We present a novel framework for analyzing and optimizing the outage performance of a multiple amplify-and-forward relay network, where $N$ relays, each with finite cache capacity, are deployed to assist the transmission from a base station equipped with $L$ antennas to the destination. In this network, if the requested file is cached at the relays and the associated relay channels are in good condition, then the data are directly transmitted from the relays to the destination; otherwise, traditional two-hop transmission is performed. Based on this concept, we propose a relay-selection criterion to choose the best relay, which maximizes the received signal-to-noise ratio at the destination. For this criterion, we derive exact and asymptotic analytical expressions for the system outage probability in Nakagami- $m$ fading. To further improve the network outage performance, we optimize the cache placement, which minimizes the outage probability. Numerical and simulation experiments are performed to validate the analysis and demonstrate the conditions where the proposed cache placement strategy outperforms the existing ones.

Visible light communication (VLC)-based simultaneous positioning and orientating (SPAO), using received signal strength (RSS) measurements, is studied in this paper. RSS-based SPAO for VLCs of great challenge as it is essentially a non-convex optimization problem due to the nonlinear RSS model. To address this non-convexity challenge, a novel particle-assisted stochastic search (PASS) algorithm is proposed. The proposed PASS-based SPAO scheme does not require the knowledge of the height of receiver, the perfect alignment of transceiver orientations or inertial measurements. This is a huge technical improvement over the existing VLC localization solutions. The algorithmic convergence is established to justify the proposed ASS algorithm. In addition, a closed-form Cramer-Rao lower bound (CRLB) on localization error is derived and analyzed to gain insights into how the VLC-based SPAO performance is related to system configurations. It is shown that the receiver's position and orientation accuracy is linear with signal-to-noise ratio and direction information. In addition, the position accuracy decays with six powers of the transceiver distance, while the orientation accuracy decays with four powers of the transceiver distance. Finally, simulation results verify the performance gain of the proposed PASS algorithm for VLC-based SPAO.

This article investigates a wireless-powered intelligent radio environment, where a fractional nonlinear energy harvesting (NLEH) is proposed to enable an intelligent reflecting surface (IRS)-assisted wireless-powered Internet of Things (WP IoT) network. The IRS engages in downlink wireless energy transfer (WET) and uplink wireless information transfer (WIT). We aim to improve the overall performance of the considered network, and the approach is to maximize its sum throughput subject to constraints of two different types of IRS beam patterns and time durations. To solve the formulated problem, we first consider the Lagrange dual method and Karush–Kuhn–Tucker (KKT) conditions to optimally design the time durations in closed form. Then, a quadratic transformation (QT) is proposed to iteratively transform the fractional NLEH model into the subtractive form, where the IRS phase shifts are optimally derived by the complex circle manifold (CCM) method in each iteration. Finally, numerical results are demonstrated to promote the proposed scheme in comparison to the benchmark schemes, where the benefits are induced by the IRS compared with the benchmark schemes.

In this paper, we consider a wireless powered cooperative non-orthogonal multiple access (NOMA) relay network, in which one source is supposed to send independent messages to two users with the assistance of one energy-constrained relay that harvests energy from the source. Firstly, we study the minimum power consumption at the source node to fulfill the least required transmission rates by two users in both time switching relaying (TSR) strategy and power splitting relaying (PSR) one. Secondly, when the relay is provisioned with data buffer and energy storage, the long-term average power consumption minimization problem is formulated to take into account of the data and energy queue causality, peak transmit power constraint, and transmission mode selection. By using Lyapunov optimization framework, a novel buffer-aided transmission scheme (BATS) is proposed to asymptotically approach the optimal solution. Our analysis shows that, the PSR outperforms the TSR in terms of the realized energy efficiency, and BATS can be utilized to further improve the energy efficiency. It is disclosed that, there is an inherent trade-off between the long-term power consumption and the average queuing delay. In addition, larger user rates or less power consumption can be realized if a larger delay can be tolerated.

In this paper, we investigate the buffer-aided twoway wireless energy harvesting (EH) relay network comprising of two users and one relay, where two users exchange information with the assistance of the relay via three phases of EH, multiple-access, and broadcast. By transforming the opportunistic scheduling design problem into an equivalent convex problem, we present the adaptive design for the buffer-aided two-way EH relay network to maximize the long-term achievable rate region. To fulfill the delay sensitive transmission requirements, a delay-aware adaptive transmission (DAAT) scheduling scheme is proposed to guarantee the average end to end delivery delay by employing the Lyapunov optimization framework. Our analysis discloses that, the average achievable rate region of the twoway wireless EH relay network can be improved when fully considering the potentials by deploying data buffer and energy storage at the relay. There exists an inherent tradeoff among the achievable sum rate, the delivery delay, and the power consumption. It is shown that, when a certain time delay is tolerable, the DAAT scheduling scheme is able to realize a rate region arbitrarily close to the achievable rate region of the bufferaided two-way wireless EH relay network.

In datacenter networks, bandwidth-demanding elephant flows without deadline and delay-sensitive mice flows with strict deadline coexist. They compete with each other for limited network resources, and the effective scheduling of such mix-flows is extremely challenging. We propose a deep reinforcement learning with private link approach (DRL-PLink), which combines the software-defined network and deep reinforcement learning (DRL) to schedule mix-flows. DRL-PLink divides the link bandwidth and establishes some corresponding private-links for different types of flows to isolate them such that the competition among different types of flows can decrease accordingly. DRL is used to adaptively and intelligently allocate bandwidth resources for these private-links. Furthermore, to improve the scheduling policy, DRL-PLink introduces the novel clipped double Q-learning, exploration with noise, and prioritized experience replay technology for DDPG to address function approximation error, to induce lager and more randomness for exploration, as well as more effective and efficient experience replay in DRL respectively. The experiment results under actual datacenter network workloads (including Web search and data mining workload) indicate that DRL-PLink can effectively schedule mix-flows at a small system overhead. Compared with ECMP, pFabric, and Karuna, the average flow completion time of DRL-PLink decreased by 77.79%, 65.61%, and 23.34% respectively, when the deadline meet rate is increased by 16.27%, 0.02%, and 0.836% respectively. Additionally, DRL-PLink can also well achieve load balance between paths.