Cisco Systems (Canada)

A lack of accessible data has historically restricted malware analysis research, and practitioners have relied heavily on datasets provided by industry sources to advance. Existing public datasets are limited by narrow scope - most include files targeting a single platform, have labels supporting just one type of malware classification task, and make no effort to capture the evasive files that make malware detection difficult in practice. We present EMBER2024, a new dataset that enables holistic evaluation of malware classifiers. Created in collaboration with the authors of EMBER2017 and EMBER2018, the EMBER2024 dataset includes hashes, metadata, feature vectors, and labels for more than 3.2 million files from six file formats. Our dataset supports the training and evaluation of machine learning models on seven malware classification tasks, including malware detection, malware family classification, and malware behavior identification. EMBER2024 is the first to include a collection of malicious files that initially went undetected by a set of antivirus products, creating a "challenge" set to assess classifier performance against evasive malware. This work also introduces EMBER feature version 3, with added support for several new feature types. We are releasing the EMBER2024 dataset to promote reproducibility and empower researchers in the pursuit of new malware research topics.

Musculoskeletal tumors present a diagnostic challenge due to their rarity, histological diversity, and overlapping imaging features. Accurate characterization is essential for effective treatment planning and prognosis, yet current diagnostic workflows rely heavily on invasive biopsy and subjective radiologic interpretation. This review explores the evolving role of radiogenomics and machine learning in improving diagnostic accuracy for bone and soft tissue tumors. We examine integrating quantitative imaging features from MRI, CT, and PET with genomic and transcriptomic data to enable non-invasive tumor profiling. AI-powered platforms employing convolutional neural networks (CNNs) and radiomic texture analysis show promising results in tumor grading, subtype differentiation (e.g., Osteosarcoma vs. Ewing sarcoma), and predicting mutation signatures (e.g., TP53, RB1). Moreover, we highlight the use of liquid biopsy and circulating tumor DNA (ctDNA) as emerging diagnostic biomarkers, coupled with point-of-care molecular assays, to enable early and accurate detection in low-resource settings. The review concludes by discussing translational barriers, including data harmonization, regulatory challenges, and the need for multi-institutional datasets to validate AI-based diagnostic frameworks. This article synthesizes current advancements and provides a forward-looking view of precision diagnostics in musculoskeletal oncology.

The rapid evolution of IoT-driven ICT products demands a holistic safety framework that addresses both inherent hazards and cybersecurity threats. While the Dynamic Hazard-Based Safety Engineering (D-HBSE) which been proposed in 2024 could effectively manage safety risks associated with the internal hazards of product itself, it lacks explicit integration of cybersecurity considerations, leaving critical gaps in mitigating safety incidents caused by intentional malicious attacks. This paper completes the “Control-Oriented Safety” trilogy by proposing an integrated framework that harmonizes Hazard Analysis and Risk Assessment (HARA) with Threat Analysis and Risk Assessment (TARA) build on D-HBSE context. It also introduces System-Theoretic Process Analysis (STPA), to demonstrate how safety and cybersecurity co-assessments systematically address risk arising from both endogenous hazards and exogenous threats. Finally, a case study on a server fan speed anomaly illustrates the framework's effectiveness in mitigating risks such as overheating caused by sensor faults or firmware exploits. This work establishes a modern ICT product safety assessment framework aligned with the philosophy of Composition-Based Safety View (CBSV), ensuring comprehensive compliance in the IoT era.

The Serial Peripheral Interface (SPI) protocol plays a crucial role in wearable and IoT devices, enabling high-speed communication between microcontrollers and peripherals such as sensors, displays, and connectivity modules. With the increasing complexity of modern devices and system-on-chip (SoC) designs, robust verification methods are essential to ensure functionality and reliability. This paper utilizes the Universal Verification Methodology (UVM) to develop a scalable and reusable testbench for SPI verification. The process encompasses test planning, simulation, emulation, and top-level verification to validate multi-slave coordination and error-handling scenarios. The results demonstrate the critical importance of UVM in ensuring the performance and dependability of SPI in advanced electronics, contributing to the reliable integration of the protocol in future devices. The verification results demonstrated a functional coverage of 83.33% and 100% assertion coverage, confirming our approach’s robustness. Analysis of the uncovered functional bins revealed that specific edge cases, such as timing violations and multi-slave arbitration conflicts, require additional test scenarios for full verification. Furthermore, our testbench successfully identified and handled critical fault conditions, such as clock jitter, bus contention, and framing errors, ensuring reliable SPI operation in real-world deployments. These findings highlight the effectiveness of UVM-based verification in improving the reliability and robustness of SPI communication in modern low-power, resource-constrained embedded systems.

