The mobile communication enabled by cellular networks is the one of the main foundations of our modern society. Optimizing the performance of cellular networks and providing massive connectivity with improved coverage and user experience has a considerable social and economic impact on our daily life. This performance relies heavily on the configuration of the network parameters. However, with the massive increase in both the size and complexity of cellular networks, network management, especially parameter configuration, is becoming complicated. The current practice, which relies largely on experts' prior knowledge, is not adequate and will require lots of domain experts and high maintenance costs. In this work, we propose a learning-based framework for handover parameter configuration. The key challenge, in this case, is to tackle the complicated dependencies between neighboring cells and jointly optimize the whole network. Our framework addresses this challenge in two ways. First, we introduce a novel approach to imitate how the network responds to different network states and parameter values, called auto-grouping graph convolutional network (AG-GCN). During the parameter configuration stage, instead of solving the global optimization problem, we design a local multi-objective optimization strategy where each cell considers several local performance metrics to balance its own performance and its neighbors. We evaluate our proposed algorithm via a simulator constructed using real network data. We demonstrate that the handover parameters our model can find, achieve better average network throughput compared to those recommended by experts as well as alternative baselines, which can bring better network quality and stability. It has the potential to massively reduce costs arising from human expert intervention and maintenance.

Federated learning (FL) has recently emerged as an innovative paradigm to train models among distributed agents. Conventional FL considers the center as an aggregator and trains from distributed data, while the collected global information at the center is not effectively utilized. Thus, the restricted information from local observations may limit the model accuracy. If FL can introduce data sets from the network server, the distributed models may be largely improved by the extra global information. Since network agents may not be completely trusted, the center cannot directly broadcast its raw data for security concern. Then, how to combine the central sets with FL? In this article, we propose to add a learning model at the center, which obtains the central sets as input. The outputs can be transmitted to network agents and integrated into local models instead of the raw data. The central and local models could be trained to form an integration for intelligent inference. Then, what is the integrated performance gain comparing with the original horizontal FL (HFL) and how to implement it? To figure out these two problems, we propose the vertical-HFL (VHFL) scheme, where models of the center and agents are trained collaboratively. We further analyze its convergence and the related communication channel, proposing the theoretical bounds to guide the network implementation of VHFL. Some simulation results will demonstrate the effectiveness of our proposed VHFL scheme. It is expected that VHFL will be an important block for the next generation of smart services.

Cellular network configuration plays a critical role n network performance. In current practice, network configuration depends heavily on field experience of engineers and often remains static for a long period of time. This practice is far from optimal. To address this limitation, online-learning-based approaches have great potentials to automate and optimize network configuration. Learning-based approaches face the challenges of learning a highly complex function for each base station and balancing the fundamental exploration-exploitation tradeoff while minimizing the exploration cost. Fortunately, in cellular networks, base stations (BSs) often have similarities even though they are not identical. To leverage such similarities, we propose kernel-based multi-BS contextual bandit algorithm based on multi-task learning. In the algorithm, we leverage the similarity among different BSs defined by conditional kernel embedding. We present theoretical analysis of the proposed algorithm in terms of regret and multi-task-learning efficiency. We evaluate the effectiveness of our algorithm based on a simulator built by real traces.

Cellular network performance depends heavily on the configuration of its network parameters. Current practice of parameter configuration relies largely on expert experience, which is often suboptimal, time-consuming, and error-prone. Therefore, it is desirable to automate this process to improve the accuracy and efficiency via learning-based approaches. However, such approaches need to address several challenges in real operational networks: the lack of diverse historical data, a limited amount of experiment budget set by network operators, and highly complex and unknown network performance functions. To address those challenges, we propose a collaborative learning approach to leverage data from different cells to boost the learning efficiency and to improve network performance. Specifically, we formulate the problem as a transferable contextual bandit problem, and prove that by transfer learning, one could significantly reduce the regret bound. Based on the theoretical result, we further develop a practical algorithm that decomposes a cell's policy into a common homogeneous policy learned using all cells' data and a cell-specific policy that captures each individual cell's heterogeneous behavior. We evaluate our proposed algorithm via a simulator constructed using real network data and demonstrates faster convergence compared to baselines. More importantly, a live field test is also conducted on a real metropolitan cellular network consisting 1700+ cells to optimize five parameters for two weeks. Our proposed algorithm shows a significant performance improvement of 20%.

The data transmission performance of a network protocol is closely related to the amount of available information about the network state. In general, more network state information results in better data transmission performance. However, acquiring such state information expends network bandwidth resource. Thus, a trade-off exists between the amount of network state information collected, and the improved protocol performance due to this information. A framework has been developed in the previous efforts to study the optimal trade-off between the amount of collected information and network performance. However, the effect of information delay is not considered in the previous analysis. In this paper, we extend the framework to study the relationship between the amount of collected state information and the achievable network performance under the assumption that information is subject to delay. Based on the relationship we could then obtain the optimal resource allocation between the data transmission and network state information acquisition in a time-varying network. We have considered both memoryless and memory-exploited scenarios in our framework. Structures of the Pareto optimal information collection and decision-making strategies are discussed. Examples of multiuser scheduling and multi-hop routing are used to demonstrate the framework's application to practical network protocols.

The minimum amount of information that should be supplied to transmitters to resolve traffic conflicts in a multiple access system is investigated in this paper. The arriving packets are modeled as the random points of a homogeneous Poisson point process distributed within a unit interval. The minimum information required is equal to the minimum entropy of a random partition that separates the points of the Poisson point process. Only a lower bound of this minimum is known in previous work. We provide an upper bound of this minimum entropy, and the gap with the existing lower bound is shown to be smaller than log2e bits. The upper bound asymptotically achieves the minimum entropy required to resolve per unit traffic. We then analyze the control information used to resolve the traffic conflicts in the splitting algorithm and in the slotted-ALOHA protocol, and identify their gaps with the theoretic bound.

Network protocol performance is closely related to the available information about the network state. However, acquiring such information expends network bandwidth resource. Thus a trade-off exists between the amount of information collected about the network state, and the improved protocol performance due to this information. A framework has been developed to study the optimal trade-off between the amount of collected information and network performance. However, the effect of information delay is not considered. In this paper, we extend the framework to study the impact of information delay on the value of network state information to network protocols, based on which optimal periodic information update policies could be obtained. The framework is illustrated by an example of multiuser scheduling, and observations about the impact of information delay on network protocols are obtained.

Network protocol performance is closely related to knowledge about the network state. However, acquiring such knowledge expends network bandwidth resource. Thus a trade off exists between the amount of bandwidth resource expended in acquiring knowledge about network state, and the improved protocol performance due to such knowledge. Previous work used rate distortion theory to calculate the minimum information required for certain network performance. However, this limit is asymptotic and might not be achievable due to the introduced infinite delay. This work develops a non-asymptotic framework to find a practical bound of the required information for certain network performance, and the strategies for implementing network information collection. The framework is illustrated by a wireless scheduling problem to show the quantitative relationship between collected traffic information and network throughput. Furthermore, we calculate the effective data rate by considering the overhead of network information collection, and determine the optimal resource allocation between information collection and data transmission.