In distributed online optimization over a computing network with heterogeneous nodes, slow nodes can adversely affect the progress of fast nodes, leading to drastic slowdown of the overall convergence process. To address this issue, we consider a new algorithm termed Distributed Any-Batch Mirror Descent (DABMD), which is based on distributed Mirror Descent but uses a fixed per-round computing time to limit the waiting by fast nodes to receive information updates from slow nodes. DABMD is characterized by varying minibatch sizes across nodes. It is applicable to a broader range of problems compared with existing distributed online optimization methods such as those based on dual averaging, and it accommodates time-varying network topology. We study two versions of DABMD, depending on whether the computing nodes average their primal variables via single or multiple consensus iterations. We show that both versions provide strong theoretical performance guarantee, by deriving upperbounds on their expected dynamic regret, which capture the variability in minibatch sizes. Our experimental results show substantial reduction in cost and acceleration in convergence compared with the known best alternative.

We study the problem of joint offloading decision and resource allocation for mobile cloud networks with a computing access point (CAP) and a remote cloud center. We consider the case where the task computing requirement is not fully known before their execution. We aim to jointly optimize the offloading decisions as well as the allocation of computation and communication resources, to minimize a weighted sum of the average cost and cost variation. The problem is formulated as a mixed-integer program. We propose an efficient algorithm, termed Task Offloading and Resource Allocation with Uncertain Computing (TORAUC), and show that it always converges to a Karush-Kuhn-Tucker (KKT) point of an alternate form of the original problem, which has its binary constraints removed but guarantees an offloading decision solution that is arbitrarily close to binary. We extend TORAUC to TORAUC-MP for the case of a multi-processor CAP. Through trace-based simulation, we study the performance of TORAUC and TORAUC-MP. We observe that TORAUC is nearly optimal, and both algorithms substantially outperform several alternatives.

Millimeter-wave (mm-wave) communication is a promising technology for supporting extremely high data rates in the next generation wireless networks. Mm-wave signals experience high path loss and directional transmission is required to compensate the severe channel attenuation. The special characteristics of the mm-wave propagation arise opportunities as well as challenges for the network resource allocation problems. In this paper, a fair user association, beamwidth selection and power allocation problem for indoor mm-wave networks is studied. The objective of the optimization problem is to maximize the minimum user throughput to provide a fair resource distribution among the users. In our model, we take into account the unique mm-wave communications features, namely beam alignment procedure and directional transmission. Simulation results confirm the superior performance of the proposed solution compared to the existing approaches.

Millimeter-wave (mm-wave) communication is a promising technology for the next generation cellular networks. Motivated by the huge available bandwidth at these bands, it can be used to support the fastest-growing demands for the mobile data traffic such as multimedia applications. However, there are some challenges on the network connectivity at the mm-wave frequency bands. The high path loss, the limited diffraction capability due to the short wavelength, and having difficulties in penetrating through solid materials necessitate the line-of-sight and multi-hop communication in mm-wave networks. In this paper, we develop a multi-hop routing protocol which maximizes the sum quality of the uncompressed high-definition video applications for the device-to-device connections. The quality is measured as a function of the rate in this paper. We take into account the unique characteristics of the mm-wave propagation in our model. Simulation results show that the proposed algorithm achieves the optimal solution for high cooperation probability. It is also verified by the simulation that our algorithm outperforms the max-min flow routing protocol solution.

Millimeter-wave (mm-wave) communication is a promising technology for the next generation wireless networks. Motivated by the immense amount of bandwidth at these bands, it can be used to support the quality of service requirements for the bandwidth-intensive purposes, like the backhaul demands of the small cell base stations. However, the high path loss, the limited penetration ability and the intermittent connectivity necessitate utilizing the multi-hop transmission techniques to maintain network connectivity. In this paper, the outage performance of the mm-wave relaying systems is studied. We take into account the unique propagation characteristics of the mm-wave bands, namely the intermittent connectivity, and obtain the closed-form expression for the outage probability of the mm-wave multi-hop regenerative relaying system. Moreover, the closed-form approximation for the outage probability in a dual-hop nonregenerative case is also derived. The analytical expressions are verified by the simulation results.

With multimedia dominating the main traffic load of the wireless networks, device-to-device (D2D) communication is an efficient way of data offloading. By the growth of video data traffic, quality of experience (QoE), which is a user-centric measure, has become the focus of the research academia. In this paper, the video quality and fluency are considered as the factors that influence the user experience in D2D video streaming services. Aiming to enhance the user experience, we propose a QoE-aware power allocation for D2D video transmissions. The resource allocation target is to maximize the video quality while minimizing the data rates variations over the time-varying wireless channels. We then show that using dual decomposition technique, the problem can be implemented in a distributed way by the exchange of demand and price among the network and D2D users. Simulation results demonstrate that the proposed scheme can significantly improve the QoE in D2D video services.