Prof. Dr.
Murugaraj Odiathevar

Task offloading strategies in Mobile Edge Computing (MEC) aim to reduce computation delay and energy consumption of mobile devices by offloading tasks to edge servers, which is key to improving MEC system performance and user experience. Recent efforts have focused on utilizing deep reinforcement learning (DRL) but DRL-based offloading strategies struggle to achieve optimal decisions in limited iterations due to complexity. Therefore, to address the complexity challenge, this paper proposes a Prune-based Deep reinforcement learning Offloading Algorithm (PDOA) to enhance MEC system performance. First, we construct a dynamic MEC system model and formulate the task offloading problem as a Markov decision process to minimize the total cost of the MEC system. Next, we propose a prune-based DRL offloading algorithm, which prunes DRL models to reduce the complexity and improve learning efficiency, thereby lowering system costs. The experimental results show that PDOA reduces the computational cost of MEC systems significantly compared with other methods and lowers system costs by over 10%. This optimization approach provides a novel research perspective for applying DRL models in MEC.

Network data constantly evolves with new network applications and protocols. There is a need for robust techniques to detect anomalous behaviour. Offline models trained with static data lose validity when new variants of traffic emerge. They require retraining but the need for ground truth and lengthy training times make this task challenging. Meanwhile, online models which detect outliers in streaming data are susceptible to the curse of dimensionality and natural variability. Today’s anomalies may be tomorrow’s new traffic and existing methods do not provide a way to differentiate between them. We propose a framework that makes the most of both approaches: an offline deep learning model extracts features of normal traffic and provides a bias for an online outlier detection model to select data for training. The online model retains its previously learnt knowledge and retrains itself with new data. Online thresholds are updated in a drifting manner and the Mann-Whitney U test is incorporated to prevent inaccurate updates. We perform analysis on the scores, develop heuristics to detect new traffic and evaluate using three deep learning models and four outlier detection methods on the UNSW-NB15 and CTU-13 datasets. The framework improves upon any individual offline or online models in isolation.

The challenge of anomaly detection is to obtain an accurate understanding of expected behaviour which is intensified when the data are distributed heterogeneously. Transmitting raw data to a central site incurs high communication overhead and raises privacy issues. The concept of Edge AI allows computation to be performed at the edge site allowing for quick decision making in mission critical scenarios such as self-driving cars. A model is learnt locally and its parameters are transmitted and aggregated. However, existing methods of aggregation do not account for variance and heterogeneous distribution of data. They also do not consider edge constraints such as limited computational, memory and communication capabilities of edge devices. In this work, a fully Bayesian approach is employed by means of a Bayesian Random Vector Functional Link AutoEncoder being incorporated with Expectation Propagation for distributed training. Our anomaly detection system operates without any transmission of raw data, is robust under inhomogeneous network densities and under uneven and biased data distributions. It allows for asynchronous updates to converge in a few iterations and is a relatively simple neural network addressing edge constraints without compromising on performance as compared to existing more complex models.

Bufferbloat, or excessive queuing delay under load, is a noticeable quality of service degradation that occurs when latency-sensitive traffic experiences the effects of increased packet buffering delays in network devices. This phenomenon leads to increased latency and reduced network performance, particularly affecting real-time voice and video traffic, online gaming, and other interactive applications that demand low-latency at all times. Bufferbloat also poses serious risks to future ambitions of latency-critical applications such as telesurgery, autonomous vehicles, and virtual reality, as it undermines network consistency and creates significant operational issues for service providers to manage. Bufferbloat is especially detrimental to Wireless Internet Service Providers (WISPs), due to their often ad-hoc and dynamic nature. To better characterise the prevalence of bufferbloat, we analyse a real-world WISP network and propose "Polus", a framework for detecting and characterising adverse network conditions caused by the phenomenon.