This paper studies an important rate allocation problem that arises in many networked and distributed systems: steady-state traffic rate allocation from multiple sources to multiple service nodes when both (i) the access-path delay on each source-node route is rate-dependent (capacity-constrained) and convex, and (ii) each service node (also capacity-constrained) experiences a load-dependent queueing delay driven by aggregate load from all sources. We show that the resulting flow-weighted end-to-end delay minimization is a convex program, yielding a global system-optimal solution characterized by KKT conditions that equalize total marginal costs (a path marginal access term plus a node congestion price) across all utilized routes. This condition admits a Wardrop-type interpretation: for each source, all utilized options equalize total marginal cost, while any option with strictly larger total marginal cost receives no flow. Building on this structure, we develop a lightweight distributed pricing-based algorithm in which each service node locally computes and broadcasts a scalar congestion price from its observed aggregate load, while each source updates its traffic split by solving a small separable convex allocation problem under the advertised prices. Numerical illustrations demonstrate convergence of the distributed iteration to the centralized optimum and highlight the trade-offs induced by jointly modeling access and service congestion.

We present a comprehensive study on applying machine learning to detect distributed Denial of service (DDoS) attacks using large-scale Internet of Things (IoT) systems. While prior works and existing DDoS attacks have largely focused on individual nodes transmitting packets at a high volume, we investigate more sophisticated futuristic attacks that use large numbers of IoT devices and camouflage their attack by having each node transmit at a volume typical of benign traffic. We introduce new correlation-aware architectures that take into account the correlation of traffic across IoT nodes. We extensively analyze the proposed architectures by evaluating five different neural network models trained on a dataset derived from a 4060-node real-world IoT system. We observe that long short-term memory (LSTM) and a transformer-based model, in conjunction with the architectures that use correlation information of the IoT nodes, provide higher performance (in terms of F1 score and binary accuracy) than the other models and architectures, especially when the attacker camouflages itself by following benign traffic distribution on each transmitting node. For instance, by using the LSTM model, the distributed correlation-aware architecture gives 81% F1 score for the attacker that camouflages their attack with benign traffic as compared to 35% for the architecture that does not use correlation information. We validate the effectiveness of our proposed detection mechanism by implementing it on a real testbed. We also investigate the performance of heuristics for selecting a subset of nodes to share their data for correlation-aware architectures to meet resource constraints.

In recent years, there has been a significant surge in the deployment of Internet of Things (IoT) devices, which has consequently escalated security threats, notably Distributed Denial of Service (DDoS) attacks. Our prior research developed an LSTM-based framework for detecting futuristic DDoS attacks but largely relied on simulated datasets [1]. To bridge this gap, we designed a Raspberry Pi (RPi) testbed that mimics the complexities of large-scale IoT networks. This setup allows us to simulate realistic DDoS attacks originating from IoT devices and evaluate the effectiveness of various DDoS detection techniques. Specifically, using this RPi testbed, we validated the effectiveness of our LSTM-based framework in identifying futuristic DDoS attacks, observing an F1 score ranging between 0.8 and 0.86 depending on the aggressiveness of the DDoS attack.

In this article, we propose a lightweight blockchain-inspired framework-Magnum-as a magazine of transfer learning models in blocks. We propose the storage of these blocks on proximal fog nodes to simplify access to pretrained base models by industrial plants to tune them before deployment. We design Magnum for B5G-enabled scenarios to reduce the block transfer time. We formulate a demand-centric distribution scheme to further reduce the search and access time by adopting a nonlinear program model and solving it using the branch-and-bound method. Through extensive experiments and comparison with state-of-the-art solutions, we show that Magnum retains the accuracy of the models and present its feasibility with a maximum CPU and memory usage of 80% and 6%, respectively. Additionally, while Magnum requires a maximum of 10 s for writing models as large as 17 Mb on the blocks, it requires 16 ms for fetching the same.