Apostolos Destounis

Dynamic Power Allocation for Broadband Multi-Beam Satellite Communication Networks

Broadband satellite communication networks, operating at Ka band and above, play a significant role to worldwide telecommunication infrastructure, providing backhaul and direct-to-user satellite services. Rain attenuation at these frequencies is the most dominant impairment factor that degrades the performance of the system. The dynamic reconfiguration of multi-beam satellite antennas has recently been proposed as a fading mitigation technique. In this paper, the problem of finding the power allocation among the beams of such an antenna, aiming that the number of subscribers not receiving the desired quality of service is minimized is addressed for a satellite communications system with a GEO satellite serving fixed earth terminals. An algorithm for the dynamic power allocation using a physical-mathematical model for rain attenuation prediction is proposed. Extended simulation results highlight the improvement of the dynamic scheme over the static allocation based on long term rain attenuation characteristics.

Streaming big data meets backpressure in distributed network computation

We study network response to a stream of queries that require computations on remotely located data, and we seek to characterize the network performance limits in terms of maximum sustainable query rate that can be satisfied. The available network setup consists of (i) a communication network graph with finite-bandwidth links over which data is routed, (ii) computation nodes with certain computation capacity, over which computation load is balanced, and (iii) network nodes that need to schedule raw and processed data transmissions. Our aim is to design a universal methodology and distributed algorithm to adaptively allocate resources in order to support maximum query rate. The proposed algorithms extend in a nontrivial way the backpressure (BP) algorithm to take into account computations carried out in the presence of query streams. They contribute to the fundamental understanding of network computation performance limits when the query rate is limited by both the communication bandwidth and the computation capacity, a classical setting that arises in streaming big data applications in network clouds and fogs.

Controlling flow reconfigurations in SDN

Software-Defined Network (SDN) controllers include mechanisms to globally reconfigure the network in order to respond to a changing environment. While iterative methods are employed to solve flow optimization problems, demands arrive or leave the system changing the optimization instance and requiring further iterations. In this paper, we focus on the general class of iterative solvers considering an exponential decrease over time in the optimality gap. Assuming dynamic arrivals and departures of demands, the computed optimality gap at each iteration Q(t) is described by an auto-regressive stochastic process. At each time slot the controller may choose to apply the current iteration to the network or not. Applying the current iteration improves the optimality gap but requires flow reconfiguration which hurts QoS and system stability. To limit the reconfigurations, we propose two control policies that minimize the flow allocation cost while respecting a network reconfiguration budget. We validate our model by experimenting with a realistic network setting and using standard Linear Programming tools used in the SDN industry. We show that our policies provide a practical means of keeping the optimally gap small within a given reconfiguration constraint.

Traffic-Aware Training and Scheduling for MISO Wireless Downlink Systems

In this paper, the problem of feedback and active user selection in multiple-input single-output (MISO) wireless systems such that the systems stability region is as big as possible is examined. The focus is on a system in a Rayleigh fading environment where zero forcing precoding is used to serve all active users in every slot. Acquisition of the channel states is done via uplink training in time division duplexing mode by the active users. Clearly, only a subset of users can perform uplink training and the selection of this subset is a challenging and interesting problem especially in MISO systems. The stability regions of a baseline centralized scheme and two novel decentralized policies are examined analytically. In the decentralized schemes, the transmitter broadcasts periodically the queue state information and the users contend for the channel in a carrier sense multiple access-based manner with parameters based on the outdated queue state information and real-time channel state information. We show that, using infrequent signaling between the base station and the users, the decentralized policies outperform the centralized policy. In addition, a threshold-based user selection and training scheme for discrete-time contention is proposed. The results of this paper imply that, as far as stability is concerned, the users must be involved in the active user selection and feedback/training decision. This should be leveraged in future communication systems.

