For implementation of massive multiple-input multiple-output (MIMO) cellular systems in frequency division duplex (FDD) mode, accurate estimation of downlink channel state information (CSI) is necessary, but full radio channel reciprocity between the uplink and downlink does not exist in that mode. Existing work on estimating downlink CSI in FDD massive MIMO systems has considered such approaches as angle-of-arrival reciprocity, compressive sensing, using second-order channel statistics (particularly the channel covariance matrix (CCM)), and machine learning using deep neural networks (DNNs). Typical DNN-based approaches are unsuitable for this problem because DNNs require large datasets, thousands of training epochs, and are susceptible to environmental variations. To overcome these shortcomings, we develop a conditional generative adversarial network (CGAN) approach to uplink-to-downlink mapping of both CCMs and CSI. To apply this method, we convert the uplink and downlink CCMs/CSI to images and employ CGAN techniques previously applied to image translation. The normalized mean square error performance of the proposed CGAN is evaluated for several array sizes for both CCM and CSI mapping. For uplink-to-downlink CSI mapping, we also examine the spectral efficiency performance of our CGAN-based method, as well as the impact of pilot reuse; both simulated and measured CSI data are considered. Our results demonstrate performance improvement over existing algorithms.

Conventional massive multiple-input multiple-output (MIMO) systems provide high spectral efficiency, throughput, and energy efficiency, but suffer from high inter-cell interference and poor cell-edge coverage. Cell-free massive MIMO addresses these shortcomings by geographically distributing access points (APs), each with one or more antennas, that form a virtual massive MIMO array instead of co-locating all antennas at a base station. This distributed AP placement significantly reduces the average distance between a user equipment (UE) and an AP. Existing work mainly considers the ideal canonical case where each UE is served by all APs, which is impractical because of limited fronthaul capacity and finite computational resources. Therefore, to make the network scalable to an arbitrary size, a user-centric approach should be adopted, where each UE is served by a personalized cluster of nearby APs. However, the clustering problem is combinatorially complex and may be too time consuming to solve optimally when the UEs are in motion. Therefore, in this work, we develop a multi-agent reinforcement learning (MARL) algorithm for AP selection and clustering, where each AP is an agent in the MARL algorithm and trained to near-optimally select for itself which UEs to serve. Simulation results demonstrate our MARL algorithm outperforms typical AP selection algorithms such as greedy selection as well as more sophisticated ones from the literature, and also provides performance comparable to the ideal canonical case.

Network slicing to support multi-tenancy plays a key role in improving the performance of 5G networks. In this paper, we propose a novel framework for network slicing with the goal of maximizing the expected utilities of tenants in the backhaul and Radio Access Network (RAN), where we reconfigure slices according to the time-varying user traffic and channel states. Upon the arrival of new statistics from users and channels and considering the expected utility from serving users of a slice and the reconfiguration cost, we formulate a sparse optimization problem to reconfigure resources for network slices with the maximum isolation of reserved resources. The formulated optimization is non-convex and difficult to solve. We use the group LASSO regularization and successive upper-bound minimization techniques to solve this problem by iteratively solving a sequence of convex approximations of the original problem. Simulation results verify that our approach outperforms the existing state of the art method.

Estimating or predicting the downlink channel state information (CSI) is extremely important for practical implementation of frequency division duplex (FDD) massive MIMO. Estimation of downlink CSI from uplink CSI using second order channel statistics, namely the channel covariance matrix (CCM), is a promising approach. However, published work so far has rarely applied machine learning techniques to solve this problem using CCMs, most probably due to the unavailability of a direct mapping function or parametric model for supervised learning to convert from uplink to downlink CCMs. In this paper, we develop a conditional generative adversarial network (CGAN) method for uplink-to-downlink CCM conversion. To apply the CGAN-based method, we convert the uplink and downlink CCMs to images and use image translation techniques for CGANs. The normalized mean square error performance of the proposed CGAN is evaluated for several antenna array sizes and with both perfect and imperfect knowledge of the CCMs. Our results demonstrate performance improvement over existing algorithms.

Network Function Virtualization (NFV) is an efficient approach to simplify and accelerate the deployment of diverse network services. A critical challenge lies in mapping Virtual Network Functions (VNFs) to high-volume servers, resource allocation, and traffic routing. In this paper, we study the joint problem of VNF placement on servers and traffic engineering for a network spanning multiple subnetworks. Each subnetwork is owned and controlled by a different administrator. We formulate the joint problem of routing and VNF placement cost minimization subject to flow demands where processing flows in local subnetworks is encouraged. To ensure sensitive information of administrators remains private and to cut the implementation cost, a scalable and decentralized approach based on the proximal Alternating Direction Method of Multipliers (ADMM) is proposed. Extensive numerical evaluations show the efficiency of our approach against existing work.

