Standing on the Shoulders of Giants: AI-Driven Calibration of Localisation Technologies

Khan, Aftab; Farnham, Tim; Kou, Roget; Raza, Usman; Premalal, Thajanee; Stanoev, Aleksandar; Thompson, William

doi:10.1109/globecom38437.2019.9013500

Cited by 8 publications

(5 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given that these devices should provide readings as accurate as possible regarding the recording of the location from which the interpersonal distance could be deduced, the technologies not meeting these requirements were not considered, following the specificities of the scientific literature 42,49,50 . As a result, the real‐time locating system (RTLS) based on ultra‐wideband (UWB) was the most adequate technology for the purpose of this research 49,51‐55 . Based on both the comparisons conducted by other authors 56 and our analysis and available resources, a Decawave ® 57 equipment was chosen as there were experimental precedents in several spaces and situations 58‐65 .…”

Section: Methodsmentioning

confidence: 99%

“…42,49,50 As a result, the real-time locating system (RTLS) based on ultra-wideband (UWB) was the most adequate technology for the purpose of this research. 49,[51][52][53][54][55] Based on both the comparisons conducted by other authors 56 and our analysis and available resources, a Decawave ®57 equipment was chosen as there were experimental precedents in several spaces and situations. [58][59][60][61][62][63][64][65] This equipment is an ultra-wideband network constituted by fixed nodes (anchors) and moving nodes (tags).…”

Section: Detection Devicesmentioning

confidence: 99%

See 1 more Smart Citation

Distances of transmission risk of COVID‐19 inside dwellings and evaluation of the effectiveness of reciprocal proximity warning sounds

2020

View full text Add to dashboard Cite

One of the main modes of transmission and propagation of COVID‐19 (SARS‐CoV‐2) is the direct contact with respiratory droplets transmitted among individuals at a certain distance. There are indoor spaces, such as dwellings, in which the transmission risk is high. This research aims to record and analyze risk close contacts in this scope, experimentally assessing the effectiveness of using electronic proximity warning sound devices or systems. For this purpose, the methodology is based on monitoring the location of the occupants of a dwelling. Then, the days in which a proximity warning sound system is installed and activated are compared to the days in which the system is not activated. The results stressed the significant reduction of time and number of close contacts among individuals when the warning was activated. Regarding the relation between the number and the duration of close contacts, together with the reductions mentioned, the possibility of making certain predictions based on the distributions obtained is proved. All this contributes to the progress in the prevention of COVID‐19 transmission because of close contacts in dwellings.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Detection Devicesmentioning

confidence: 99%

Distances of transmission risk of COVID‐19 inside dwellings and evaluation of the effectiveness of reciprocal proximity warning sounds

2020

View full text Add to dashboard Cite

show abstract

“…The two AoAs alongside with phase and amplitude information of subcarriers contained in the CSI make up the training features of the WiFi data set. While the UWB data set features include channel impulse response (CIR), preamble symbol accumulation (PSA) and distance estimates [10].…”

Section: The Proposed Localisation Frameworkmentioning

confidence: 99%

“…The choice of training features of UWB data has been well screened in [10]. We adopt the same selection in our evaluation.…”

Section: B Uwb Data Setmentioning

confidence: 99%

Bayesian Optimisation-Assisted Neural Network Training Technique for Radio Localisation

Liu¹,

Li²,

Zhu³

2022

Preprint

View full text Add to dashboard Cite

Radio signal-based (indoor) localisation technique is important for IoT applications such as smart factory and warehouse. Through machine learning, especially neural networks methods, more accurate mapping from signal features to target positions can be achieved. However, different radio protocols, such as WiFi, Bluetooth, etc., have different features in the transmitted signals that can be exploited for localisation purposes. Also, neural networks methods often rely on carefully configured models and extensive training processes to obtain satisfactory performance in individual localisation scenarios. The above poses a major challenge in the process of determining neural network model structure, or hyperparameters, as well as the selection of training features from the available data. This paper proposes a neural network model hyperparameter tuning and training method based on Bayesian optimisation. Adaptive selection of model hyperparameters and training features can be realised with minimal need for manual model training design. With the proposed technique, the training process is optimised in a more automatic and efficient way, enhancing the applicability of neural networks in localisation.

show abstract

“…The small network size and the limited number of parameters in SNNs, result in a shorter training time and a computationally-efficient deployment for indoor localisation applications [23]. However, in the case of WiFi CSI, simple neural networks cannot perform effective signal extraction and accurately estimate target locations [2].…”

Section: A Shallow Neural Networkmentioning

confidence: 99%

Wireless Localisation in WiFi using Novel Deep Architectures

Cui

Khan

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

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 singlelayer 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.

show abstract

Standing on the Shoulders of Giants: AI-Driven Calibration of Localisation Technologies

Cited by 8 publications

References 5 publications

Distances of transmission risk of COVID‐19 inside dwellings and evaluation of the effectiveness of reciprocal proximity warning sounds

Distances of transmission risk of COVID‐19 inside dwellings and evaluation of the effectiveness of reciprocal proximity warning sounds

Bayesian Optimisation-Assisted Neural Network Training Technique for Radio Localisation

Wireless Localisation in WiFi using Novel Deep Architectures

Contact Info

Product

Resources

About