The popularity of concurrent transmissions (CT) has soared after recent studies have shown their feasibility on the four physical layers specified by BLE 5, hence providing an alternative to the use of IEEE 802.15.4 for the design of reliable and efficient low-power wireless protocols. However, to date, the extent to which physical layer properties affect the performance of CT has not yet been investigated in detail. This article fills this gap and provides an extensive study on the impact of the physical layer on CT-based solutions using IEEE 802.15.4 and BLE 5. We first highlight through simulation how the impact of errors induced by relative carrier frequency offsets on the performance of CT highly depends on the choice of the underlying physical layer. We then confirm these observations experimentally on real hardware and with varying environmental conditions through an analysis of the bit error distribution across received packets, unveiling possible techniques to effectively handle these errors. We further study the performance of CT-based data collection and dissemination protocols in the presence of RF interference on a large-scale testbed, deriving insights on how the employed physical layer affects their dependability.

There is an increasing demand for accurate indoor localisation that require minimal infrastructure and wide coverage. All radio technologies commonly used for localisation have trade-offs and there is no clear winner. Accurate systems such as ultra-wideband (UWB) radio, due to their short range of tens of meters, require dense deployment of beacons, incurring huge cost. Such systems out-match the 2.4GHz narrowband technologies with their 10cm accuracy against a few metres with the latter. However, long range technologies such as LoRa, while operating in the 2.4GHz band, can provide a few kilometers of range with fewer beacons in ideal conditions and are therefore cost effective whilst providing greater coverage. We argue in this paper that the desire of having the best of both worlds, i.e., long range, low infrastructure costs, and sub-meter accuracy, can be achieved using embedded machine learning, often referred to as TinyML. To this extent, we run an extensive data collection campaign for two radio technologies i.e., UWB and LoRa, and train machine learning models to improve their localization performance. We then deploy these models on the tiniest of devices powered by ARM Cortex M4 microcontrollers. To the best of our knowledge, we are the first to demonstrate that on-device machine learning can significantly improve localization accuracy. In our case, UWB and LoRa based systems have been shown to improve their localization accuracy by ~20% and ~70% respectively, through learning and essentially calibrating with a more accurate optical camera system without inheriting their weaknesses.

Intelligent, large-scale IoT ecosystems have become possible due to recent advancements in sensing technologies, distributed learning, and low-power inference in embedded devices. In traditional cloud-centric approaches, raw data is transmitted to a central server for training and inference purposes. On the other hand, Federated Learning migrates both tasks closer to the edge nodes and endpoints. This allows for a significant reduction in data exchange while preserving the privacy of users. Trained models, though, may under-perform in dynamic environments due to changes in the data distribution, affecting the model’s ability to infer accurately; this is referred to as concept drift. Such drift may also be adversarial in nature. Therefore, it is of paramount importance to detect such behaviours promptly. In order to simultaneously reduce communication traffic and maintain the integrity of inference models, we introduce FLARE, a novel lightweight dual-scheduler FL framework that conditionally transfers training data, and deploys models between edge and sensor endpoints based on observing the model’s training behaviour and inference statistics, respectively. We show that FLARE can significantly reduce the amount of data exchanged between edge and sensor nodes compared to fixed-interval scheduling methods (over 5x reduction), is easily scalable to larger systems, and can successfully detect concept drift reactively with at least a 16x reduction in latency.

This paper provides details of the collaborative robotics testbed platform that has been developed within the UMBRELLA project. The testbed is part of a larger open Industrial IoT testbed which is currently being deployed in the UK. The aim of the testbed is to permit flexible experimentation using different end devices (including smart city sensing and robot nodes) to evaluate algorithms or new practical application scenarios. For the collaborative robotics testbed this relates to warehouse robotics, which can move pallets of different sizes and shapes. The testbed also includes simulator facilities for validation of algorithms prior to deployment on the robot nodes. The nodes support a rich set of sensors, actuators, and wireless communication technologies. The software architecture is based on Docker containers and ROS2 DDS middleware for flexible and extensible evolution to support future sensors or network technologies. This will be provided as an open testbed to support research, experimentation and evaluation of swarm robotics and other Industrial IoT use-cases.

