In this paper, we aim to design an autonomous tracking system for a swarm of unmanned aerial vehicles (UAVs) to localize a radio frequency (RF) mobile target. In the system, UAVs equipped with omnidirectional received signal strength (RSS) sensors can cooperatively search the target with a specified tracking accuracy. To achieve fast localization and tracking in the highly dynamic channel environment (e.g., time-varying transmit power and intermittent signal), we formulate a flight decision problem as a constrained Markov decision process (CMDP) with the main objective of avoiding redundant UAV flight path. Then, we propose an enhanced multi-agent reinforcement learning to coordinate multiple UAVs performing real-time target tracking. The core of the proposed scheme is a feedback control system that takes into account the uncertainty of the channel estimate. We prove that the proposed algorithm can converge to the optimal decision. Our simulation results show that the proposed scheme outperforms standard Q-learning and multi-agent Q-learning algorithms in terms of searching time and successful localization probability.

Received signal strength (RSS) fingerprint-based indoor localization has received increasing popularity over the past decades. However, it suffers from the high calibration effort for fingerprint collection. In this paper, a Centralized indooR localizatioN method using Pseudo-label (CRNP) is proposed, which employs a small set of labeled data (RSS fingerprint) along with large volumes of unlabeled data (RSS values without coordinates) to reduce the workload of labeled data collection and improve the indoor localization performance. However, the rich location data is large in quantity and privacy sensitive, which may lead to high network cost (i.e., data transmission cost, data storage cost) and potential privacy leakage for data transmission to the central server. Therefore, a decentralized indoor localization method incorporating CRNP and federated learning is devised, which keeps the location data on local users’ devices and improves the shared CRNP model by aggregating users’ updates of the model. The experiment results demonstrate that (i) the proposed CRNP enables to improve the indoor localization accuracy by using unlabeled crowdsourced data; (ii) the designed decentralized scheme is robust to different data distribution and is capable to reduce the network cost and prevent users’ privacy leakage.

With offloading the tasks that mobile users (MUs) running in their mobile devices (MDs) to the data centers of remote public clouds, mobile cloud computing (MCC) can greatly improve the computing capacity and prolong the battery life of MDs. However, the data centers of remote public cloud are generally far from the MUs, thus long delay will be caused due to the transmission from the base station to the public clouds over the Internet. Mobile edge computing (MEC) is recognized as a promising technique to augment the computation capabilities of MDs and shorten the transmission delay. Nevertheless, compared with the traditional MCC and MEC generally has a limited number of cloud resources. Therefore, making a choice on offloading task to the MCC or MEC is a challenging issue for each MU. In this paper, we investigate service selection in a mobile cloud architecture, in which MUs select cloud services from two cloud service providers (CSPs), i.e., public cloud service provider (PSP) and an edge cloud service provider (ESP). We use M/M/$\infty $ queue and M/M/1 queue to model PSP and ESP, respectively. We analyze the interaction of the two CSPs and MUs by adopting Stackelberg game, in which PSP and ESP set the prices first, and then the MUs decide to select cloud services based on performances and prices. In particular, we study the relationship between PSP and ESP in the simultaneous-play game (SPG) scenario, in which they compete to set prices of their cloud services simultaneously. Our numerical results show that MUs prefer to select service from the edge cloud if the number of tasks they run is small. In another hand, more tasks will be offloaded to the remote public cloud if the number of tasks they run becomes large.

Current TCP/IP based network is suffering from the tremendous usage of IP. Recently, content centric network (CCN) is proposed as an alternative of the future network architecture. In CCN, data itself, which is authenticated and secured, is a name and can be directly requested at the network level instead of using IP and DNS. Moreover, routers in CCN have caching abilities. Then end users can obtain the data from routers instead of remote server if the content has been stored in the routers, thus the overall network performance could be improved by reducing the transmission hops. Orthogonally, video plays a more and more important role nowadays and dominates the network traffic. Response time of each video request greatly affects the quality of user experience (QoE), users may even abandon the requested video service if they have to wait for long time before the video playback. Hence how to provide fast-start video delivery in CCN is critical. In this paper, we target to provide users fast-start video delivery in CCN. Specifically, we design a new caching policy for popularity-aware video caching in topology-aware CCN. And we propose to encode the video using scalable video coding (SVC) for fast-start video delivery and cache each video layer separately following the designed caching policies. Given an assigned weight by users, the tradeoff between the waiting time and received video quality is studied. Simulations are conducted to verify the performances and the results show that the proposed scheme outperforms state-of-the-art schemes significantly in typical scenarios.

Previous studies have shown that the motion intention recognition for lower limb prosthesis mainly focused on the identification of performed gait. However, the bionic prosthesis needs to know the next movement at the beginning of a new gait, especially in complex operation environments. In this paper, an upcoming locomotion prediction scheme via multilevel classifier fusion was proposed for the complex operation. At first, two motion states, including steady state and transient state, were defined. Steady-state recognition was backtracking of a completed gait, which would be used as prior knowledge of motion prediction. In steady-state recognition, surface electromyographic (sEMG) and inertial sensors were fused to improve recognition accuracy; five typical locomotion modes were recognized by random forest classifier with over 97.8% accuracy. The transient state was defined as an observation period at the initial stage of upcoming movement, in which only the sEMG signal was recorded due to the limitation of sliding window length. LightGBM classifier was validated to outperform other methods in the accuracy and prediction time of transient-state recognition. Finally, a simplified HMM model based on prior knowledge and observation result was constructed to predict upcoming locomotion. The results indicated that the locomotion prediction was over 91% accuracy. The proposed scheme implements the locomotion prediction at the initial stage of each gait and provides critical information for the gait control of lower limb prosthesis.

Gait phase detection is an important procedure in the application of many lower limb auxiliary robots. Gait phase detection algorithms that utilize surface electromyography (sEMG) signals have been developed to overcome unnatural walking. However, traditional studies mostly consider precise gait walking events when the subject is focusing only on walking, and the accuracy of traditional algorithms is susceptible to more realistic gait scenarios, such as walking with cognitive tasks. In this paper, the gait phase detection is considered in more realistic and challenging scenarios, in which the subject is walking while performing cognitive tasks. A kernel linear discriminant analysis (LDA)-based nonlinear fusion model is proposed for gait recognition, which can effectively reduce the error caused by cognitive tasks, making it an ideal model for gait phase detection while cognitive tasks are performed in the process of walking. Furthermore, an Internet of Things (IoT) framework is incorporated to reduce the gait phase detection algorithm’s process time by offloading the data from local sEMG sensors to an IoT server with powerful computation capability. The experiments have been conducted to validate our proposed algorithms, demonstrating that the boundaries between stance period and swing period are more blurred when walking with cognitive tasks.