Many IoT use cases have ultra-low latency and strong security requirements. But achieving both simultaneously is challenging. In this paper, as a use case, we consider the authentication of IoT devices for every transaction and develop a fast and secure authentication protocol. Our key idea is to leverage highly secure Physically Unclonable Functions (PUFs) and high-speed programmable switch and offload PUF-based authentication protocol to the switch. By doing so, it enables authentication of every transaction at network speed. In this paper, we demonstrate the feasibility of our idea by offloading the authentication protocol to a programmable switch with Tofino chip. Our preliminary experiments show that protocol offloading reduces authentication latency by 2-4 times and scales to a few hundred thousand IoT devices.

Accelerating PUF-based Authentication Protocols Using Programmable Switch

Detecting compromised IoT devices: Existing techniques, challenges, and a way forward

The adoption of Software-Defined Networks (SDN) and the shift towards programmable data planes have led to better network management. However, this has not been accompanied with the implementation of robust host authentication or access control mechanisms to improve network security and prevent unauthorized access to the network. The current literature has explored the implementation of the widely adopted authentication mechanism - port knocking in SDN to address the former. However, they suffer from two major drawbacks making them vulnerable to MITM (Man-In-The-Middle) attacks: unsecured transfer of the port knocking sequences between the SDN controller and hosts, and the lack of host identity verification mechanisms post port knocking authentication. This paper introduces P4-sKnock: a P4 based two level host authentication and access control mechanism. The first level introduces encrypted dynamic port knocking to secure the transfer of port knocking sequences over a compromised channel by encrypting them. Further, a challenge-response host identity verification mechanism is introduced as a second level authentication measure following which a host can be authorized, quarantined or blocked owing to the programmability of the P4 switch providing robust access control. Experimental analysis shows that P4-sKnock can authenticate a new SDN host within 500 ms and mitigate MITM attacks like IP spoofing and replay attacks making it significantly more secure than previous P4 based port knocking authentication systems.

P4-sKnock: A Two Level Host Authentication and Access Control Mechanism in P4 based SDN

The rapid popularity of IoT devices has led to an escalating number of sophisticated cybersecurity attacks. Prior security mechanisms are inaccurate and incur high computational costs for resource-constrained IoT devices, hindering their scalability to large networks. Manufacturer Usage Description (MUD) has been introduced to overcome IoT security challenges. However, it cannot mitigate volumetric attacks in IoT networks. This paper proposes iDAM: a distributed self-learning, autonomous system to detect and mitigate volumetric attacks in IoT networks. iDAM monitors and authenticates the behavioral profiles of MUD compliant IoT devices and builds specific-device-type OC-SVM models aggregated using federated learning. The solution can cope with the occurrence of volumetric attacks at several levels of the IoT infrastructure and the compromise of the internal components of the proposed solution. We have extensively evaluated our solution with the IoT network intrusion dataset, which shows that iDAM can efficiently mitigate several volumetric attacks by detecting anomalous packet flows in the network with an AUC of 0.9597. Testing iDAM against a real-time SYN flood attack in an experimental setup and its ability to quickly mitigate the attack solidifies the conclusion that it can be deployed in a real-time environment to detect and mitigate volumetric attacks effectively.

iDAM: A Distributed MUD Framework for Mitigation of Volumetric Attacks in IoT Networks

Egocentric vision data has become popular due to its unique way of capturing first-person perspective. However they are lengthy, contain redundant information and visual noise caused by head movements which disrupt the story being expressed through them. This paper proposes a novel visual feature and gaze driven approach to retarget egocentric videos following the principles of cinematography. This approach is divided into two parts: activity based scene detection and performing panning and zooming to produce visually immersive videos. Firstly, visually similar frames are grouped using DCT feature matching followed by SURF descriptor matching. These groups are further refined using the gaze data to generate different scenes and transitions occurring within an activity. Secondly, the mean 2D gaze positions of scenes are used for generating panning windows enclosing 75% of the frame content. This is done for performing zoom-in and zoom-out operations in the detected scenes and transitions respectively. Our approach has been tested on the GTEA and EGTEA gaze plus datasets witnessing an average accuracy of 88.1% and 72% for sub-activity identification and obtaining an average aspect ratio similarity (ARS) score of 0.967 and 0.73; 60% and 42% SIFT similarity index (SSI) respectively. Code available on Github.1

A Novel Visual Feature and Gaze Driven Egocentric Video Retargeting

We present DanceAnyWay, a hierarchical generative adversarial learning method to synthesize mixed-genre dance movements of 3D human characters synchronized with music. Our method learns to disentangle the dance movements at the beat frames from the dance movements at all the remaining frames by operating at two hierarchical levels. At the coarser"beat"level, it encodes the rhythm, pitch, and melody information of the input music via dedicated feature representations only at the beat frames. It leverages them to synthesize the beat poses of the target dance using a sequence-to-sequence learning framework. At the finer"repletion"level, our method encodes similar rhythm, pitch, and melody information from all the frames of the input music via dedicated feature representations and couples them with the synthesized beat poses from the coarser level to synthesize the full target dance sequence using an adversarial learning framework. By disentangling the broader dancing styles at the coarser level from the specific dance movements at the finer level, our method can efficiently synthesize dances composed of arbitrarily mixed genres and styles. We evaluate the performance of our approach through extensive experiments on both the mixed-genre TikTok dance dataset and the single-genre AIST++ dataset and observe improvements of about 2% in motion quality metrics and 1.6% - 5.9% in motion diversity metrics over the current baselines in the two datasets respectively. We also conducted a user study to evaluate the visual quality of our synthesized dances. We noted that, on average, the samples generated by our method were about 9% more preferred by the participants and had a 12% better five-point Likert-scale score over the best available current baseline in terms of motion quality and diversity.

DanceAnyWay: Synthesizing Mixed-Genre 3D Dance Movements Through Beat Disentanglement

Detection of COVID-19 has been a global challenge due to the lack of proper resources across all regions. Recently, research has been conducted for non-invasive testing of COVID-19 using an individual's cough audio as input to deep learning models. However, these methods do not pay sufficient attention to resource and infrastructure constraints for real-life practical deployment and the lack of focus on maintaining user data privacy makes these solutions unsuitable for large-scale use. We propose a resource-efficient CoviFL framework using an AIoMT approach for remote COVID-19 detection while maintaining user data privacy. Federated learning has been used to decentralize the CoviFL CNN model training and test the COVID-19 status of users with an accuracy of 93.01 % on portable AIoMT edge devices. Experiments on real-world datasets suggest that the proposed CoviF L solution is promising for large-scale deployment even in resource and infrastructure-constrained environments making it suitable for remote COVID-19 detection.

Aneesh Bhattacharya

Papers

P4-sKnock: A Two Level Host Authentication and Access Control Mechanism in P4 based SDN

iDAM: A Distributed MUD Framework for Mitigation of Volumetric Attacks in IoT Networks

A Novel Visual Feature and Gaze Driven Egocentric Video Retargeting

DanceAnyWay: Synthesizing Mixed-Genre 3D Dance Movements Through Beat Disentanglement

CoviFL: Edge-Assisted Federated Learning for Remote COVID-19 Detection in an AIoMT Framework