Intent Based Networking (IBN) is a novel paradigm that aims to create a closed-loop automation towards building a self-governed network. An inherent part of IBN is the network assurance, which tries to automatically detect any performance drift and propose a remedy solution. Following this trend, in this paper we propose a novel IBN -enabled network assurance approach in a data center setting, in order to predict key resource utilization metrics and to help network administrators to proactively take corrective actions when needed. Specifically, we propose an incremental learning approach for data center assurance, designed to accommodate the dynamic nature of infrastructures. Experimental findings obtained from real-world datasets substantiate the effectiveness of our approach, demonstrating its ability to accurately forecast resource utilization even amidst highly dynamic environments.

Cloud Computing has revolutionized the information technology world and the application offering over the last two decades. At the same time recent trends in Network Function Virtualization (NFV) and Software-Defined Wide Area Networks (SD-WAN) and the combination of those with the Cloud paradigm has allowed an unprecedented shift of enterprise networking services towards the Public Cloud. Even though this network evolutionary approach brings many benefits, it still presents many drawbacks as well. The performance stability and service continuity over a black box Public Cloud infrastructure can hinder the formal service guarantees that many new emerging applications may have. To this end, in this paper, we aim to shed light on the overall performance achieved when deploying coast-to-coast and intercontinental Service Function Chains (SFCs) that interconnect geographically distributed enterprise branches over the Amazon Web Services (AWS) infrastructure. In particular, we investigate the impact of region, Virtual Machine (VM) instance, time of the day and day of the week in the overall throughput and delay attained. The obtained results show the strengths and weaknesses of entirely relying on the AWS infrastructure to offer networking services by investigating possible hidden performance bottlenecks.

ElGamal system was introduced by Taher ElGamal in his landmark paper in 1985 [1]. ElGamal is an important asymmetric cryptosystem that is based on the discrete logarithm problem. In brief, this cryptosystem relies on the difficulty of computing discrete logarithms over finite fields [1]. Towards solving discrete logarithms, Shor demonstrated the existence of a polynomial algorithm that can solve these logarithms on quantum machines [2]. In this paper we propose the Convolution Power Cryptosystem (CPC-H2); an extension to the ElGamal cryptosystem that is based on a novel hard problem. CPC-H2 is designed to defend against a Chosen Ciphertext Attack. We show that CPC-H2 is less complicated and has a higher security level than the Number Theory Research Unit (NTRU).

Greenhouses offer the ability to cultivate various crops in what would otherwise be harsh and unfavorable environments. Networked Control Systems (NCSs) can be utilized within greenhouses, to monitor and control various metrics, such as temperature, humidity, and soil moisture. While the utilization of wireless sensors offers a high degree of flexibility, it adds the risk of Access Point (AP) failure. This paper investigates the issue of AP failure in a Greenhouse NCS and proposes a fault-tolerant model. Single as well as double AP failures are considered. All scenarios in the fault-tolerant model are simulated using the Riverbed Modeler and system performance is evaluated in terms of the Packet Loss Rate (PLR). It is shown that, for all possible AP failures in the fault model, the proposed fault-tolerant architecture succeeds in meeting system requirements.

This paper proposes a novel cryptosystem (PMTRU) that is a variant of the NTRU public key cryptosystem. PMTRU combines the advantages of the NTRU and its other variant, MATRU. We prove analytically that our cryptosystem improves the speed of encryption and decryption procedures, and significantly increases the security level of the private key. Compared to NTRU and MATRU, PMTRU greatly improves resistance against brute-force and lattice-based attacks, while decreasing the memory space required to register the public key and ciphertexts.

Network Function Virtualization (NFV) along with Service Function Chaining (SFC) has proliferated the way that network functions and services are introduced, offered, and deployed. In particular, NFV and SFC are the basic components of the so called network softwarization era, promoting significant cost reductions and agility. Nonetheless, a constant debate exist regarding the network performance achieved and the holistic establishment of NFV as the key and sole component of the network infrastructure. To this end, this paper tries to shed light in remedy solutions when chaining Virtualized Network Functions (VNFs) internally on a standard x86 server using the Single-Root Input Output Virtualization (SR-IOV) technology, in order to maximize the throughput and delay achieved.