In mobile networks, network performance can be severely degraded by a signaling storm, a phenomenon characterized by control plane (C-plane) congestion due to excessive signaling messages. This paper identifies congestion of inter-site links and overload of a specific NF instance as factors leading to a signaling storm. To address these problems, we propose a combined placement and routing algorithm approach. Firstly, our joint placement and routing algorithms perform geographically distributed deployment of C-plane NF instances according to the fluctuating mobility patterns of each user equipment (UE). Secondly, our routing algorithm identifies an appropriate C-plane NF instance to manage each UE, considering both the utilization rate and the number of contexts possessed by each NF instance. Extensive simulation evaluations demonstrate that our proposed methods significantly reduce inter-site signaling messages compared to existing placement scenarios. Furthermore, the dispersion of C-plane NF instances’ utilization rate is reduced, enhancing network performance and efficiency.

With the 5th Generation mobile communication system (5G), ultrahigh capacity and ultralow latency communication are realized. Toward the next generation, i.e., 6th Generation mobile communication system (6G), even higher capacity or lower latency is needed. This paper focuses on the significant processing delay in the control plane when a User Equipment (UE) state transitions from Radio Resource Control (RRC)-IDLE to RRC-CONNECTED. On the transition, many interactions exist between a UE and a Base Station (BS) or a mobile core system, such as a connection establishment or context data exchange. We propose eliminating the effect of processing delay by introducing an Always-Connected Enablement BS (ACE-BS). We conduct experiments in an actual local 5G environment with UEs, BSs, and mobile cores, demonstrating the low latency communication by using the ACE-BS.

This study develops a federated learning (FL) framework overcoming largely incremental communication costs due to model sizes in typical frameworks without compromising model performance. To this end, based on the idea of leveraging an unlabeled open dataset, we propose a distillation-based semi-supervised FL (DS-FL) algorithm that exchanges the outputs of local models among mobile devices, instead of model parameter exchange employed by the typical frameworks. In DS-FL, the communication cost depends only on the output dimensions of the models and does not scale up according to the model size. The exchanged model outputs are used to label each sample of the open dataset, which creates an additionally labeled dataset. Based on the new dataset, local models are further trained, and model performance is enhanced owing to the data augmentation effect. We further highlight that in DS-FL, the heterogeneity of the devices’ dataset leads to ambiguous of each data sample and lowing of the training convergence. To prevent this, we propose entropy reduction averaging, where the aggregated model outputs are intentionally sharpened. Moreover, extensive experiments show that DS-FL reduces communication costs up to 99 percent relative to those of the FL benchmark while achieving similar or higher classification accuracy.

With regard to the implementation of WiFi sensing agnostic according to the availability of channel state information (CSI), we investigate the possibility of estimating a CSI matrix based on its compressed version, which is known as beamforming feedback matrix (BFM). Being different from the CSI matrix that is processed and discarded in physical layer components, the BFM can be captured using a medium-access-layer frame-capturing technique because this is exchanged among an access point (AP) and stations (STAs) over the air. This indicates that WiFi sensing that leverages the BFM matrix is more practical to implement using the pre-installed APs. However, the ability of BFM-based sensing has been evaluated in a few tasks, and more general insights into its performance should be provided. To fill this gap, we propose a CSI estimation method based on BFM, approximating the estimation function with a machine learning model. In addition, to improve the estimation accuracy, we leverage the inter-subcarrier dependency using the BFMs at multiple subcarriers in orthogonal frequency division multiplexing transmissions. Our simulation evaluation reveals that the estimated CSI matches the ground-truth amplitude. Moreover, compared to CSI estimation at each individual subcarrier, the effect of the BFMs at multiple subcarriers on the CSI estimation accuracy is validated.

Distributed inference is an emerging technology that enables inference with cutting-edge machine learning (ML) models such as deep neural networks (DNNs) on resource-constrained devices. However, narrow-band and lossy wireless networks easily create bottlenecks and increase the latency in distributed inference. This study proposes the joint optimization of an ML model and wireless communication parameters (such as transmission rate and retransmission limit) to reduce communication latency while maintaining the accuracy of inference. Our key idea is to utilize the packet-loss tolerance of ML inference that decreased the reliability but reduced communication latency. To this end, the proposed method based on the multi-armed bandit (MAB) algorithm, namely the upper confidence bound (UCB) algorithm, jointly optimizes (i) the wireless communication parameters that control the trade-off between reliability and latency in communications and (ii) the architecture of the ML model that controls the trade-off between accuracy and packet-loss reliance in ML inference. The results of computer simulations using ns3-ai show that the proposed method of joint optimization maintains the accuracy of inference as well as achieves a lower latency than when only the architecture of the ML model or communication parameters are optimized.

The distributed inference framework is an emerging technology for real-time applications empowered by cutting-edge deep machine learning (ML) on resource-constrained Internet of things (IoT) devices. In distributed inference, computational tasks are offloaded from the IoT device to other devices or the edge server via lossy IoT networks. However, narrow-band and lossy IoT networks cause non-negligible packet losses and re-transmissions, resulting in non-negligible communication latency. This study solves the problem of the incremental retransmission latency caused by packet loss in a lossy IoT network. We propose a split inference with no retransmissions (SI-NR) method that achieves high accuracy without any retransmissions, even when packet loss occurs. In SI-NR, the key idea is to train the ML model by emulating the packet loss by a dropout method, which randomly drops the output of hidden units in a neural network layer. This enables the SI-NR system to obtain robustness against packet losses. Our ML experimental evaluation reveals that SI-NR obtains accurate predictions without packet retransmission at a packet loss rate of 60%.

Federated learning (FL) enables a neural network (NN) to be trained using privacy-sensitive data on mobile devices while retaining all the data on their local storages. However, FL asks the mobile devices to perform heavy communication and computation tasks, i.e., devices are requested to upload and download large-volume NN models and train them. This paper proposes a novel unsupervised pre-training method adapted for FL, which aims to reduce both the communication and computation costs through model compression. Since the communication and computation costs are highly dependent on the volume of NN models, reducing the volume without decreasing model performance can reduce these costs. The proposed pretraining method leverages unlabeled data, which is expected to be obtained from the Internet or data repository much more easily than labeled data. The key idea of the proposed method is to obtain a "good" subnetwork from the original NN using the unlabeled data based on the lottery hypothesis. The proposed method trains an original model using a denoising auto encoder with the unlabeled data and then prunes small-magnitude parameters of the original model to generate a small but good subnetwork. The proposed method is evaluated using an image classification task. The results show that the proposed method requires 35% less traffic and computation time than previous methods when achieving a certain test accuracy.