with two types of Neural Networks (Back Propagation) and Multi-Layer Perceptron (MLP) have been used. Furthermore, the two neural networks were combined with ARMA models (individually) in order to generate the best forecasted prices. The two indexes used for forecasting are the Dow Jones Industrial Average (DJI) and the Saudi Stock Exchange Tadawul (TASI). 800 values were used to predict the future 200 values. It was found out that for such large number of predictions, MLP yields the best results, which are drastically improved when combined with ARMA forecasting.
This paper proposes to maximize the accuracy of a distributed machine learning (ML) model trained on learners connected via the resource-constrained wireless edge. We jointly optimize the number of local/global updates and the task size allocation to minimize the loss while taking into account heterogeneous communication and computation capabilities of each learner. By leveraging existing bounds on the difference between the optimal and actual training loss, we derive an expression for the objective function in terms of the local updates. The resulting convex program is solved to obtain the optimal number of local updates which is used to obtain the total updates and batch sizes for each learner. The merits of the proposed solution, which is heterogeneity aware (HA), are exhibited by comparing its performance to the heterogeneity unaware (HU) approach.