-
Optimization of Energy Consumption in Delay-Tolerant Networks
Authors:
Junran Wang,
Milena Radenkovic
Abstract:
Delay tolerant network is a network architecture and protocol suite specifically designed to handle challenging communications environments, such as deep space communications, disaster response, and remote area communications. Although DTN [1]can provide efficient and reliable data transmission in environments with high latency, unstable connections, and high bit error rates, its energy consumptio…
▽ More
Delay tolerant network is a network architecture and protocol suite specifically designed to handle challenging communications environments, such as deep space communications, disaster response, and remote area communications. Although DTN [1]can provide efficient and reliable data transmission in environments with high latency, unstable connections, and high bit error rates, its energy consumption optimization problem is still a challenge, especially in scenarios with limited resources.To solve this problem, this study combines the Epidemic[2] and MaxProp[3] routing protocols with Machine Learning Models to optimize the energy consumption of DTNs. Hundreds of simulations were conducted in the ONE simulator, and an external real-world dataset from San Francisco taxi mobility traces [54] was imported. Random Forest[4] and Gradient Boosting Machine (GBM)[5] models were employed for data analysis. Through optimization involving Hyperparameter Tuning and Feature Selection, the Random Forest model achieved an R-squared value of 0.53, while the GBM model achieved an R-squared value of 0.65.
△ Less
Submitted 5 June, 2024;
originally announced June 2024.
-
Opportunistic Delay Tolerant Routing for LED Wristbands in Music Events
Authors:
Nandita Joshi,
Milena Radenkovic
Abstract:
The existing technology used for LED wristbands does not explore the use of opportunistic delay tolerant network (DTN) routing protocols to disseminate lighting information. This paper multiple trade-offs of different criteria with the protocol (Spray & Focus and PRoPHETv2) changes with varying music event scenario configurations & capacities using the ONE simulator. When considering variable musi…
▽ More
The existing technology used for LED wristbands does not explore the use of opportunistic delay tolerant network (DTN) routing protocols to disseminate lighting information. This paper multiple trade-offs of different criteria with the protocol (Spray & Focus and PRoPHETv2) changes with varying music event scenario configurations & capacities using the ONE simulator. When considering variable music event durations, it was shown that PRoPHETv2 was more desirable for longer durations however Spray & Focus accounted for network congestion issues that could occur for smaller buffer sizes in the wristbands. When considering variable patterns of audience density, Spray & Focus had higher delivery probabilities whereas PRoPHETv2 had near zero results. Further research can be conducted in incorporating buffer removal management to the protocols as well as experimenting with different congestion targeted DTN protocols.
△ Less
Submitted 4 June, 2024;
originally announced June 2024.
-
SubwayMeshDTN: Exploring Opportunistic Delay Tolerant Routing Protocols when Disseminating Emergency Alerts on a Smart City Subway Network
Authors:
Bruce Estivil,
Milena Radenkovic
Abstract:
This paper seeks to understand the effectiveness of using multi-dimensional opportunistic delay-tolerant network (DTN) routing protocols, specifically Epidemic and MaxProp, in the context of New York City (NYC) metropolitan subway network. We examine how efficiently emergency messages spread through mobile, self-configuring, edge-based movement patterns on the train network to understand and propo…
▽ More
This paper seeks to understand the effectiveness of using multi-dimensional opportunistic delay-tolerant network (DTN) routing protocols, specifically Epidemic and MaxProp, in the context of New York City (NYC) metropolitan subway network. We examine how efficiently emergency messages spread through mobile, self-configuring, edge-based movement patterns on the train network to understand and propose solutions for improving communication in subterranean environments. Since DTNs are able to store, carry and forward messages through intermediate edges, this paper benchmarks both Wi-Fi and Bluetooth topologies to compare and critically evaluate movement patterns, latency, overheads and delivery rates on pseudo-realistic underground traces. We also show that the accordion effect is predominant in these networks, and therefore, the most effective protocol configurations vary.
△ Less
Submitted 13 May, 2024; v1 submitted 4 May, 2024;
originally announced May 2024.