Adaptive clustering and CSI acquisition for FDD massive MIMO systems with two-level precoding

In this paper, we target a massive multiple-input/multiple-output (MIMO) system operating in frequency-division duplexing (FDD) mode, assuming the adoption of a two-level linear precoding strategy at the BS. We propose a novel strategy to effectively acquire the channel state information (CSI) at the base station (BS). In particular, we devise a cross-layer dynamic algorithm for user grouping, CSI acquisition and user scheduling that takes into account fairness considerations, application characteristics and quality of service (QoS) constraints of the users. We assess the merit of the proposed algorithm for a proportional fairness objective by comparing its performance with what is achieved by a relevant baseline algorithm in which user grouping is static and based only on the second order statistics, i.e., joint space division and multiplexing (JSDM). Our numerical findings illustrate that the proposed algorithm outperforms the baseline in terms of both fairness and speed of convergence to a steady state, and for different network topologies.

Traffic-aware training and scheduling for the 2-user MISO broadcast channel

In this paper we study the stability region of the 2-user MISO broadcast channel where the transmitter employs Zero Forcing precoding when both users are scheduled, taking into account the time overheads needed for uplink channel training. We show that, with proper signalling design, combining a decentralized policy with the baseline centralized one for user selection can increase the stability region of the system.

On Queue-Aware Power Control in Interfering Wireless Links: Heavy Traffic Asymptotic Modelling and Application in QoS Provisioning

In this work, we address the problem of power allocation for interfering transmitter-receiver pairs so that the probability that each queue length exceeds a specified threshold is fixed at a desired value. One application is satisfying QoS requirements in a dense cellular network. We deal with this problem using heavy traffic approximation techniques which lead to an asymptotic model of a (controlled) stochastic differential equation. The proposed power control strategy consists of allocating most of the power according to the states of the channel and a smaller fraction according to the queue lengths, for which we find a closed-form expression. We first consider a scenario where all channel realizations and queue lengths are known instantaneously to every transmitter. Then, the algorithm is extended to the case where only local SINR feedback is available and when queue length information is shared with delays among the transmitters. These models and results are also extended to the case where the transmitters are equipped with multiple antennas. Finally, the applicability in practical system settings are discussed and simulation results are provided to illustrate the performance of the proposed method.

A randomized probing scheme for increasing the stability region of multicarrier systems

In this work we address the problem of channel probing in a multicarrier downlink wireless network where in order to collect CSI feedback from each user at a channel, a fraction of the available time for transmission is used. This means that the time left to transmit is getting smaller. We study the aspect of stability of such a system and we find a randomized algorithm which can guarantee an expansion of the stability region with respect to full probing and prior works. In addition, we investigate a special case of a probing scheme that does not require knowledge of the statistics of the channels and can still enlarge the stability region of the system. Simulations show the performance of the proposed scheme.

A heavy traffic approach for queue-aware power control in interfering wireless links

In this work, we address the problem of power allocation for interfering transmitter-receiver pairs so that the probability that each queue length exceeds a specified threshold is fixed at a desired value. One application is satisfying QoS requirements in a dense cellular network. We address this problem using heavy traffic approximation techniques which lead to an asymptotic model described by a (controlled) stochastic differential equation. The power control strategy consists in allocating most of the power according to the wireless channel state and a smaller fraction according to the queue lengths. Simulation results in a simple setting illustrate that the proposed control policy can yield desirable results in practical systems.

Routing with blinkers: Online throughput maximization without queue length information

We study a service provisioning system where arriving jobs are routed in an online fashion to any of the available servers; typical applications include datacenters, Internet switches, and cloud computing infrastructures. A common goal in these scenarios is to balance the load across the servers and achieve maximum throughput. For example, the classical online policy Join-the-Shortest-Queue (JSQ) routes an arriving job to the server with the shortest instantaneous queue length. Although JSQ has desirable properties, it requires coordination between the routers and the servers in the form of queue length reports, which prohibits its practical usability in many scenarios. In this paper we study the practical case of “routing with blinkers”, where no coordination is allowed between the routers and the service provisioning system, and the routers act in an individual manner with limited view of the system state. Every router keeps a log of delays of all jobs it has routed in the past; these are delayed estimates of the actual server queue length. Although easy to acquire, such information is a highly inaccurate depiction of the system state and hence it is unclear whether it is enough to achieve maximum performance. Motivated by the fact that a reasonable policy such as Join-the-Shortest-Delay fails to achieve maximum throughput, we propose a novel routing policy that “samples” the servers periodically and achieves maximum throughput, subject to a condition for the service discipline of the server.