The growing edge cloud computing paradigm allows flexible handling of latency-sensitive and computation-intensive applications operating on user devices as the Internet of Things and 5G technologies gain in popularity. Microservices based on container technology are regarded as a potential architecture when applied to edge computing because of their lightweight and layered image properties. However, many current studies on the combination of the two simply treat microservices as a replacement for traditional virtual machine architecture without fully utilizing its advantages. In addition to discussing the impact of image loading strategy on neighboring time slots, this paper also focuses on the advantages of microservices layered image sharing. Our research in this paper studies the microservice deployment and task offloading of a mobility-aware things-edge-cloud system, and a deep reinforcement learning-based algorithm is proposed in this work to make decisions that optimize the system’s long-term throughput and delay utility.

Motivated by the necessity of acquiring wall parameters for through-the-wall radar, a novel and general time domain model is proposed to measure the thickness and permittivity of single-layered slab-shaped materials, by exploiting the delays of the two surface reflections in the bistatic radar scheme. First, the two surface delays are formulated as functions of the unknown permittivity and thickness, as well as the accessible incident angle, and a nonlinear equation set is formed. Then based on a geometric analysis, in two separate bistatic delay tests with different antenna separations and standoff distances, the condition of identical incident angle is established. As such, the intra-wall delay is the same for the two delay tests, leading to a significant simplification and a closed-form solution to the equation set. Finally, a three-antenna test setup is constructed, with which the desired parameters can be acquired conveniently and accurately by performing bistatic tests at a set of standoff distances. Simulation and experiment show that our method can achieve high accuracy and strong robustness against noise.

With the increasing popularity of the internet of things (IoT) and 5G, emerging things-edge-cloud computing (TEC) paradigm provides a flexible way for execution of delay-sensitive and computation-intensive applications running on the user equipment (UE). By offloading these workloads to the mobile edge computing (MEC) or mobile cloud computing (MCC) server, the quality of experience, e.g., the execution delay, could be greatly improved. Nevertheless, conventional battery-powered devices face the challenge of battery exhaustion for task offloading. Using renewable energy via energy harvesting (EH) technologies has become a promising way to power these devices. In this paper, we investigate a multi-user green TEC system with EH UEs, each has a task buffer with limited capacity. A joint offloading decision and resource allocation problem is formulated, which addresses the long-term average execution delay, the task dropping and the long-term average energy cost constraint. A low-complexity online algorithm is proposed leveraging Lyapunov optimization framework and matroid theory, which jointly decides the offloading decision, the MEC server CPU frequencies and the transmit power for computation offloading. A unique advantage of this algorithm is that the decisions depend only on the current system state without requiring distribution information of the arrival tasks, wireless channel state, and EH processes. The implementation of the algorithm only requires to solve a deterministic problem in each time slot. Simulation results show that our proposed algorithm makes a best trade-off between minimizing the long-term average generalized delay and satisfying the long-term average energy cost constraint. Impacts of various parameters on the delay and energy cost performance are also discussed.

To suppress background clutter in short-range millimeter wave multiple-input multiple-output (MIMO) imaging, an adaptive and cost-efficient approach is proposed based on the availability of the background observation. By analyzing the origins of the background clutter, gain variation, and blocking effect are identified as the primary culprits responsible for the background clutter variation, which are treated separately in the proposed method. The gain variation is compensated with an adaptive power weighting (APW) procedure, where the gain variance is accurately and rapidly measured using the short-range profile segment associated with the interelement direct coupling. The clutter residual due to the blocking effect after APW is further removed by the combination of two zero-thresholding processes and coherence-factor-based image enhancement. The proposed method entails relatively light computational burden at the cost of acceptable memory increment, and is, thus, more amenable to real-time imaging requirement. Both simulations and experimental imaging tests on an MIMO array demonstrate the effectiveness and cost-efficiency of the proposed approach.