-
Autonomous Vehicle Networks for More Reliable Truck Tracking in Challenged High Mountain Roads, Tunnels and Bridges Environments
Authors:
Junhao Chen,
Milena Radenkovic
Abstract:
The popularity of online shop** has challenged the existing express tracking. How to provide customers with reliable and stable express tracking has become one of the important issues that express companies need to solve now. The current stage of courier tracking is not ideal in challenging environments such as mountain roads, tunnels and city centres. Therefore, the project aims to overcome the…
▽ More
The popularity of online shop** has challenged the existing express tracking. How to provide customers with reliable and stable express tracking has become one of the important issues that express companies need to solve now. The current stage of courier tracking is not ideal in challenging environments such as mountain roads, tunnels and city centres. Therefore, the project aims to overcome the challenging environment and achieve stable express tracking, and proposes the Ya'an scenario and conducted multiple experiments. We show that opportunistic DTN-aware protocols are feasible solution for trucks to maintain stable communication in challenging environments, and nodes maintain extremely high message delivery rates and average delays that can maintain communication.
△ Less
Submitted 3 April, 2024;
originally announced April 2024.
-
Exploring Opportunistic Routing for Remote Sea Emergencies
Authors:
Cleon Liew,
Milena Radenkovic
Abstract:
This paper explores the Opportunistic Routing Protocols in the context of remote sea emergency scenarios, using the MH370 plane crash as a case study (OppNetMH370). We studied the likelihood of successful transmissions of emergency messages to response services where communication methods are inadequate in remote sea areas. The study focuses on two opportunistic routing protocols, where their perf…
▽ More
This paper explores the Opportunistic Routing Protocols in the context of remote sea emergency scenarios, using the MH370 plane crash as a case study (OppNetMH370). We studied the likelihood of successful transmissions of emergency messages to response services where communication methods are inadequate in remote sea areas. The study focuses on two opportunistic routing protocols, where their performances are evaluated based on key metrics including average latency and delivery probability. Our study reveals the challenges associated with the current communication technology in remote areas and proposes potential enhancements for future simulations. The findings contribute to understanding the limitations of existing communication strategies in remote sea areas and offers insights on the future development and improvements to the routing protocols.
△ Less
Submitted 3 April, 2024;
originally announced April 2024.
-
NeuraLunaDTNet: Feedforward Neural Network-Based Routing Protocol for Delay-Tolerant Lunar Communication Networks
Authors:
Parth Patel,
Milena Radenkovic
Abstract:
Space Communication poses challenges such as severe delays, hard-to-predict routes and communication disruptions. The Delay Tolerant Network architecture, having been specifically designed kee** such scenarios in mind, is suitable to address some challenges. The traditional DTN routing protocols fall short of delivering optimal performance, due to the inherent complexities of space communication…
▽ More
Space Communication poses challenges such as severe delays, hard-to-predict routes and communication disruptions. The Delay Tolerant Network architecture, having been specifically designed kee** such scenarios in mind, is suitable to address some challenges. The traditional DTN routing protocols fall short of delivering optimal performance, due to the inherent complexities of space communication. Researchers have aimed at using recent advancements in AI to mitigate some routing challenges [9]. We propose utilising a feedforward neural network to develop a novel protocol NeuraLunaDTNet, which enhances the efficiency of the PRoPHET routing protocol for lunar communication, by learning contact plans in dynamically changing spatio-temporal graph.
△ Less
Submitted 7 April, 2024; v1 submitted 29 March, 2024;
originally announced March 2024.
-
On mistakes we made in prior Computational Psychiatry Data driven approach projects and how they jeopardize translation of those findings in clinical practice
Authors:
Milena Čukić Radenković,
David Pokrajac,
Victoria Lopez
Abstract:
After performing comparison of the performance of seven different machine learning models on detection depression tasks to show that the choice of features is essential, we compare our methods and results with the published work of other researchers. In the end we summarize optimal practices in order that this useful classification solution can be translated to clinical practice with high accuracy…
▽ More
After performing comparison of the performance of seven different machine learning models on detection depression tasks to show that the choice of features is essential, we compare our methods and results with the published work of other researchers. In the end we summarize optimal practices in order that this useful classification solution can be translated to clinical practice with high accuracy and better acceptance.
△ Less
Submitted 11 June, 2020;
originally announced June 2020.