The exponential growth of data traffic volume dominates the demand for the next generation mobile networks (5G). The consistent and satisfied quality of experience (QoE) is one of the leading challenges of provisioning services in 5G mobile networks. Thus, in this paper, we propose a novel adaptive prefetching scheme to compensate the undesired transmission conditions in the 5G mobile network by extending the content prefetching concept from the users to the network. Specifically, an optimization problem is proposed for a prefetching scheme that adaptively retrieves users' data to access nodes and (or) user equipments (UEs) before the actual requests according to the network status, QoE status, predicted data rates, and mobility patterns of users. For the sake of tractability, the prefetching problem is transferred to a convex problem that can be solved efficiently. Accordingly, to implement the proposed schemes in the 5G network, system interactions among entities in the network are designed to realize prefetching-related functions. A signaling protocol to support the adaptive prefetching scheme is also presented. Simulation results show that an adaptive prefetching scheme can improve the network performance significantly.

To support multiple on-demand services over several fixed communication networks, the network operators must allow flexible customization and fast provision of their network resources. One effective approach is network virtualization, whereby each service is mapped to a virtual subnetwork providing dedicated on-demand support. In practice, each service consists of a pre specified sequence of functions, called a service function chain (SFC). Moreover, each function in a SFC can only be provided by some given network nodes. Thus, to support a given service, we must select network function nodes according to the SFC, and determine the routing strategy through the function nodes in the specified order. A crucial problem that needs to be addressed is how to optimally allocate the network resources while satisfying multiple service requirements specified by the service function chains, subject to link and node capacity constraints. In this paper, we formulate the problem as a mixed binary linear program and establish its NP-hardness. Furthermore, we propose an efficient penalty successive upper bound minimization algorithm to solve the problem. We also present simulation results to demonstrate the effectiveness of the proposed algorithm.

Traffic engineering (TE) problem is a central component of the next generation cloud-based wireless networks. In this paper, we study a new resource allocation scheme for effective traffic engineering under practical constraints such as the finite buffer size at each node. To reduce the computational effort required in the existing single-slot TE approaches and to deal with practical hardware limitations on link flows and buffers, we propose a two time-scale, low-complexity TE algorithm which incorporates a novel link scheduling component. The algorithm can be distributedly implemented. Simulation results demonstrate the effectiveness of the proposed algorithm.

In this paper, we consider the problem of admission control of wireless virtual network (VN) service requests in a multi-service network with service-specific service function chains (SFC). SFC imposes traversal constraints on flows, i.e., each flow must visit certain service-specific nodes in a specific order. We leverage the fact that all flows within a service have common traversal constraints and propose a “service-based” admission control approach. For both of the service-based and flow-based approaches, we propose novel joint optimizations of backhaul and access networks for the admission control of VN service requests. We show that our optimization formulation is convex, hence computationally efficient and tractable. We also show that the proposed method is applicable to general backhaul and access networks. The simulation results show that the service-based approach incurs no loss in optimality while greatly reducing problem complexity compared to the flow-based counterpart.

In this paper, we present low-complexity algorithms for cache placement problem in software-defined radio access networks. Our key design target is that the packet delay from any cache to its associated users is below a threshold. This feature is a significant improvement compared to conventional approaches relying on the average hop count. We introduce a traffic engineering solution for joint multiserver selection and rate allocation. This algorithm is used to minimize the cache placement cost function, such as delivery cost per bit. Our cache placement method allows prediction of user throughput and thus the network throughput. Extensive simulation results verify that our method can give optimal or near optimal network throughput while guaranteeing packet delay performance.

To support multiple on-demand services over fixed communication networks, network operators must allow flexible customization and fast provision of their network resources. One effective approach to this end is network virtualization, whereby each service is mapped to a virtual subnetwork providing dedicated on-demand support to network users. In practice, each service consists of a prespecified sequence of functions, called a service function chain (SFC), while each service function in a SFC can only be provided by some given network nodes. Thus, to support a given service, we must select network function nodes according to the SFC and determine the routing strategy through the function nodes in a specified order. A crucial network slicing problem that needs to be addressed is how to optimally localize the service functions in a physical network as specified by the SFCs, subject to link and node capacity constraints. In this paper, we formulate the network slicing problem as a mixed binary linear program and establish its strong NP-hardness. Furthermore, we propose efficient penalty successive upper bound minimization (PSUM) and PSUM-R(ounding) algorithms, and two heuristic algorithms to solve the problem. Simulation results are shown to demonstrate the effectiveness of the proposed algorithms.