Edge Computing is one of the core technology of 5G networks. Edge computing deploys servers at the edge of the wireless access network, sinking cloud computing capabilities to the edge of the network, sharing the computing pressure of mobile users nearby, and improving the computing power of the entire network. Energy consumption is one of the important research issues of edge computing. At present, research on edge computing focuses on the energy consumption of terminal device, while little attention is paid to the energy consumption of edge servers. In this paper, considering the above two kinds of energy consumption, a global energy optimization strategy based on delay constraint in edge computing environment is proposed. Specifically, first, we use queuing theory to analyze the average delay of each terminal device and edge cloud processing computing tasks in the Internet of Things network, and the average delay of the entire system processing computing tasks. Secondly, we use the average delay as a constraint to establish a mathematical model for minimizing the total energy consumption of the device and the server. Then, we design a genetic algorithm-based offloading computation optimization algorithm to solve the above problems, so as to obtain the number of running servers in the edge cloud and the offload probability of IoT devices. Finally, the goal of minimizing the energy consumption of the overall system under the time delay constraint is achieved. The simulation experiment verifies the effectiveness of the energy optimization strategy.

In the energy‐harvesting wireless sensor networks (EH‐WSNs) with Time division multiple access (TDMA), it is challenging to assign time slots to the nodes because energy shortage causes some nodes unable to transmit in their time slots, resulting in the inefficiency in slot usage and the increase of the data packet delay. To overcome this problem, the slot assignment in TDMA is required to consider energy packet arrivals, where an energy packet is defined as the amount of energy that suffices for one transmission. In this study, the authors investigate the EH‐WSN with hybrid energy sources, in which the nodes harvest energy from the fixed inter‐arrival time (FIAT) and the random inter‐arrival time (RIAT) energy sources. After deriving the slot‐hitting ratios (SHRs) of energy packet arrivals for both FIAT and RIAT energy sources, they propose the SHR‐based TDMA (SHR‐TDMA) scheme. Then, they derive the delay arising from the awaiting slot (DAFAS) and formulate the DAFAS minimisation problem for the SHR‐TDMA. Solution to the DAFAS minimisation problem makes the slots optimally assigned according to the characteristics of the energy packet arrivals at the nodes. The simulation results show that the SHR‐TDMA outperforms the existing TDMA schemes in terms of DAFAS.

In the battery‐free wireless sensor network (BF‐WSN) that harvests radio signal energy, data delivery suffers from a longer delay arising from the energy‐harvesting period. It is significant to develop an energy‐efficient, low‐delay, and reliable data gathering scheme for the BF‐WSN. The goodput‐maximised data delivery scheme (GDDS) is proposed to reliably collect time‐constrained data in the IEEE 802.15.4‐based BF‐WSN. Under the GDDS, the sink's operation period consists of multiple data gathering cycles with each incorporating three phases: charging the nodes, assigning channel occupation time for the nodes, and receiving packets from the nodes. The scheme of accumulating correct data blocks (SACDB) is used in the third phase for the sink to gather data from the nodes. The authors develop an analytical model for the SACDB, from which they derive the time and the energy consumed in transmitting a packet. Then, they derive the goodput and the energy efficiency under the proposed GDDS. The GDDS aims at maximising the goodput by optimising the charging period, the number of data blocks, and the maximum number of transmission trials under the constraint on data gathering time. Simulation results show the GDDS outperforms the existing schemes in terms of the goodput and energy efficiency.

RF Energy harvesting technology has been proved one of the effective approaches for powering battery-free wearable devices in wireless body area networks. However, excessive electromagnetic radiation is harmful to human body. In this paper, we consider real-time healthcare scenario where wearable devices worn by mobile users collect their physiological data in real time and multiple wireless chargers are deployed for energy provision. Our goal is to minimize the maximal radiation degree among mobile users while maintaining normal work of wearable devices via adaptive power control of wireless chargers. We first discrete the users' moving trajectories and transform the stubborn problem into a docile one. Then we propose a distributed algorithm with interaction of wireless chargers, wearable devices and base station to solve it. Our proposed real-time power control scheme achieves an approximation ratio of (1+e) in general case. Furthermore, one special case is discussed. Simulation results reveal that our scheme is efficient and the maximal radiation degree among users can be reduced by almost 20\% as compared to the baseline scheme.

RF energy transfer, as the main far-field wireless energy transfer technology in wireless sensor networks, allows the relatively long charging distance from wireless charger to sensor nodes. Existing charging schemes based on a mobile RF energy charger neglect the energy consumption of the charger and its limited battery capacity. Motivated by this, we consider the practical charging scenario where energy-constrained mobile charger (MC) travels along a constrained long trajectory in the network area to wirelessly power the sensors, with multiple depots (for the energy provision of MC) deployed on the trajectory to achieve high energy efficiency. In this paper, we introduce VCEC, a Velocity-Control scheme of Energy-constrained mobile Charger to maximize the minimum charged energy in nodes after MC passes through the whole trajectory. Specifically, we first simplify the initial velocity-control problem to a tractable one by discretizing the trajectory into segments and propose a distributed algorithm to solve it. Then, we present a segment merging algorithm for the real-world applications. Our VCEC scheme achieves an approximation ratio of (1-θ)(1+ ε)-1. Simulations and test-bed experiments are conducted to show that VCEC promotes the bottleneck node's charged energy by at least 20% as compared to the baseline scheme where MC moves at a constant speed.