The popularity of concurrent transmissions (CT) has soared after recent studies have shown their feasibility on the four physical layers specified by BLE 5, hence providing an alternative to the use of IEEE 802.15.4 for the design of reliable and efficient low-power wireless protocols. However, to date, the extent to which physical layer properties affect the performance of CT has not yet been investigated in detail. This paper fills this gap and provides the first extensive study on the impact of the physical layer on CT-based solutions using IEEE 802.15.4 and BLE 5. We first highlight through simulation how the impact of errors induced by de-synchronization and beating on the performance of CT highly depends on the choice of the underlying physical layer. We then confirm these observations experimentally on real hardware through an analysis of the bit error distribution across received packets, unveiling possible techniques to effectively handle these errors. We further study the performance of CT-based flooding protocols in the presence of radio interference on a large-scale, and derive important insights on how the used physical layer affects their dependability.

This paper studies the indoor localisation of WiFi devices based on a commodity chipset and standard channel sounding. First, we present a novel shallow neural network (SNN) in which features are extracted from the channel state information (CSI) corresponding to WiFi subcarriers received on different antennas and used to train the model. The single-layer architecture of this localisation neural network makes it lightweight and easy-to-deploy on devices with stringent constraints on computational resources. We further investigate for localisation the use of deep learning models and design novel architectures for convolutional neural network (CNN) and long-short term memory (LSTM). We extensively evaluate these localisation algorithms for continuous tracking in indoor environments. Experimental results prove that even an SNN model, after a careful handcrafted feature extraction, can achieve accurate localisation. Meanwhile, using a well-organised architecture, the neural network models can be trained directly with raw data from the CSI and localisation features can be automatically extracted to achieve accurate position estimates. We also found that the performance of neural network-based methods are directly affected by the number of anchor access points (APs) regardless of their structure. With three APs, all neural network models proposed in this paper can obtain localisation accuracy of around 0.5 metres. In addition the proposed deep NN architecture reduces the data pre-processing time by 6.5 hours compared with a shallow NN using the data collected in our testbed. In the deployment phase, the inference time is also significantly reduced to 0.1 ms per sample. We also demonstrate the generalisation capability of the proposed method by evaluating models using different target movement characteristics to the ones in which they were trained.

RPL, the de-facto standard for low power and lossy networks, forms multi-hop routing structures between network nodes and a single root. It intrinsically minimises the number of hops across the mesh, however this can result in large distances between adjacent forwarding nodes. Consequently, during forwarding operations, the received signal strength can be close to the receiver sensitivity threshold and result in frequent packet losses due to link fluctuations. RECLAIM, our proposed approach, overcomes this challenge by transmitting routebuilding messages at a reduced power level at first while switching back to the normal transmission power level during the actual data communication phase. This results in more reliable links per hop and ensures link budget above the receiver sensitivity threshold. This however creates more hops and a longer routing path in order to reach destination. Hence, RECLAIM further applies efficient and non-conflicting 6TiSCH scheduling method to coordinate those communication events in a non conflicting manner which do not collide or cause interference to each other, resolving the issue that has traditionally prevented this approach. Our detailed simulation shows that RECLAIM drops 60 times less packets than standard RPL, and achieves 99.9999% packet delivery ratio.

High accuracy localisation technologies exist but are prohibitively expensive to deploy for large indoor spaces such as warehouses, factories, and supermarkets to track assets and people. However, these technologies can be used to lend their highly accurate localisation capabilities to low-cost, commodity, and less-accurate technologies. In this paper, we bridge this link by proposing a technology-agnostic calibration framework based on artificial intelligence to assist such low-cost technologies through highly accurate localisation systems. A single-layer neural network is used to calibrate a less accurate technology using a more accurate one such as Bluetooth Low Energy (BLE) using Ultra-Wideband (UWB) and UWB using a professional motion tracking system. On a real indoor testbed, we demonstrate an improvement in accuracy of approximately 70% for BLE and 50% for UWB. Not only does the proposed approach require a very short measurement campaign, the low complexity of a single-layer neural network makes it ideal for deployment on constrained devices typically used for localisation purposes.

