Houda Labiod

Koffi Ismael Ouattara, Ioannis Krontiris, Theo Dimitrakos et al.

Trustworthiness has become a key requirement for the deployment of artificial intelligence systems in safety-critical applications. Conventional evaluation metrics, such as accuracy and precision, fail to appropriately capture uncertainty or the reliability of model predictions, particularly under adversarial or degraded conditions. This paper introduces the Parallel Trust Assessment System (PaTAS), a framework for modeling and propagating trust in neural networks using Subjective Logic (SL). PaTAS operates in parallel with standard neural computation through Trust Nodes and Trust Functions that propagate input, parameter, and activation trust across the network. The framework defines a Parameter Trust Update mechanism to refine parameter reliability during training and an Inference-Path Trust Assessment (IPTA) method to compute instance-specific trust at inference. Experiments on real-world and adversarial datasets demonstrate that PaTAS produces interpretable, symmetric, and convergent trust estimates that complement accuracy and expose reliability gaps in poisoned, biased, or uncertain data scenarios. The results show that PaTAS effectively distinguishes between benign and adversarial inputs and identifies cases where model confidence diverges from actual reliability. By enabling transparent and quantifiable trust reasoning within neural architectures, PaTAS provides a foundation for evaluating model reliability across the AI lifecycle.

Hengchuan Tan, Maode Ma, Houda Labiod et al.

Wireless Mesh Networks (WMNs) have emerged as a key technology for the next generation of wireless networking due to its self-forming, self-organizing and self-healing properties. However, due to the multi-hop nature of communications in WMN, we cannot assume that all nodes in the network are cooperative. Nodes may drop all of the data packets they received to mount a Denial of Service (DoS) attack. In this paper, we proposed a lightweight trust detection mechanism called Trusted Entropy and Dempster Shafer (TEDS) to mitigate the effects of blackhole attacks. This novel idea combines entropy function and Dempster Shafer belief theory to derive a trust rating for a node. If the trust rating of a node is less than a threshold, it will be blacklisted and isolated from the network. In this way, the network can be assured of a secure end to end path free of malicious nodes for data forwarding. Our proposed idea has been extensively tested in simulation using network simulator NS-3 and simulation results show that we are able to improve the packet delivery ratio with slight increase in normalized routing overhead.

Heng Chuan Tan, Maode Ma, Houda Labiod et al.

Trust models that rely on recommendation trusts are vulnerable to badmouthing and ballot-stuffing attacks. To cope with these attacks, existing trust models use different trust aggregation techniques to process the recommendation trusts and combine them with the direct trust values to form a combined trust value. However, these trust models are biased as recommendation trusts that deviate too much from one's own opinion are discarded. In this paper, we propose a non-biased trust model that considers every recommendation trusts available regardless they are good or bad. Our trust model is based on a combination of 2 techniques: the dissimilarity test and the Dempster-Shafer Theory. The dissimilarity test determines the amount of conflict between 2 trust records, whereas the Dempster-Shafer Theory assigns belief functions based on the results of the dissimilarity test. Numerical results show that our trust model is robust against reputation-based attacks when compared to trust aggregation techniques such as the linear opinion pooling, subjective logic model, entropy-based probability model, and regression analysis. In addition, our model has been extensively tested using network simulator NS-3 in an Infrastructure-based wireless mesh networks and a Hybrid-based wireless mesh networks to demonstrate that it can mitigate blackhole and grayhole attacks.

Hengchuan Tan, Maode Ma, Houda Labiod et al.

Public key infrastructure (PKI) is the most widely used security mechanism for securing communications over the network. However, there are known performance issues, making it unsuitable for use in vehicular networks. In this paper, we propose a secure and authenticated key management protocol (SA-KMP) to overcome the shortcomings of the PKI. The SA-KMP scheme distributes repository containing the bindings of the en-tity's identity and its corresponding public key to each vehicle and road side unit. By doing so, certificate exchanges and certificate revocation lists are eliminated. Furthermore, the SA-KMP scheme uses symmetric keys derived based on a 3-D-matrix-based key agreement scheme to reduce the high computational costs of using asymmetric cryptography. We demonstrate the efficiency of the SA-KMP through performance evaluations in terms of transmission and storage overhead, network latency, and key generation time. Analytical results show that the SA-KMP is more scalable and outperforms the certificate-based PKI. Simulation results indicate that the key generation time of the SA-KMP scheme is less than that of the existing Elliptic Curve Diffie--Hellman and Diffie--Hellman protocols. In addition, we use Proverif to prove that the SA-KMP scheme is secure against an active attacker under the Dolev and Yao model and further show that the SA-KMP scheme is secure against denial of service, collusion attacks, and a wide range of other malicious attacks.