We consider the distributed network routing problem in a large-scale hierarchical network whereby the nodes are partitioned into subnetworks, each managed by a network controller (NC), and there is a central NC to coordinate the operation of the distributed NCs. We propose a semi-asynchronous routing algorithm for such a network, whereby the computation is distributed across the NCs and is parallel within each NC. A key feature of the algorithm is its ability to handle a certain degree of asynchronism: the distributed NCs can perform their local computation asynchronously at different processing speed. The efficiency of the proposed algorithm is validated through numerical experiments.

We consider the problem of traffic engineering (TE) in a software-defined radio access network (SD-RAN) consisting of backhaul and radio access parts. The role of TE is to route traffic from sources to their corresponding destinations in the network based on its knowledge of network topology and link capacities. Unlike wired networks, where links have fixed capacities, the capacity of a wireless link in a radio access network depends on the amount of resource (e.g., frequency, time, etc.) assigned to it by wireless link scheduler. Therefore, TE for SD-RAN also involves optimization of link/user scheduling. We propose a cross-layer solution by decoupling the SD-RAN TE problem into a wired network TE problem and a user scheduling problem using alternating direction method of multipliers (ADMM). The ADMM iterations are guaranteed to converge if the objective function of the SD-RAN TE problem is convex. The simulation results show that our proposed online implementation closely follows the optimal solution in a dynamic setting where individual flows independently join/leave the network.

We consider traffic engineering (TE) in networks where quality of service (QoS)-guaranteed as well as best effort (BE) services are present. We propose a joint TE optimization technique for both QoS and BE traffic using multi-objective optimization framework. We show that joint TE optimization results in more efficient network resource usage compared to sequential TE methodology where TE decision (finding paths and traffic splitting among paths) is made for QoS-guaranteed traffic first by assigning appropriate network resources and then TE decision is made for BE traffic according to remaining network resources. Furthermore, we consider QoS traffic demand fluctuations and propose a joint TE problem formulation to address both QoS traffic demand fluctuations and efficient network resource usage for BE traffic.

We consider central traffic engineering in a software-defined radio access network (SD-RAN) consisting of routers, wireless access nodes, and wired/wireless links. We present a multi-path traffic engineering formulation for downlink transmission considering both backhaul and wireless access constraints. We show large throughput and quality of experience (QoE) gains over traditional single-path transmission. In order to increase robustness against unpredicted changes in the radio access network, we propose using link buffer status as feedback to traffic engineering and present a heuristic method for traffic engineering to adjust flow allocations according to buffer status observation.

We consider a cloud-based heterogeneous network of base stations (BSs) connected via a backhaul network of routers and wired/wireless links with limited capacity. The optimal provision of such networks requires proper resource allocation across the radio access links in conjunction with appropriate traffic engineering within the backhaul network. In this paper, we propose an efficient algorithm for joint resource allocation across the wireless links and flow control over the entire network. The proposed algorithm, which maximizes the min-rate among all the transmitted commodities, is based on a decomposition approach that leverages both the alternating direction method of multipliers (ADMM) and the weighted-MMSE (WMMSE) algorithm. We show that this algorithm is easily parallelizable and converges globally to a stationary solution of the joint optimization problem. The proposed algorithm can also be extended to networks with multi-antenna nodes and other utility functions.

Tomlinson-Harashima precoding (THP) for downlink MU-MIMO channel can be modeled as user signal perturbation (constellation expansion) followed by linear precoding. The effective linear precoder in standard THP is merely a function of the MU-MIMO channel, which does not provide the flexibility for user power allocation. A modified THP algorithm is proposed so that the effective linear precoder can be arbitrary. As a special case, zero-forcing THP (ZF-THP) with arbitrary user power allocation is considered for which optimal user power allocation is derived to maximize the weighted sum-rate of users.

For a state-dependent DMC with input alphabet X and state alphabet S where the i.i.d. state sequence is known causally at the transmitter, it is shown that by using at most |X||S|-|S|+1 out of |X||S|input symbols of the Shannon's associated channel, the capacity is achievable. As an example of state-dependent channels with side information at the transmitter, M-ary signal transmission over AWGN channel with additive Q-ary interference where the sequence of i.i.d. interference symbols is known causally at the transmitter is considered. For the special case where the Gaussian noise power is zero, a sufficient condition, which is independent of interference, is given for the capacity to be log2M bits per channel use. The problem of maximization of the transmission rate under the constraint that the channel input given any current interference symbol is uniformly distributed over the channel input alphabet is investigated. For this setting, the general structure of a communication system with optimal precoding is proposed.