In this paper, a dataset of IoT network traffic is presented. Our was generated by utilising the Gotham testbed, an emulated large-scale Internet Things (IoT) designed to provide realistic and heterogeneous environment for security research. The testbed includes 78 devices operating on various protocols, including MQTT, CoAP, RTSP. Network captured in Packet Capture (PCAP) format using tcpdump, both benign malicious were recorded. Malicious through scripted attacks, covering variety attack...

The vast increase of Internet Things (IoT) technologies and the ever-evolving attack vectors have increased cyber-security risks dramatically. A common approach to implementing AI-based Intrusion Detection Systems (IDSs) in distributed IoT systems is a centralised manner. However, this may violate data privacy prohibit IDS scalability. Therefore, intrusion detection solutions ecosystems need move towards decentralised direction. Federated Learning (FL) has attracted significant interest...

Data integrity becomes paramount as the number of Internet Things (ioT) sensor deployments increases. Sensor data can be altered by benign causes or malicious actions. Mechanisms that detect drifts and irregularities prevent disruptions bias in state an IoT application. This paper presents LE3D, ensemble framework drift estimators capable detecting abnormal behaviours. Working collaboratively with surrounding ioT devices, type (natural/abnormal) also identified reported to end-user. The...

Federated Learning (FL) is fast becoming one of the most prevalent distributed learning techniques focused on privacy preservation and communication efficiency for large-scale Internet Things (IoT) deployments. FL a approach to training models devices. Since local data remains on-device, through network reduced. However, in IoT environments or resource constrained networks, typical approaches significantly suffer performance due longer times. In this paper, we propose two methods further...

Finding on-street parking in congested urban areas is a challenging chore that most drivers worldwide dislike. Previous vehicle traffic studies have estimated around thirty percent of vehicles travelling inner city are made up searching for vacant space. While there hardware sensor based solutions to monitor occupancy real-time, instrumenting and maintaining such wide system substantial investment. In this paper, novel activity detection method, called ParkUs, introduced tested with the aim...

Deep learning (DL) models have emerged as a promising solution for the Internet of Things (IoT). However, due to their computational complexity, DL consume significant amounts energy, which can rapidly drain battery and compromise performance IoT devices. For sustainable operation, we consider an edge device with rechargeable energy harvesting (EH) capabilities. In addition stochastic nature ambient source, rate is often insufficient meet inference requirements, causing drastic degradation...

Recent studies show that a key contributor to congestion and increased CO2 emissions within cities are drivers searching (or cruising) find vacant on-street parking space. It has been shown approximately (depending on the city) 20-30% of vehicles in congested urban areas were cruising space with search time varying order several minutes. In city Bristol alone, we have shown, using our collected trip publicly available census data over 790 metric tons is generated every year due cruising. At...

Over the last half century, proportion of humans living in cities has dramatically risen from around a third to just over half. As continue rise popularity, demand for basic services such as transportation increases. The automobile been dominant method inner city many across globe, resulting increased congestion and air pollution. rises, so does number vehicles, which leads greater competition publicly available parking spaces. Use land can be an inefficient use space, it is expensive, both...

A dataset of street light images is presented. Our consists $\sim350\textrm{k}$ images, taken from 140 UMBRELLA nodes installed in the South Gloucestershire region UK. Each node on pole a lamppost and equipped with Raspberry Pi Camera Module v1 facing upwards towards sky bulb. collects an image at hourly intervals for 24h every day. The data collection spans period six months. logged as single entry along Global Positioning System (GPS) coordinates lamppost. All entries have been...

Data-enabled cities are recently accelerated and enhanced with automated learning for improved Smart Cities applications. In the context of an Internet Things (IoT) ecosystem, data communication is frequently costly, inefficient, not scalable lacks security. Federated Learning (FL) plays a pivotal role in providing privacy-preserving efficient Machine (ML) frameworks. this paper we evaluate feasibility FL Street Light Monitoring application. evaluated against benchmarks centralised (fully)...

Edge computing is rapidly changing the IoT-Cloud landscape. Various testbeds are now able to run multiple Docker-like containers developed and deployed by end-users on edge devices. However, this capability may allow an attacker deploy a malicious container host compromise it. This paper presents dataset based Linux Auditing System, which contains benign activity. We two scenarios, denial of service privilege escalation attack, where adversary uses device. Furthermore, we user in parallel...

