Network slicing often involves instantiation of certain network functionality into network nodes, possibly subject to function chaining constraints. In this paper, we introduce the novel concept of elastic service function chain (SFC), which is an ordered list of virtual functions that may be optional or recursive, and we address network slicing with such chaining constraints as a Software Defined Topology (SDT) problem. We formulate the SDT problem as a combinatorial optimization problem that includes a multi-commodity flow problem and a bin packing problem as sub-problems. Since it inherits NP hardness from the bin packing sub problem, we develop a heuristic algorithm to tackle it. The algorithm's effectiveness and performance are evaluated via simulation study.

This article explores a method to dynamically size the backhaul links (shared resources) used to support cooperative decoding and to forward decoded data, in the case of limited bandwidth of the backhaul. The practical approach provides a scheduler with a mean to know the backhaul limitations and to use it adequately while interacting with a traffic engineering optimizer which reacts to changes in demands of the radio access network balancing where to allocate cooperative backhaul resources to maximize fairness and throughput of users.

Simultaneous bidirectional transmission and reception of signals on a single channel in a full duplex (FD) system is gaining increased attention from both academia and industry. Recent advances in radio transceiver design have been able to minimize the self interference (SI) to a great extent making realization of a FD transceiver possible. For FD deployment in a multi-cell environment, minimizing SI is not sufficient due to the presence of a complex array of network-wide interference. This paper investigates the critical deployment challenges for a dense urban FD multi- cell network and provides solutions for successful deployment of FD cellular networks. Extensive simulation results show that the proposed user selection and power control algorithms are able to increase the FD performance gain by more than 80% when compared to the traditional half duplex (HD) counterpart.

We explore the effect of imperfect knowledge on multipoint downlink traffic engineering (TE) (with dynamic point selection DPS) computed centrally and jointly with power control (PC) optimization versus distributed PC techniques. Instead of considering a single metric that shows qualities in a single dimension, we observe both performance and resilience by observing how many users get quality of experience (QoE) — or otherwise, how many users would complain that they do not receive the expected service. We identify that centralized traffic engineering provides the best performance (highest class of service) with distributed power control in a practical (imperfect) network. Although this has a clear drawback as it comes with a low resilience in face of bad service admission. We conclude that fifth generation networks will need to be able to dynamically change optimizing strategies to face different loading of the networks.

In this paper, we propose a step-wise optimal low power node (LPN) deployment algorithm for LTE heterogeneous networks. Our proposed LPN deployment algorithm takes signal quality, relative loading among macro cells and LPN cells, and user density into account. Simulation results show that the proposed algorithm can achieve close to 100% coverage gain and more than 50% average cell throughput gain over a random LPN deployment scheme.

This paper presents an optimization framework for downlink transmission parameters of mobile cellular systems. A typical network optimization approach is to divide the network into disjoint clusters of base-stations (BS). Optimization is then performed within each cluster for important parameters such as transmit power, precoder, etc. This approach is widely adopted in academic research and industrial standard bodies, e.g. 3GPP LTE-Advanced. While reducing the optimization complexity, this strategy suffers from a performance limit due to interference from the nodes outside cluster. We thus propose a framework to overcome this limit by allowing clusters to exchange their parameters and optimization information via low-rate and non-zero delay backhauls. System-level simulations for LTE downlink transmit power optimization show that the proposed optimization model, while having the low-complexity of cluster-based approach, could nearly achieve the performance of network-wise optimization. Therefore, this model is particularly suitable for optimization of 4G and beyond-4G cellular networks.

Channel inversion, and its minimum mean square error (MMSE) variation, are low complexity methods for space division multiple access (SDMA) in multiple input multiple output broadcast channel (MIMO-BC). As the channel matrix deviates from orthogonal, these methods result in a waste of transmit power. This paper proposes a trellis precoding method (across time and space) to improve the power efficiency. Adopting a 4-state trellis shaping method, the complexity of the proposed method, which is entirely at the transmitter side, is equivalent to the search in a trellis with 4Nstates where N is the number of transmit antennas. Numerical results are presented showing that the achievable gains, which depend on the channel realization, can be significantly higher than the traditional shaping gain which is limited to 1.53 dB.