In traditional centralized machine learning, transmitting raw data to a cloud center incurs high communication costs and raises privacy concerns. This is particularly challenging in mobile edge environments, where devices are dynamic and resource-constrained. Federated Learning (FL) addresses these issues by allowing devices to train models locally and upload parameters to a central server without sharing local data. However, limited wireless channel resources and dynamic transmission performance make communication overhead a major bottleneck of FL in mobile edge environments. Concerning this challenge, this survey provides a comprehensive summary of methods to improve communication efficiency in FL, focusing on: 1) minimizing communication complexity to reduce total transmission volume, 2) scheduling resources appropriately to improve training efficiency, 3) utilizing over-the-air computation (OTA) to integrate computation into communication for accommodating the computation/communication characteristics of FL in mobile edge environments. Thus, this work analyzes research from the perspective of convergence and data heterogeneity to reduce communication rounds by optimizing algorithm performance. We hope that this survey could offer insights into communication-efficient FL for future research.

In RFID-enabled applications, when a tag is put into use and associated with a specific object, the category-related information (e.g., the brands of clothes) about this object might be preloaded into the tag's memory as required. Since such information reflects the category attributes, all tags in the same category carry the identical category information. To collect this information, we do not need to repeatedly interrogate each tag; one tag's response in a category is sufficient. In this paper, we investigate the new problem of category information collection in a multi-category RFID system, which is referred to as information sampling. We propose an efficient two-phase sampling protocol (TPS). By quickly zooming into a category and isolating a tag from this category, TPS is able to sample a category by broadcasting only 7.5-bit polling vector (very efficient when compared to the 96-bit tag ID). We theoretically analyze the protocol performance and discuss the optimal parameter settings that minimize the overall execution time. Extensive simulations show that TPS outperforms the benchmark, greatly improving the sampling performance.

In RFID systems, tag grouping is to inform each tag in the interrogator's RF field which group it belongs to by assigning them a group ID. With this information, the reader can write the same data into a group of tags in one transmission by taking their shared group ID as the destination address, which plays a vital role in improving inventory efficiency. Existing work proposes some efficient protocols but suffers from two defects: slot waste and data redundancy. To address these problems, we in this paper propose a two-phase grouping scheme that improves the grouping efficiency via slot reuse as well as data compression. For slot waste, we try to reconcile collision slots and turn some of them into useful again by re-assigning a new slot to collision tags, which increases the number of useful slots and reduces slot waste. For data redundancy, we encode group IDs with lossless coding algorithms and reduce the data redundancy caused by repetitive transmission of group IDs. Two coding methods, Huffman coding and arithmetic coding, are adopted to shorten the length of the bit vector comprised by group IDs. Theoretical analyses and extensive simulations show that our best protocol improves the time efficiency by 51.9% compared with the state-of-the-art.

Airlines commonly need to take into consideration maximizing their profit while designing the hub-and-spoke network to obtain more market share and promote healthy development of aviation industry. Hence, in this article, we study the problem of multiple-allocation HUb and spoke network design for ROuting flight flows to maximize airline profit utility (HURO). That is, given a set of airport nodes, a set of flight flows with known origin positions and destination positions, finding a limited number of hub edges to transfer flows and determining routing allocation mode considering customer preference such that the overall transportation profit utility is maximized. To address HURO problem, we first consider a relaxed version of HURO (HURO-R for short). We prove that HURO-R falls into the realm of maximizing a submodular set function subject to a cardinality constraint, and propose an algorithm with a constant approximation ratio. Next, we design a two-level algorithm framework with a constant approximation ratio to address HURO. Besides, we consider variants of HURO, HURO-C and HURO-RU, and design approximation algorithms to address them. We conduct simulation experiments on standard dataset and field experiments to verify our theoretical findings. The results shows that our proposed algorithm can outperform other comparison algorithms by 75.28 percent.

Radio Frequency IDentification (RFID) technology has been widely used in missing tag identification to reduce the economic loss caused by theft. Although many advanced works have been proposed, they cannot work properly in an open RFID system with unexpected tags. That is because the unexpected tags may reply to the reader when it is the missing tags’ turn, such that the replies are mistakenly treated as the missing tags’, leading to failure to detect the missing tags. To solve the problem, this article proposes an order-based missing tag identification protocol (OMTI) that efficiently identifies all missing tags even with the presence of unexpected tags. The key of OMTI is that we dynamically assign each tag an exclusive slot rather than use random hashes in traditional design. Namely, OMTI off-line serializes all known tags in advance by assigning each of them a unique and continuous slot, on-line identifies missing tags by checking tags’ responses in each slot, and dynamically updates the slot assignment to ensure the high efficiency of the next execution. Besides, we enhance OMTI to make it embrace newly added tags and spread to the multi-reader case from the single one. The simulation results demonstrate that OMTI can significantly reduce the identification time compared to the state-of-the-art.

The increasing complexity and computational demands of deep neural networks (DNNs) pose significant challenges and deployment on resource-constrained devices due to substantial latency and considerable energy consumption. Multi-exit DNNs have emerged as a promising solution, enabling simple tasks to exit early at intermediate network layers, thereby reducing inference latency and improving efficiency. However, relying solely on the computational capacity of end devices is often insufficient to meet performance requirements. Edge computing, by offloading part of the model’s computation to edge servers, has become a key solution to address this issue. Despite the potential of multi-exit DNNs in edge computing environments, two major challenges remain: model partitioning and resource allocation on edge servers. Existing research often focuses on model partitioning strategies or resource allocation strategies in isolation, neglecting the mutual optimization between the two. Moreover, energy consumption, a critical performance metric, is frequently overlooked in the optimization process. To address these issues, this paper proposes a joint optimization framework for model partitioning and resource allocation, integrating multi-exit DNNs and incorporating a deep reinforcement learning (DRL)-based optimization algorithm. Experimental results demonstrate that the proposed method significantly reduces inference costs and enhances system performance.