The use of mobile unmanned aerial vehicles (UAV) to collect data from wireless sensor networks has attracted great attention recently. It is often necessary to relate the stream of sensed data to the deployed location of the data producing sensor nodes. This paper addresses the problem of localizing these nodes using a mobile UAV that is unaware of its own location. This, in contrast to existing works, does not assume a GPS equipped UAV. Each sensor node receives beacon packets transmitted by the UAV at random positions and records an RSSI vector. The L1norm distance of two RSSI vectors, which is theoretically proved to be linearly related to the distance between two nodes, is used for ranging. An existing location estimator is then applied to localize target nodes. Extensive simulations in different environments validate high localization accuracy of the proposed algorithm in order of few meters even under noisy channel conditions.

The Time Synchronised Channel Hopping (TSCH) MAC, adopted in popular WirelessHART, ISA100.11a and IEEE802.15.4e, is emerging as the de-facto technique for critical wireless industrial monitoring and control applications. However, much of its benefits are tied to the ability of the protocols and the scheduling techniques to optimally route traffic and orchestrate communication over individual links in a non- conflicting manner. Existing routing and scheduling approaches employ complex algorithms requiring excessive signalling overhead. While some of them can build optimal solutions under static conditions, they lack the agility to adapt to changes in network topology, traffic, and link qualities. To overcome these limitations, we propose BOOST, a protocol that blends TSCH with opportunistic routing and an effortless scheduling technique. BOOST complements the native support of spatial and channel diversity of the former with receiver diversity of the latter, pushing the envelope of network reliability to over 99.99%. Its ultra-lightweight autonomous scheduling mechanism adapts the communication schedule to network changes instantly with almost zero signalling overhead when other solutions struggle with tens of seconds or even minutes of delay with low reliability during this period.

The LoRaWAN based Low Power Wide Area networks aim to provide long-range connectivity to a large number of devices by exploiting limited radio resources. The Adaptive Data Rate (ADR) mechanism controls the assignment of these resources to individual end-devices by a runtime adaptation of their communication parameters when the quality of links inevitably changes over time. This paper provides a detailed performance analysis of the ADR technique presented in the recently released LoRaWan Specifications (v1.1). We show that the ADR technique lacks the agility to adapt to the changing link conditions, requiring a number of hours to days to converge to a reliable and energy-efficient communication state. As a vital step towards improving this situation, we then change different control knobs or parameters in the ADR technique to observe their effects on the convergence time.

The need for low power, long range and low cost connectivity to meet the requirements of IoT applications has led to the emergence of Low Power Wide Area (LPWA) networking technologies. The promise of these technologies to wirelessly connect massive numbers of geographically dispersed devices at a low cost continues to attract a great deal of attention in the academic and commercial communities. Several rollouts are already underway even though the performance of these technologies is yet to be fully understood. In light of these developments, tools to carry out 'what-if analyses' and predeployment studies are needed to understand the implications of choices that are made at design time. While there are several promising technologies in the LPWA space, this paper specifically focuses on the LoRa/LoRaWAN technology. In particular, we present LoRaWANSim, a simulator which extends the LoRaSim tool to add support for the LoRaWAN MAC protocol, which employs bidirectional communication. This is a salient feature not available in any other LoRa simulator. Subsequently, we provide vital insights into the performance of LoRaWAN based networks through extensive simulations. In particular, we show that the achievable network capacity reported in earlier studies is quite optimistic. The introduction of downlink traffic can have a significant impact on the uplink throughput. The number of transmit attempts recommended in the LoRaWAN specification may not always be the best choice. We also highlight the energy consumption versus reliability trade-offs associated with the choice of number of retransmission attempts.