We apply Federated Learning (FL) to the problem of indoor localisation in a real-world multiple residential house scenario. Fingerprinting Received Signal Strength Indicator (RSSI) was used as method. show that, given minimal amount fine-tuning allowed by constraint on size gradient step fit round FL, shared model learned this way has strong performance all houses and is stable with respect randomness weight initialisation. Not unexpectedly, inferior an individual learning approach....

A dataset of sensor measurements is presented. Our contains discrete 8 IoT devices located in various places a research lab at the University Bristol. Nordic nRF52840 DK periodically collects environmental data, such as temperature, humidity, pressure, gas, room light intensity, accelerometer; including also measurement quality indicator. The were taken every 10 seconds over six-month period between February and September 2022. In addition, we provide Received Signal Strength Indicator...

The way we travel is changing rapidly and Cooperative Intelligent Transportation Systems (C-ITSs) are at the forefront of this evolution. However, adoption C-ITSs introduces new risks challenges, making cybersecurity a top priority for ensuring safety reliability. Building on premise, paper an envisaged Cybersecurity Centre Excellence (CSCE) designed to bolster researching, testing, evaluating C-ITSs. We explore design, functionality, challenges CSCE's testing facilities, outlining...

Federated learning (FL) systems face performance challenges in dealing with heterogeneous devices and non-identically distributed data across clients. We propose a dynamic global model aggregation method within Asynchronous Learning (AFL) deployments to address these issues. Our scores adjusts the weighting of client updates based on their upload frequency accommodate differences device capabilities. Additionally, we also immediately provide an updated clients after they local models reduce...

Federated learning (FL) systems face performance challenges in dealing with heterogeneous devices and non-identically distributed data across clients. We propose a dynamic global model aggregation method within Asynchronous Learning (AFL) deployments to address these issues. Our scores adjusts the weighting of client updates based on their upload frequency accommodate differences device capabilities. Additionally, we also immediately provide an updated clients after they local models reduce...

With the ever-increasing reliance on digital networks for various aspects of modern life, ensuring their security has become a critical challenge. Intrusion Detection Systems play crucial role in network security, actively identifying and mitigating malicious behaviours. However, relentless advancement cyber-threats rendered traditional/classical approaches insufficient addressing sophistication complexity attacks. This paper proposes novel 3-stage intrusion detection system inspired by...

The growing demand for intelligent applications beyond the network edge, coupled with need sustainable operation, are driving seamless integration of deep learning (DL) algorithms into energy-limited, and even energy-harvesting end-devices. However, stochastic nature ambient energy sources often results in insufficient harvesting rates, failing to meet requirements inference causing significant performance degradation energy-agnostic systems. To address this problem, we consider an on-device...

Wireless vehicular ad-hoc networks comprised solely of city taxis are investigated for their ability to deliver data across an urban environment. Openly available taxi trace datasets Rome (Italy) and San Francisco (USA) combined with respective building footprint road network topology from OpenStreetMap, generate a realistic systems level model V2V network. Analysis LOS NOLOS constraints on wireless transmission range suggests minimum threshold 50m is applicable ensure in over 90% cases....

The Open Radio Access Network (O-RAN) is a burgeoning market with projected growth in the upcoming years. RAN has highest CAPEX impact on network and, most importantly, consumes 73% of its total energy. That makes it an ideal target for optimisation through integration Machine Learning (ML). However, energy consumption ML frequently overlooked such ecosystems. Our work addresses this critical aspect by presenting FROST – Flexible Reconfiguration method Online System Tuning solution...

A dataset of street light images is presented. Our consists $\sim350\textrm{k}$ images, taken from 140 UMBRELLA nodes installed in the South Gloucestershire region UK. Each node on pole a lamppost and equipped with Raspberry Pi Camera Module v1 facing upwards towards sky bulb. collects an image at hourly intervals for 24h every day. The data collection spans period six months. logged as single entry along Global Positioning System (GPS) coordinates lamppost. All entries have been...

Data integrity becomes paramount as the number of Internet Things (IoT) sensor deployments increases. Sensor data can be altered by benign causes or malicious actions. Mechanisms that detect drifts and irregularities prevent disruptions bias in state an IoT application. This paper presents LE3D, ensemble framework drift estimators capable detecting abnormal behaviours. Working collaboratively with surrounding devices, type (natural/abnormal) also identified reported to end-user. The proposed...