-
Machine Learning Approaches for Detecting the Depression from Resting-State Electroencephalogram (EEG): A Review Study
Authors:
Milena Čukić Radenković,
Victoria Lopez Lopez
Abstract:
In this paper, we aimed at reviewing present literature on employing nonlinear analysis in combination with machine learning methods, in depression detection or prediction task. We are focusing on an affordable data-driven approach, applicable for everyday clinical practice, and in particular, those based on electroencephalographic (EEG) recordings. Among those studies utilizing EEG, we are discus…
▽ More
In this paper, we aimed at reviewing present literature on employing nonlinear analysis in combination with machine learning methods, in depression detection or prediction task. We are focusing on an affordable data-driven approach, applicable for everyday clinical practice, and in particular, those based on electroencephalographic (EEG) recordings. Among those studies utilizing EEG, we are discussing a group of applications used for detecting the depression based on the resting state EEG (detection studies) and interventional studies (using stimulus in their protocols or aiming to predict the outcome of therapy). We conclude with a discussion and review of guidelines to improve the reliability of developed models that could serve the improvement of diagnostic and more accurate treatment of depression.
△ Less
Submitted 6 September, 2019;
originally announced September 2019.
-
Machine learning approaches in Detecting the Depression from Resting-state Electroencephalogram (EEG): A Review Study
Authors:
Milena Cukic Radenkovic
Abstract:
In this paper, we aimed at reviewing several different approaches present today in the search for more accurate diagnostic and treatment management in mental healthcare. Our focus is on mood disorders, and in particular on the major depressive disorder (MDD). We are reviewing and discussing findings based on neuroimaging studies (MRI and fMRI) first to get the impression of the body of knowledge a…
▽ More
In this paper, we aimed at reviewing several different approaches present today in the search for more accurate diagnostic and treatment management in mental healthcare. Our focus is on mood disorders, and in particular on the major depressive disorder (MDD). We are reviewing and discussing findings based on neuroimaging studies (MRI and fMRI) first to get the impression of the body of knowledge about the anatomical and functional differences in depression. Then, we are focusing on less expensive data-driven approach, applicable for everyday clinical practice, in particular, those based on electroencephalographic (EEG) recordings. Among those studies utilizing EEG, we are discussing a group of applications used for detecting of depression based on the resting state EEG (detection studies) and interventional studies (using stimulus in their protocols or aiming to predict the outcome of therapy). We conclude with a discussion and review of guidelines to improve the reliability of developed models that could serve improvement of diagnostic of depression in psychiatry.
△ Less
Submitted 26 March, 2019;
originally announced March 2019.
-
Congestion Aware Spray And Wait Protocol: A Congestion Control Mechanism For The Vehicular Delay Tolerant Network
Authors:
Chuka Oham,
Milena Radenkovic
Abstract:
In the last few years, the Vehicular Ad-hoc Network (VANET) has come to be an important area of research. Significant research has been conducted to improve the performance of VANETS. One output of further research conducted on VANET is the Vehicular Delay Tolerant Network (VDTN). It is an application of the mobile DTN where nodes relay messages in the network using a store-carry-forward approach.…
▽ More
In the last few years, the Vehicular Ad-hoc Network (VANET) has come to be an important area of research. Significant research has been conducted to improve the performance of VANETS. One output of further research conducted on VANET is the Vehicular Delay Tolerant Network (VDTN). It is an application of the mobile DTN where nodes relay messages in the network using a store-carry-forward approach. Due to its high mobility, it suffers frequent disconnections and also congestions at nodes which leads to message drops. To minimize the rate of message drops and so optimize the probability of message delivery so that drivers are increasingly aware of the situation of the road, we propose a congestion control mechanism: Congestion Aware Spray and Wait (CASaW) protocol in this work so as to optimize the rate of message delivery to its destination and so increase the awareness of drivers in the vehicular environment thereby improve road safety. The results have shown that our proposition performed better than other classical VDTN protocols in terms of message delivery probability and rate of packet drops performance measures. We used the Opportunistic Networking Environment (ONE) simulator to implement the classical VDTN protocols: the PROPHET protocol, the Epidemic protocol, the MaxProp protocol and the Spray and Wait Protocol. The simulation scenarios shows a better performance for the congestion control mechanism we propose as it maintains a good message delivery rate as well as minimize the rate of packet losses thereby optimizing the chances of messages getting to their destinations and so improve road safety.
△ Less
Submitted 7 January, 2016;
originally announced January 2016.