Data prediction in wireless sensor networks replaces the commonly used (periodic) data reporting with a model, updated (infrequently) at the sink to accurately reproduce real data trends. This technique abates up to 99% of application messages; yet, recent work has shown it achieves "only" up to a 7x lifetime improvement when executed atop a mainstream network stack (e.g., CTP + BoX-MAC), as the idle listening and topology maintenance in the latter are ill-suited to the sparse traffic induced by data prediction. This paper presents a novel network stack designed for data prediction, CRYSTAL, that exploits synchronous transmissions to quickly and reliably transmit model updates when these occur (infrequently but often concurrently), and minimizes overhead during the (frequent) periods with no updates. Based on 90-node experiments in the Indriya testbed and with 7 public datasets, we show that CRYSTAL unleashes the full potential of data prediction, achieving per-mille duty cycle with perfect reliability and very small latency.

Data prediction is proposed in wireless sensor networks (WSNs) to extend the system lifetime by enabling the sink to determine the data sampled, within some accuracy bounds, with only minimal communication from source nodes. Several theoretical studies clearly demonstrate the tremendous potential of this approach, able to suppress the vast majority of data reports at the source nodes. Nevertheless, the techniques employed are relatively complex, and their feasibility on resource-scarce WSN devices is often not ascertained. More generally, the literature lacks reports from real-world deployments, quantifying the overall system-wide lifetime improvements determined by the interplay of data prediction with the underlying network. These two aspects, feasibility and system-wide gains, are key in determining the practical usefulness of data prediction in real-world WSN applications. In this paper, we describe derivative-based prediction (DBP), a novel data prediction technique much simpler than those found in the literature. Evaluation with real data sets from diverse WSN deployments shows that DBP often performs better than the competition, with data suppression rates up to 99 percent and good prediction accuracy. However, experiments with a real WSN in a road tunnel show that, when the network stack is taken into consideration, DBP only triples lifetime-a remarkable result per se, but a far cry from the data suppression rates above. To fully achieve the energy savings enabled by data prediction, the data and network layers must be jointly optimized. In our testbed experiments, a simple tuning of the MAC and routing stack, taking into account the operation of DBP, yields a remarkable seven-fold lifetime improvement w.r.t. the mainstream periodic reporting.

Time series forecasting aims at improving energy efficiency in wireless sensor networks (WSNs) by reducing the amount of data traffic. One such technique has each node generate a model that predicts the sampled data. When the actual, sensed data deviates from the model, a new model is generated and transmitted to the sink. Reductions in application data traffic as high as two orders of magnitude can be achieved. However, our experience in applying such forecasting in a real world deployment shows that the actual lifetime improvement is significantly less due to networking overheads. The study reported here reveals that careful, coordinated network parameter tuning can leverage the reduced traffic of forecasting techniques to increase lifetime without compromising application performance.

Model-driven data acquisition techniques aim at reducing the amount of data reported, and therefore the energy consumed, in wireless sensor networks (WSNs). At each node, a model predicts the sampled data; when the latter deviate from the current model, a new model is generated and sent to the data sink. However, experiences in real-world deployments have not been reported in the literature. Evaluation typically focuses solely on the quantity of data reports suppressed at source nodes: the interplay between data modeling and the underlying network protocols is not analyzed. In contrast, this paper investigates in practice whether i) model-driven data acquisition works in a real application; ii) the energy savings it enables in theory are still worthwhile once the network stack is taken into account. We do so in the concrete setting of a WSN-based system for adaptive lighting in road tunnels. Our novel modeling technique, Derivative-Based Prediction (DBP), suppresses up to 99% of the data reports, while meeting the error tolerance of our application. DBP is considerably simpler than competing techniques, yet performs better in our real setting. Experiments in both an indoor testbed and an operational road tunnel show also that, once the network stack is taken into consideration, DBP triples the WSN lifetime-a remarkable result per se, but a far cry from the aforementioned 99% data suppression. This suggests that, to fully exploit the energy savings enabled by data modeling techniques, a coordinated operation of the data and network layers is necessary.