In the Internet of Things, wireless sensor networks (WSNs) are widely deployed. In recent years, wirelessly powered WSNs or battery‐free WSNs (BF‐WSNs), in which the nodes harvest energy from radio signals in the environment, have been emerging to support sustainable operation for WSNs. It is significant to design an energy‐efficient data delivery scheme for the BF‐WSNs. In this study, the authors propose the prefix code based scheme (PCBS) to save energy in data delivery by making use of the energy consumption disparity (ECD) between transmitting/receiving bit 0 and bit 1 in the existing non‐modulation baseband transmission or carrier‐modulation based passband transmission. The authors formulate an optimisation problem and use genetic algorithm to find its solution so that the energy‐efficient prefix codebook is obtained. The codebook dilutes the ECD by containing the energy‐consuming bit as few as possible, and the PCBS maps each m ‐bit data block into a prefix codeword in the codebook to conduct energy‐efficient transmission at the transmitter and vice versa at the receiver. Both the experiments on wireless identification sensing platform and the simulations demonstrate that the proposed PCBS outperforms the existing schemes in terms of energy saving.

In Battery-Free Wireless Sensor Network (BF-WSN), nodes are powered by the energy harvested from ambience instead of batteries. The nodes may frequently suffer from insufficient energy so that they need to alternate normal operation (such as transmitting/receiving, etc.) with harvesting energy. This brings in extra packet delay for the nodes to deliver data to the sink(s). Therefore, delivering data with shorter delay is a critical concern for BF-WSN nodes. In this paper, we first define the weight of wireless link that takes into account interference among links, balancing data load among the subtrees of the data gathering tree, and energy harvesting rate (EHR) of the nodes. Then, using the defined weight, we present the heuristic algorithm to build data gathering tree by letting the wireless link with a smaller weight join the tree prior to the ones with greater weights so that the wireless links either tending towards interference with the other ones, bringing in unbalance data load in the subtrees, or having smaller EHR are deferred to join the tree, thus reducing packet delay. Simulation results show the proposed data gathering scheme outperforms the existing scheme in terms of the delay per packet reaching the sink.

Wireless nanosensor networks (WNSNs), which consist of a lot of nanosensors with size of just a few hundred nanometers and are able to detect and sense new types of events at the nanoscale, are promising for a lot of unique applications like intrabody drug delivery systems, air pollution surveillance, etc. One important feature of WNSNs is that the nanosensors are highly energy-constrained, which makes it essential to develop energy efficient protocols for different layers of such networks. This paper focuses on a WNSN with on-off keying (OOK) modulation and explores the problem of transmission energy minimization in it. We first propose a general minimum transmission energy (MTE) coding scheme, which maps m-bit symbols into n-bit codewords with the least number of high-bits and thus results in the lowest energy consumption per symbol for any given m and n. We further determine the optimal setting of symbol length m and codeword length n in the MTE coding scheme so as to achieve the minimum energy consumption per data bit, which serves as the lower bound of transmission energy consumption in such WNSNs. Numerical results are provided to demonstrate the efficiency of the MTE coding scheme.

In an IEEE 802.11 based mobile ad hoc network (MANET), a network node accesses a common wireless channel through the distributed coordinate function (DCF), which is provided at the medium access control (MAC) layer of the IEEE 802.11 standard. The binary exponential backoff (BEB) algorithm, which uses slotted contention windows, plays an important role in the DCF. This paper develops a mathematical model for analyzing the performance of the BEB algorithm, which takes into account the packet loss probability of a wireless link. Based on the developed mathematical model, we derive the backoff probability distribution of a node, the average number of backoffs of a node, the average size of a contention window, and the average packet delay. Moreover, we attempt to find the optimal value of the initial contention window size, to which a node resets its contention window size after a successful transmission, in order to avoid the oscillation of the contention window size and thus maximize the utilization of the wireless channel.