Optimizing resource orchestration in network slicing is essential for the performance of diverse applications in 5G edge-cloud networks. This paper introduces a novel approach utilizing multi-agent reinforcement learning (MARL) with a dueling double deep Q-network (D3QN) to efficiently manage dynamic resource provisioning to the different traffic flows. We model a network slicing environment with applications generating stochastic resource demands, simulating real-world virtual network patterns over physical infrastructure. Our MARL-based scheme adapts to the varying needs of traffic flows, balancing compute and memory resource allocation under limited information. Comparative analysis demonstrates the superiority of our approach over traditional static methods, particularly for ultra-reliable low-latency communication (URLLC) traffic flows, by minimizing latency and enhancing resource efficiency. The effectiveness of the proposed framework is validated through extensive simulations, which demonstrate up to 45% higher average utility for URLLC traffic flows and 18% improvement in overall resource efficiency compared with baseline strategies. These results confirm that the framework can simultaneously ensure stringent service requirements and enhance system-wide performance in Next-Generation networks.

Achieving ultra-reliable low-latency communication (URLLC) in next-generation radio access networks (RAN) is crucial for mission-critical applications, such as industrial robotics and remote healthcare. However, high traffic loads in RAN environments may lead to resource contention and overload, jeopardizing latency and reliability requirements. To address this, Open RAN (O-RAN) architecture provides a flexible, softwaredefined framework that can manage admission control and resource allocation strategies. This paper introduces a joint admission control and resource provisioning framework tailored for O-RAN environments, using a constrained reinforcement learning model to dynamically balance user admissions and allocate resources to those admitted. By selectively granting service requests and efficiently managing resources, our approach mitigates latency violations and improves energy efficiency under high traffic conditions. Simulation results indicate substantial performance gains over traditional methods, demonstrating the potential of reinforcement learning to optimize URLLC performance in resource-constrained NextG networks.

User equipments (UEs) sharing a RAN slice can generate bursty or unpredictable traffic that degrades the quality of service (QoS) for other slice members. We present triage slicing, a control mechanism that detects disruptive UEs using an online long short term memory (LSTM) predictor and migrates them to a monitored slice with constrained, risk proportional resources. Implemented on the Open AI Cellular (OAIC) testbed, triage slicing rapidly contains disruptive UEs, preserves benign QoS, and incurs minimal control overhead.

Aerial Acoustic Communication (AAC) has attracted much attention recently due to its ubiquitous device support. However, the transmission rate has largely throttled the boom of AAC-enabled applications. In this paper, we propose a high-speed AAC system that modulates messages over the entire bandwidth but is unobtrusive to end users. We achieve high bandwidth efficiency through a dynamic modulation mechanism. Specifically, we propose a high-order phase-based modulation mechanism on existing audio channels. Additionally, we leverage the auditory masking effect to embed data streams over the unused frequency band. Through a sophisticated power analysis and frequency allocation scheme, the changes made by these two modulation techniques on existing audio can be imperceptible to human ears. Furthermore, we utilize the loose orthogonal modulation on the inaudible channel to further boost the transmission rate. Our system prototype reveals that we can achieve up to 2 kbps link rate.

MOTIVATION: Accurately predicting cancer survival is crucial for helping clinicians to plan appropriate treatments, which largely improves the life quality of cancer patients and spares the related medical costs. Recent advances in survival prediction methods suggest that integrating complementary information from different modalities, e.g. histopathological images and genomic data, plays a key role in enhancing predictive performance. Despite promising results obtained by existing multimodal methods, the disparate and heterogeneous characteristics of multimodal data cause the so-called modality gap problem, which brings in dramatically diverse modality representations in feature space. Consequently, detrimental modality gaps make it difficult for comprehensive integration of multimodal information via representation learning and therefore pose a great challenge to further improvements of cancer survival prediction. RESULTS: To solve the above problems, we propose a novel method called cross-aligned multimodal representation learning (CAMR), which generates both modality-invariant and -specific representations for more accurate cancer survival prediction. Specifically, a cross-modality representation alignment learning network is introduced to reduce modality gaps by effectively learning modality-invariant representations in a common subspace, which is achieved by aligning the distributions of different modality representations through adversarial training. Besides, we adopt a cross-modality fusion module to fuse modality-invariant representations into a unified cross-modality representation for each patient. Meanwhile, CAMR learns modality-specific representations which complement modality-invariant representations and therefore provides a holistic view of the multimodal data for cancer survival prediction. Comprehensive experiment results demonstrate that CAMR can successfully narrow modality gaps and consistently yields better performance than other survival prediction methods using multimodal data. AVAILABILITY AND IMPLEMENTATION: CAMR is freely available at https://github.com/wxq-ustc/CAMR. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Cancer survival prediction can greatly assist clinicians in planning patient treatments and improving their life quality. Recent evidence suggests the fusion of multimodal data, such as genomic data and pathological images, is crucial for understanding cancer heterogeneity and enhancing survival prediction. As a powerful multimodal fusion technique, Kronecker product has shown its superiority in predicting survival. However, this technique introduces a large number of parameters that may lead to high computational cost and a risk of overfitting, thus limiting its applicability and improvement in performance. Another limitation of existing approaches using Kronecker product is that they only mine relations for one single time to learn multimodal representation and therefore face significant challenges in deeply mining rich information from multimodal data for accurate survival prediction. RESULTS: To address the above limitations, we present a novel hierarchical multimodal fusion approach named HFBSurv by employing factorized bilinear model to fuse genomic and image features step by step. Specifically, with a multiple fusion strategy HFBSurv decomposes the fusion problem into different levels and each of them integrates and passes information progressively from the low level to the high level, thus leading to the more specialized fusion procedure and expressive multimodal representation. In this hierarchical framework, both modality-specific and cross-modality attentional factorized bilinear modules are designed to not only capture and quantify complex relations from multimodal data, but also dramatically reduce computational complexity. Extensive experiments demonstrate that our method performs an effective hierarchical fusion of multimodal data and achieves consistently better performance than other methods for survival prediction. AVAILABILITY AND IMPLEMENTATION: HFBSurv is freely available at https://github.com/Liruiqing-ustc/HFBSurv. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.