The fifth-generation (5G) New Radio Vehicle-to-Everything (NR-V2X) technology supports unicast and groupcast with broadcast communications. These new features enable a vehicle to reliably exchange messages with another vehicle or a group of vehicles. However, these advanced features also create opportunities for malicious actors. We propose two stealthy jamming attacks that exploit the predictable physical layer and the vulnerable decentralized radio resource scheduling algorithm of NR-V2X. In these attacks, assailants leverage the unicast feature for secure coordination to attack the broadcast communications among vehicles. We conduct simulations in a highway scenario. Simulation results demonstrate that our proposed attacks can significantly decrease the packet reception ratio (PRR) by 32.65% and increase the packet inter-reception (PIR) by 30.99%. Simulation results also reveal that in specific attack windows and when vehicle density increases, the impact of our proposed attacks nearly doubles. To the best of our knowledge, this is the first work that exploits the unicast/groupcast mode of NR-V2X to carry out an attack.

The popularity of WiFi keeps increasing because of its capability to deliver a cost-efficient broadband Internet connection. However, a new problem emerges when many basic service sets (BSSs) are deployed in the same area, which increases inter-access point (AP) contention and co-channel interference. Inter-AP contention will decrease the overall throughput in the area and cause long application delay. The rise of bandwidth-hungry and low-latency applications like virtual and augmented reality, online gaming, and video streaming makes it urgent to solve this problem. The IEEE 802.11be Working Group (TGbe) proposes a new feature, coordinated spatial reuse (CSR). This feature allows concurrent transmissions of multiple APs through coordination between APs, reducing inter-AP contention and co-channel interference. However, existing proposed CSR algorithms use one-way coordination where the AP that initiates the CSR transmits at its maximum power and other APs optimize their transmit (TX) power accordingly, resulting in poor throughput for other APs. To solve this problem, we propose a new bidirectional CSR algorithm, Enhanced Coordinated Spatial Reuse (ECSR), which collaboratively determines the TX power for all APs by setting an appropriate interference tolerance limit. To the best of our knowledge, this is the first work that proposes bidirectional multiple AP coordination for CSR. We evaluate our proposed algorithm in dense enterprise scenarios. Simulation results show that our proposed design achieves three times higher throughput than the traditional carrier sense multiple access (CSMA) technique and four times more than the existing one-way CSR. Our experimental results also show that none of the stations participating in ECSR transmissions have low throughput.

Voice Assistant Devices (VADs) such as Alexa, Google Now, and Siri have become increasingly popular because of their various voice-enabled features, including online shopping, controlling smart home appliances, accessing banking services, and more. However, it also brings unique security issues like voice replay attacks, where an attacker can generate a malicious voice command via a compromised playback device near the VAD. The usage of VADs in users' daily tasks makes such attack detection more important. To detect such attacks, we propose a defense system leveraging the impact of playback devices on the bass frequency region (0 to 500 Hz) of multi-channel audios. Currently, no prior work has exploited the playback device's impact on the bass frequency region of the multi-channel audios to prevent voice replay attacks. Specifically, our system divides the bass region of each channel into ten sub-bands and computes the total percentage of the overall signal's power presented in each sub-band. To make the system more robust against advanced audio attacks, we also extract the Modified Group Delay Function (MODGDF) cepstral coefficients of the bass area for phase features. Then, our proposed system applies a support vector machine (SVM) classifier to infer whether a human or a compromised playback device initiates the voice command. The system is then tested against a public multi-channel replay attack dataset. The system performance is checked in four different environmental conditions, achieving a maximum of 1.21 % Equal Error Rate (EER). Our experimental results also show that incorporating more audio channels improves the attack detection performance.